uses
science for a changing world
Prepared in cooperation with the U.S. Environmental Protection Agency
Measurement and Monitoring for the 21st Century Initiative
User's Guide to the Collection and Analysis of Tree
Cores to Assess the Distribution of Subsurface
Volatile Organic Compounds
Scientific Investigations Report 2008-5088
U.S. Department of the Interior
U.S. Geological Survey

-------
Cower,  U.S. Geological Survey hydrologist using an increment borer to obtain a tree core (digitally modified photograph taken
        by Allison Vroblesky).

-------
                   to the                 and
of                 to            the                   of
By Don A. Vroblesky
Prepared in cooperation with the U.S. Environmental Protection Agency
Measurement and Monitoring for the 21st Century Initiative
Scientific Investigations Report 2008-5088
U.S.          of the Interior
U.S.          Sureef

-------
U.S.
DIRK KEMPTHORNE, Secretary

U.S.
Mark D.  Myers, Director
U.S. Geological Survey, Reston, Virginia: 2008
For product and ordering information:
World Wide Web: http://www.usgs.gov/pubprod
Telephone: 1-888-ASK-USGS

For more information on the USGS—the Federal source for science about the Earth, its natural and living resources,
natural hazards, and the environment:
World Wide Web: http://www.usgs.gov
Telephone: 1-888-ASK-USGS
Suggested citation:
Vroblesky, D.A., 2008, User's guide to the collection and analysis of tree cores to assess the distribution of subsurface
volatile organic compounds: U.S. Geological Survey Scientific Investigations Report 2008-5088, 59 p.
(available online at http://pubs.waterusgs.gov/sir2008~5088)

-------
                                                                                                                Hi
Contents

Abstract	1
Introduction	1
    Advantages of Tree Coring as a Tool to Examine Subsurface Volatile Organic Compound
           Concentrations	2
    Limitations of Tree Coring as a Tool to Examine Subsurface Volatile Organic Compound
           Concentrations	2
    Acknowledgments	3
Part 1.  Methodology for Collection and Analysis of Tree Cores	3
    Tree-Core Collection	3
    Tree-Core Analysis	7
    Quality Control and Assurance	8
Part 2.  Historical Perspectives and Technical Considerations	8
    Historical  Perspectives	8
    Technical  Rationale for Methodology	9
         Tree-Core Collection	9
         Tree-Core Transferto Sample Vials and Storage	10
         Tree-Core Equilibration Time	11
         Tree-Core Analysis	13
         Quality Control and Assurance	14
    Factors Influencing Volatile Organic Compound Concentrations in Tree Cores	15
         Types of Volatile Organic  Compounds	15
         Subsurface Volatile Organic Compound Concentrations	16
         Differences Among Tree Species	18
         Rooting Depth and Depth to the Contaminated Horizon	18
         Subsurface Lithology	19
         Concentration Differences Around the Tree Trunk	19
         Volatilization Losses from the Tree Trunk	20
         Rainfall Infiltration as a Dilution Mechanism	20
         Seasonal Influences	21
         Sorption	21
         Within-Tree Volatile Organic Compound Degradation	22
Summary	22
References Cited	23
Appendix 1. Case Studies	29
    Case Study 1: Chlorinated Ethenes from Ground Water in Tree Trunks
           (Summarized from D.A. Vroblesky, C.T. Nietch, and J.T.  Morris)	29
    Case Study 2: Tree Coring as a Guide to Well Placement, Solid Waste
           Management Unit 17, Naval Weapons Station Charleston, South Carolina,
           2002 (by D.A. Vroblesky and C.C. Casey)	32
    Case Study 3: Operable Unit 4, Riverfront Superfund Site,  Franklin County, Missouri
           (by J. Schumacher)	34
    Case Study 4: MTBE and BTEX in Trees above Gasoline-Contaminated Ground Water
           (by J.E. Landmeyer, D.A. Vroblesky, and P.M. Bradley)	38
Appendix 2: Air Sample Analysis for Volatile Organic Compounds (by S. Clifford)	43

-------
IV
                         Figures

                            1-4.   Photographs showing:
                                    1,   Typical tree-coring tool	3
                                    2,   Comparison of nonporous, diffuse-porous, and ring-porous wood	5
                                    3.   Crimping tool, 20-milliliter crimp-top serum vial, and 40-millilliter volatile
                                          organic analysis fVQA} vial used for collecting tree cores	6
                                    4,   Crimp-top serum vial containing tree core	6
                            5-8.   Graphs showing:
                                    5,   Loss of trichloroethene overtime from uncapped 20-milliliter serum vials,
                                          Naval Weapons Station Charleston, South Carolina, 2006	10
                                    6,   Headspace trichloroethene concentrations at 24 hours and 19 days
                                          of storage in sealed serum vials containing tree cores from a
                                          trichloroethene-contaminated site showing generally slight concentration
                                          increases in most cores and no evidence of volatilization loss overtime	11
                                    7,   Trichloroethene concentrations over time in unheated and heated
                                          sealed serum vials containing loblolly pine tree cores from a
                                          trichloroethene-contaminated site showing little concentration change in
                                          cores from some trees after 5 to 6 minutes and an increase in sensitivity
                                          by field heating and analyzing the cores, Solid Waste Management Unit 17,
                                          Charleston, South Carolina	12
                                    8,   Field-heated  and unheated (A) trichloroethene and (B) tetrachloroethene
                                          concentrations in 20-milliliter serum vials containing cores from
                                          trees growing over ground-water contamination showing that field
                                          heating the sample vials can produce VOC concentrations that are in the
                                          range of or higher than concentrations obtained by allowing the samples
                                          to equilibrate at room temperature overnight	13
                           9-10.   Maps showing:
                                    9,   Trees cored at the Nyanza Chemical Waste Dump Superfund Site, Ashland,
                                          Massachusetts, August 2006, showing  proximity to wells and to buildings
                                          where vapor-intrusion investigations were conducted in 2004	17
                                   10,   1,1,1-Trichloroethane concentrations in tree cores in 2006 showing
                                          correspondence to the combined results of 2003 and 2006 soil-gas
                                          investigations at the Merriam Manufacturing Company property,
                                          Durham Meadows Superfund Site, Connecticut	17
                             11.   Graph showing trichloroethene removal from the rhizosphere of bald cypress
                                    seedlings as (A) nanomoles per minute through the above-ground part
                                    of the plant and (B) fractional trichloroethene loss from carboy water
                                    during the summer and winter	21

                         Tables

                              1.   Examples of nonporous, diffuse-porous, and ring-porous trees	5
                              2.   Field data to be collected during tree-coring investigation	7
                              3.   Trichloroethene concentrations in tree cores collected within about 5 minutes
                                    of each other, using the same core barrel with no decontamination between
                                    cores, showing no carryover between core collections	14
                              4.   Chemical formula, Henry's Law constant, and  log octanol-water partition
                                    coefficients for selected volatile organic compounds reported in tree-coring
                                    investigations	15

-------
Conversion Factors and Datums
Multiply

inch (in.)
inch (in.)
foot (ft)
mile (mi)

square foot (ft2)
square mile (mi2)

gallon (gal)
cubic inch (in3)
cubic yard (yd3)

foot per day (ft/d)
foot per year (ft/yr)
gallon per minute (gal/min)
gallon per day (gal/d)
inch per year (in/yr)

foot per mile (ft/mi)
By
Length
2.54
25.4
0.3048
1.609
Area
0.09290
2.590
Volume
3.785
16.39
0.7646
Flow rate
0.3048
0.3048
0.06309
0.003785
25.4
Hydraulic gradient
0.1894
To obtain

centimeter (cm)
millimeter (mm)
meter (m)
kilometer (km)

square meter (m2)
square kilometer (km2)

liter (L)
cubic centimeter (cm3)
cubic meter (m3)

meter per day (m/d)
meter per year (m/yr)
liter per second (L/s)
cubic meter per day (m3/d)
millimeter per year (mm/yr)

meter per kilometer (m/km)
Temperature in degrees Celsius (°C) may be converted to degrees Fahrenheit (°F) as follows:
                                        °F=(1.8x°C)+32
Temperature in degrees Fahrenheit (°F) may be converted to degrees Celsius (°C) as follows:
                                         °C=(°F-32)/1.8
Vertical coordinate information is referenced to the North American Vertical Datum of 1988 (NAVD 88).
Horizontal coordinate information is referenced to North American Datum of 1983 (NAD 83).
Altitude, as used in this report, refers to distance above the vertical datum.
Concentrations of chemical constituents in water are given either in milligrams per liter (mg/L) or
micrograms per liter (ug/L).

-------
VI
                                               in this

                        BTEX           benzene, toluene, ethylbenzene, and xylene
                        cDCE           cis-1,2-dichloroethene
                        cni/yr           centimeter per year
                        DNAPL         dense non-aqueous phase liquid
                        BCD            electron-capture detector
                        GC             gas chromatograph
                        GC/MS          gas chromatograph/mass spectrometry
                        g/L             grani per liter
                        HSA            headspace analysis
                        kg              kilogram
                        Kmv             octanol-water partition coefficient
                        Kwmd            equilibrium distribution of compounds between tree tissue and water
                        L/d             liter per day
                        L/nr/yr          liter per square meter per year
                        mg/g            milligram per gram
                        mL             milliliter
                        rnpfa            miles per hour
                        MTBE          methyl tot-butyl ether
                        OS WER         Office of Solid Waste and Emergency Response
                        OU             operable unit
                        PC A            1,1,2,2-tetrachloroethane
                        PCE            tetrachloroethene
                        P1D             photoionization detection
                        ppbv            parts per billion by volume
                        ppmv           parts per million by volume
                        PRG            preliminary remediation goal
                        PT             purge-and-trap analysis
                        PVC            polyvinyl chloride
                        RCF            root concentration factor
                        SWMU          Solid Waste Management Unit
                        TC A            1,1,1 -trichloroethane
                        TCE            trichloroethene
                        1MB            trimethylbenzene
                        TSCF           transpiration stream concentration factor
                        |!g-h/kg          microgram in headspace per kilogram of wet core
                        |ig/kg           microgram per kilogram
                        |,ig/L            microgram per liter
                        |iL              microliter
                        USEPA          U.S. Environmental Protection Agency
                        USGS           U.S. Geological  Survey
                        VC             vinyl chloride
                        VOA            volatile organic analysis
                        VOC            volatile organic compound

-------
                                                                                          ¥11
and                  of             in this
Common name
Alder
Apple, pear
Ash
Aspen, cottonwood
Bald cypress
Beech
Birch
Blue beech
Buckeye
Cedar
Coffee tree
Cottonwood. Eastern
Cottonwood. Fremont
Cottonwood. narrow leaf
Dogwood
Douglas fir
Him, American
Elm, Cedar
Elm, Chinese
Eucalyptus
False cypress
Fig, laurel
Fir
Ginkgo
Gum
Hackbcrry, Southern
Hemlock
Holly
Scientific name
Alnus sp.
Primus sp.
Fraxinus sp.
Populus sp.
Taxodium sp.
Fagus sp.
B etui a sp.
Carpinus sp.
Aesculus sp.
Thuja sp.
Gymnocladus dioicus
Populus deltoides
Populus fremontii
Populus angustifolia James
Cornus sp.
Psendotsuga sp.
Ulmus ainericana
Ulinus crassifolia
Ulmus parvifolia
Eucalyptus sp.
Chamaecyparis sp.
Ficus microcarpa
Abies sp.
Ginkgo sp.
Nyssa sp.
Celtis laevigata
Tsuga sp.
Ilex sp.
Common name
Iloneylocust
Juniper
Juniper, Ashc
Larch
Locust
Magnolia
Maple
Maple, box elder
Mulberry
Oak
Oak. Texas live
Oak. White shin
Osage orange
Paulownia
Pine
Pine, loblolly
Poplar, hybrid
Redwood
Rosewood
Russian olive
Sassafras
Spruce
Sunflower
Sweetgum
Sycamore
Willow
Willow, Goodding's

Scientific name
Gleditsia sp.
Juniperus sp.
Juniperus ashei
Larix sp.
Robinia sp.
Magnolia sp.
Acer sp.
Acer negundo
Morns sp.
Quercus sp.
Quercusfiisiforniis
Quercus sinuata
Madura potnifera
Paulownia sp.
Finns sp.
Pinus taeda
Populus dell aides x Populus
trichocarpa
Sequoia sp.
Dalbergia sissoo
Elaeagnus angustifolia
Sassafras sp.
Picea sp.
Helianthus annuus
Liquidambar sp.
Platanus sp.
Salix sp.
Salix gooddingii


-------

-------
                          to  the                      and                   of
            to                the                          of
By Don A. Vroblesky
Abstract

    Analysis of the volatile organic compound content of tree
cores is an inexpensive, rapid, simple approach to examining
the distribution of subsurface volatile organic compound
contaminants. The method has been shown to detect several
volatile petroleum hydrocarbons and chlorinated aliphatic
compounds associated with vapor intrusion and ground-water
contamination. Tree cores, which are approximately 3 inches
long, are obtained by using an increment borer. The cores
are placed in vials and sealed. After a period of equilibration,
the cores can be analyzed by headspace analysis gas chroma-
tography. Because the roots are exposed to volatile organic
compound contamination in the unsaturated zone or shallow
ground water, the volatile organic compound concentrations in
the tree cores arc an indication of the presence of subsurface
volatile organic compound contamination. Thus, tree coring
can be used to detect and map subsurface volatile organic
compound contamination. For comparison of tree-core data at
a particular site, it is important to maintain consistent methods
for all aspects of tree-core collection, handling, and analysis.
Factors affecting the volatile organic compound concentrations
in tree cores include the type of volatile organic compound,
the tree species, the rooting depth, ground-water chemistry,
the depth to the contaminated horizon, concentration differ-
ences around the trunk related to variations in the distribution
of subsurface volatile organic compounds, concentration
differences with depth  of coring related to volatilization loss
through the bark and possibly other unknown factors, dilution
by rain, seasonal influences, sorption, vapor-exchange rates,
and wilhin-lree volatile organic compound degradation.
Introduction

    Tree roots absorb water and chemicals from the soil and
transport them up the tree trunk. Thus, the chemical content of
tree cores can be useful indicators of subsurface contamination
(Vroblesky and Yanosky, 1990; Vroblesky and others, 1992,
1999; Yanosky and Vroblesky, 1992, 1995; Yanosky and
others, 2001). A variety of volatile organic compounds
(VOCs) from subsurface contamination are known to be
taken up by plant roots into the trunks of trees. These
compounds include benzene, toluene, elhylbenzene,
xylcnc isomcrs, trimcthyl  benzene, methyl tert-butyl ether
(MTBE), 1,1,2,2-tetrachloroethane, trichloroethene (TCE),
tetrachloroethene (PCE), 1,1,1-trichloroethane (TCA), vinyl
chloride (VC), and ris-l,2-dichlorocthene (cDCE) (Burkcn,
2001; Burken and Schnoor, 1998; Ilirsh and others, 2003;
Landmeyer and others, 2000; Newman and others, 1997;
Nietch and others, 1999; Trapp and others, 2007; Vroblesky
and others, 1999, 2006). The presence of VOCs in tree trunks
can allow a reconnaissance-level mapping of the ground-water
plume by simple analysis of tree cores (Vroblesky and others,
1999, 2001, 2004). Determining the presence of subsurface
VOCs is useful for evaluating the potential ingestion risks to
human health from ground water and the potential for respira-
tion risks from vapor intrusion
into buildings.
     The purpose of this report is to provide a guide to the
use of tree coring as a tool to examine subsurface  VOCs. To
that end, the report is divided into two major parts. The first
part of the report presents basic guidelines for a tree-coring
investigation to examine subsurface VOCs. The second pail of
the report examines historical and technical issues related to
tree coring as a tool to examine subsurface contamination.
The technical considerations include rationale for various
aspects of the methodology and a discussion of factors influ-
encing VOC concentrations in tree cores. An understanding
of the factors influencing VOC concentrations in tree cores is
necessary to better plan field investigations and to understand
the meaning of the results of the investigation. In addition,
two appendixes are attached. Appendix  1 is a collection of
case studies. Appendix 2 is a protocol by the U.S. Environ-
mental Protection  Agency (USEPA) for conducting air-sample
analysis of VOCs using two different types of gas chromato-
graphs. The analytical method reported in Appendix 2 has
been tested only for selected VOCs (listed in Appendix 2).
Additional testing would be required to determine the
appropriateness of this method for the other VOCs
discussed in the report.

-------
      User's Guide to the Collection and Analysis of Tree Cores to       the Distribution of Subsurface Volatile Organic Compounds
    Funding for this guide was provided by the USEPA's
Office of Solid Waste and Emergency Response (OSWER)
through its Measurement and Monitoring Technologies for
the 21st Century Initiative http://clu-in.org/progratns/21M2/.
This initiative seeks to identify and disseminate information
on promising measurement and monitoring technologies in
response to waste management and cleanup needs. Methodolo-
gies described in this guide arc also fully compatible with
USEPA's Triad strategy to manage hazardous waste site
decisionmaking uncertainty through systematic planning,
dynamic work strategies, and real-time measurement technolo-
gies. More information on the Triad approach can be found at
http://www.triadcentral.org/.
    Tree coring to examine subsurface VOCs has several
advantages and limitations over more invasive approaches to
site investigations, such as well drilling.  The advantages and
limitations are listed below.


               of       Coring as a Tool  to


Concentrations

 1. Tree coring allows examination of unsaturated-zone
    and ground-water contamination in areas where cultural
    influences, vegetation cover, or concerns of landown-
    ers limit the ability lo use more invasive reconnaissance
    approaches to site characterization using large mechanical
    equipment.

 2. The method is  applicable to a variety of VOCs commonly
    associated with vapor intrusion and  ground-water con-
    tamination, including chlorinated solvents  and petroleum
    hydrocarbons.

 3. The presence of VOCs in tree cores is  a strong indicator
    of subsurface VOC contamination. The relative concen-
    trations of VOCs among tree cores from a particular site
    often can provide general information  on the relative
    distribution of VOC concentrations in  the subsurface.

 4. Under some conditions, tree coring can be used to detect
    VOCs in soil gas (Struckhoff, 2003; Schumacher and
    others, 2004; Struckhoff and others, 2005b), providing a
    potentially useful tool for examining soil-vapor concen-
    trations. Thus,  tree coring for VOCs may be an effective
    approach for determining areas that have relatively high
    vapor-intrusion potential.

 5. Tree coring can detect chlorinated solvents at relatively
    low concentrations. Schumacher and others (2004)
    determined that analysis of tree-core samples can be used
    to detect PCE contamination in soils at concentrations
    of several hundreds of micrograms per kilogram or less
    and PCE concentrations as low as 8 micrograms per liter
    (|ig/L) in ground water in direct contact with the roots.
 6. The method is rapid and uncomplicated. A tree core can
    be collected in less than 5 minutes. No decontamination
    of the core barrel is required, other than inspection to
    ensure that there is no paniculate carryover, such as sec-
    tions of tree core, remaining in the core barrel. The cores
    can be analyzed by the relatively simple gas chromatog-
    raphy.

 7. The samples can be analyzed for preliminary results in
    the field after equilibrating for 5 minutes or longer or by
    heating to assist in directing the tree-coring effort, or can
    be transported back to a laboratory for later analysis.

 8. If the samples are to be analyzed within a few days of col-
    lection, the  cores can be stored without refrigeration.

 9. The method is inexpensive. Increment borers can be
    purchased for a few hundred dollars and can be re-used
    for years with proper care. By contrast, well sampling is
    time consuming, and well sampling equipment can cost
    thousands of dollars. In the time it takes  to sample a well,
    several tree cores can be collected.

10. A minimal amount of field equipment is required. A large
    number of tree cores can be collected rapidly with an
    increment borer and sample containers.

11. There is evidence that parts of the aquifer where there is
    preferential, chlorinated-solvent dechlorinalion sometimes
    can be delineated by comparing parent/daughter ratios in
    tree cores (Vroblesky and others, 2004). Areas where  the
    tree-core data indicate a low parent/daughter ratio, such as
    TCE/cDCE, relative lo other areas of the site may be loca-
    tions of enhanced subsurface dechlorination.

12. Because trees collect water from the subsurface over the
    lateral and vertical extent of their root system, trees pro-
    vide information over a substantially larger volume than
    a well or soil sample. Tree coring targets contaminants in
    shallow horizons, such as the uppermost ground water, the
    capillary zone, and the unsaturated zone.
              of              as a Tool to

Concentrations

 1. Because there are a number of influences on tree-core
    VOC concentrations, the absence of VOCs in a tree core
    cannot be used to definitively show that VOC contamina-
    tion is not present. It is possible that at the site in ques-
    tion, the tree roots do not extend to the contamination
    because the tree can obtain adequate water supplies from
    a source shallower than the contamination, or because the
    contamination is otherwise inaccessible to the tree.

 2. The variety of influences on tree-core VOC concen-
    trations renders it improbable that tree-core VOC

-------
                                                     Parti. Methodology for Collection and Analysis of Tree Cores
    concentrations can be used in a quantitative way to deduce
    specific subsurface VOC concentrations. Instead, tree-
    core concentrations reflect the generalized distribution of
    high and low subsurface VOC concentrations.

 3. While the proper collection of several cores from a single
    large tree typically does not result in lasting damage to
    the tree, care should be taken to avoid excess  coring of
    individual trees to minimize stress to the tree.

 4. Because incorporation of infiltrating rainwater into the
    transpiration stream can dilute in-tree VOC concentra-
    tions, higher tree-core VOC concentrations probably will
    be obtained if the cores are collected during a relatively
    dry period rather than immediately after a rain event.
    Cores collected during the dormant-growth season may
    contain lower VOC concentrations than cores collected
    during the summer, although multiple field studies found
    that the VOC content of tree cores still can be useful
    indicators of subsurface VOC contamination even during
    the dormant-growth season (Vroblesky and others, 1999,
    2006; Richard Willey, U.S. Environmental Protection
    Agency, written commun., 2008).

 5. When using headspace gas chromatographic analysis of
    tree cores on a typical field gas chromatograph (GC), a
    number of tree-related volatile compounds elute at about
    the same time as VC, potentially complicating identifica-
    tion of that compound.
Tree-Core Collection

    Tree cores are collected by using a tree-coring tool
(fig. 1). The tree-coring tool consists of an increment borer
and a core extractor. The core extractor is a component that
easily can be misplaced in a forest because of its thin elon-
gated shape. A practical approach is to tie a brightly colored
plastic tape or cloth to the extractor (fig. 1).
 1. Choosing a tree-coring tool: Increment borers are avail-
    able in various lengths from 4 in. (inches) to 28 in., and
    in three diameters (0.169, 0.2, and 0.5 in.). Considering
    that most tree coring for VOCs involves coring to a depth
    of only about 3 in., increment borers from 8 to 10 in. long
    allow hands to be far enough from the tree so as not to
    scrape bark or come into contact with poison ivy vines
    yet not be so long as to become unwieldy. The smallest
    borer diameter is commonly used for general forestry
    applications, and the largest diameter is used when large
    amounts of wood are required for chemical analysis. Most
    of the investigations using increment borers to examine
    VOC concentrations in tree cores have used either 0.169-
    or 0.2-in. diameter borers. Increment borers also can be
    obtained in a two-thread or three-thread design. Three-
    thread designs typically advance farther per revolution
    than two-thread designs but can be more difficult to turn
    and to initially engage the wood than a two-thread design.
    Thus, two-thread designs are more suited to hardwoods,
    and three-thread designs are more suited to softwoods.
Acknowledgments

     Several people outside of the U.S. Geological Survey
(USGS) made important contributions to this publication.
Richard Willey (USEPA) functioned as project manager and
invested many hours in reviews, suggestions, and creative
input. Manuscript reviewers from the USEPA included Scott
Clifford, Charles Porfert, Anni Loughlin, Kathy Davies,
Michael Adam, and Robert Alvey. Judy Canova from the
South Carolina Department of Health and Environmental
Control also provided review comments. Joel Burken of the
University of Missouri-Rolla provided a colleague review of
the manuscript. James Landmeyer provided a colleague review
within the USGS. The suggestions and comments from these
reviewers were invaluable and, in some cases, changed the
direction of the manuscript.
Parti.  Methodology for Collection
and Analysis of Tree Cores

    Tree coring involves collecting a core from a tree and
transferring the core to sample vials. The samples can be
stored for a few days at room temperature prior to analysis. If
longer storage is required, the samples should be refrigerated.
                                Core extractor
                             with attached flagging
                Increment borer
                                                                                         3 inches
                                                                           111111111  i
                                                                          012345 centimeters
           Figure 1.  Typical tree-coring tool.

-------
     User's Guide to the Collection and Analysis of Tree Cores to        the Distribution of Subsurface Volatile Organic Compounds
2. Core-barrel sharpening:  Increment borers should be
   sharp enough to easily engage the wood. The borer needs
   sharpening when engaging the wood is difficult, when the
   borer cuts a rough core, when the edge feels dull to the
   touch, or whet) the extractor consistently fails to recover
   the core. It should not be assumed that a new borer is
   sharp; sharpening methods for increment borers can be
   found in Maeglin (1979) and Grissino-Mayer (2003).

3. Core-barrel cleaning:  New increment borers should be
   cleaned prior to use in examining tree-core VOC concen-
   trations because the new borers some limes are coated with
   a thin layer of oil. In addition, researchers involved in tree
   coring for the use of dendrochronology often lubricate the
   interior of the core barrels. Such lubrication, however, is
   inappropriate for the use of tree cores to examine VOCs.
   Borers can be cleaned with soap and water. For increment
   borers with a diameter of 0.2 in., rifle-cleaning rods with
   an attached soft cloth are effective for cleaning inside
   the barrel. A clean cloth can be pushed through the core
   barrel, sealed in a vial, allowed to equilibrate with the
   headspace, and analyzed by headspace chromatography to
   verify that the cleaning  removed potential interferences.

4. Initiating coring:  The most difficult part about tree cor-
   ing is initiating the coring. The boring should be started
   slowly and carefully to avoid sideways slippage of the bit
   against the tree trunk. One approach is to hold the borer
   shaft near the threaded bit with one hand while apply-
   ing pressure toward the Iree and turning the borer with
   the other hand. Folding chest plates and straps that wrap
   around the trunk also are commercially available as tools
   to assist in starling the bit into the tree. The coring should
   be approximately perpendicular to the tree trunk. Once the
   bit has begun drilling into the tree, both hands can be used
   to advance the borer. Increment borer drill chucks are
   available for use with a drill; however, standard 19-volt
   or less portable drills are not always powerful enough to
   collect a complete length of core.

5. Core-collection location:  When conducting a tree-coring
   survey to examine the areal distribution of VOCs at a site,
   the cores should be collected from the trees at about the
   same height. A core collected near the ground usually
   provides higher VOC concentrations than a core collected
   higher up the trunk; however, for ease of core collection
   in typical applications, a simple approach is to use mean
   breast height, a commonly used forestry term meaning
   about 4.3 feet (ft) above land surface. In general, the core
   should be collected from the side of the tree suspected
   to be closest to the target contamination body. Dead or
   damaged parts of the tree should be avoided. Refer to the
   technical considerations section of this report for a more
   complete discussion of concentration variations with
   height above the ground and with azimuthal direction.
6. Core length: The length of the core should be consistent
   among the cored trees and should include the outermost
   growth ring. A core of the outermost 3 in. (not includ-
   ing the bark) is sufficient to identify the contaminants. A
   simple approach is to mark the core barrel at a distance
   from the bit equivalent to the length of a serum vial
   (approximately 3 in., not including the bark).  By advanc-
   ing the borer until the mark is at the inner edge of the
   bark, xylem cores of uniform length can be obtained that
   fit into the sampling vials. If necessary, the cores can be
   broken to make the vials easier to seal.

7. Core removal:  The core is removed from the borer by
   means of an extractor (fig. 1). After the increment borer
   has been advanced into the tree to the desired depth, the
   extractor is fully inserted into the borer. The extractor
   should be inserted with the concave side of the extractor
   facing down. In typical forestry applications, the borer is
   then rotated one half turn counter clockwise to break the
   core, and the core is removed from the borer by pulling
   on the extractor. Often, this procedure will remove the
   entire core from the borer. In some cases, however, pan
   of the core will remain in the borer because extractors
   sometimes lose their ability to adequately grip the core or
   because part of the core becomes wedged in the bit. Prob-
   able causes of a jammed core in the core barrel include
   a dull core-barrel bit and a dirty core barrel. To facilitate
   comparison of tree-core VOC concentrations, it is prefer-
   able to maintain uniform lengths of core samples and to
   seal the entire length of required core in sampling vials
   as soon as possible. Therefore, if the extractor begins to
   fail to recover the entire core from the barrel,  an alternate
   approach should be used. The alternate approach is to
   insert the extractor fully into the borer once the depth of
   penetration has been reached, rapidly remove the borer
   from the tree, insert an unpainted golf tee into the end of
   the bit, and press it against the tree while removing the
   extractor and tree core. In most cases, this gently removes
   the entire core without damaging the cutting edge of
   the bit.

8. Parts of the tree core to be included:  Because the bark
   is not involved in transpiration, inclusion of the bark
   in the sample is not necessary. If the bark is removed,
   however, care should be taken to avoid accidental removal
   of the outermost xylem growth ring in ring-porous trees
   (table 1; fig. 2) because the outermost ring is the domi-
   nant path for water transport through the trunk (Ellmore
   and Ewers, 1986). A more detailed discussion can be
   found in the section on technical rational for methodol-
   ogy. Inclusion of the inner part of the  bark as part of the
   sample probably does not produce a substantial adverse
   effect on VOC headspace concentrations when the values
   are reported as parts per million by volume (ppmv) of
   headspace.

-------
                                                        Part 1.  Methodology for Collection and Analysis of Tree Cores
       Table 1.  Examples of nonporous, diffuse-porous, and ring-porous trees.
                     Nonporous
            Diffuse porous
           Ring porous
       Bald cypress (Taxodium sp.)
       Cedar (Thuja sp.)
       Douglas fir (Pseudotsuga sp.)
       False cypress (Chamaecyparis sp.)
       Fir (Abies sp.)
       Ginkgo (Ginkgo sp.)
       Hemlock (Tsuga sp.)
       Juniper (Juniperus sp.)
       Larch (Lam sp.)
       Pine (Pinus sp.)
       Redwood (Sequoia sp.)
       Spruce (Picea sp.)
Alder (Alnus sp.)
Apple, pear (Prunus sp.)
Aspen, cottonwood (Populus sp.)
Beech (Fagus sp.)
Birch (Betula sp.)
Blue beech (Carpinus sp.)
Buckeye (Aesculus sp.)
Dogwood (Cornus sp.)
Eucalyptus (Eucalyptus sp.)
Gum (Nyssa sp.)
Holly (Ilex sp.)
Magnolia (Magnolia sp.)
Maple (Acer sp.)
Sweetgum (Liquidambar sp.)
Sycamore (Platanus sp.)
Willow (Salix sp.)
Ash (Fraxinus sp.)
Coffee tree (Gymnocladus dioicus)
Honeylocust (Gleditsia sp.)
Locust (Robinia sp.)
Mulberry (Moms sp.)
Oak (Quercus sp.)
Osage orange (Madura pomifera)
Paulownia (Paulownia sp.)
Sassafras (Sassafras sp.)
                              A. Nonporous wood
         B. Diffuse-porous wood  C. Ring-porous wood
                            Figure 2.  Comparison of nonporous, diffuse-porous, and ring-porous
                            wood (reprinted with permission from Chaney, 2000).
 9. Timing of tree-core transfer to sample vials:  Tree
    cores should be sealed in vials immediately upon recovery
    from the tree. Volatilization loss begins immediately upon
    removal of the core from the tree. If the core is not sealed
    in a vial within several seconds of collection, it should be
    discarded, and a new core should be collected.

10. Type of sample vial: Either 20-milliliter (mL) glass
    serum vials or 40-mL volatile organic analysis (VOA)
    vials can been used for collecting tree cores for head-
    space analysis (figs. 3 and 4). Consistency should be
    maintained within an individual study area. The VOA
    vials have an advantage in that they do not require the
    use of a crimping tool; however, the crimp-top serum-vial
    cap provides a better seal than the VOA-vial cap because
    the VOA-vial seal is designed for use with water samples
    rather than air samples. In addition, for a given mass of
    TCE in a tree core, higher headspace TCE concentrations
    are found in the 20-mL vials as compared to the 40-mL
    vials because of dilution. Therefore, if the expected
    concentrations are relatively low, the vials are going to be
    stored for several days prior to analysis, or the vials will
                    be heated, the 20-mL crimp-top serum vials offer a greater
                    margin of confidence.

                11.  Sample storage: In general, TCE concentrations appear
                    to be fairly stable for several days in sealed sampling vials
                    containing tree cores; however, it is prudent to analyze
                    the cores  within a few days of collection. The samples
                    should be stored in the dark because of the potential for
                    photodegradation of the VOCs. If the tree-core samples
                    will be analyzed within a few days of sample collection,
                    the samples can be stored at room temperature prior to
                    analysis to facilitate equilibration of the headspace in the
                    vials with the VOCs in the enclosed tree core. If a longer
                    period of time before analysis is anticipated, then refriger-
                    ating the cores will reduce the potential for vapor loss and
                    decay. Refrigerated cores should be allowed to equilibrate
                    for at least a few hours at room temperature or for a few
                    minutes by heating prior to analysis.

                12. Map the  location and mark the tree:  It is important to
                    preserve a record of the location of cored trees so that the
                    analytical results can accurately be related to the site. An

-------
  User's Guide to the Collection and Analysis of Tree Cores to Assess the Distribution of Subsurface Volatile Organic Compounds
                                                 6 inches
                    TT IT
n  i
                                      10 centimeters
                                                             Crimping
                                                                  tool
          Crimp cap

         Teflon-lined
         seal
                                         20-milliliter
                                         serum vial
                                                                             Vial cap with
                                                                             Teflon-coated seal
                                             40-milliliter
                                             VOAvial
                  Figure 3.   Crimping tool, 20-milliliter crimp-top serum vial, and 40-millilliter volatile
                  organic analysis (VOA) vial used for collecting tree cores.
                                    Figure 4.   Crimp-top serum vial containing
                                    tree core.
effort should be made to mark the approximate location of
each cored tree on a site map. If necessary, the tree loca-
tions can be precisely mapped by surveying at a later date.
The cored trees should be marked in some way to facili-
tate returning to the site to follow up on the investigation.
A variety of tree-marking methods are available. The
most long-term marking method involves "engraving"
the tree identification number on an aluminum tag with a
ball-point pen and nailing the tag to the tree. Depending
on esthetic issues and how long the tree-tag needs to last,
other means of marking the tree with an identification
number include tree paint, marked wooden stakes driven
into the ground adjacent to the tree, or colorful flagging
with the identification number written on the flag tied
around the tree.
                         13.  Tree-coring damage: Although tree coring can impart
                             local damage to tree trunks, most trees are capable of
                             compartmentalizing the damage and healing within 2-3
                             years with no adverse effects. Trees that have a more dif-
                             ficult time recovering are generally those that are short-
                             lived species or suppressed individuals. Sealing the core-
                             hole does not appear to accelerate the repair and probably
                             is not necessary. A more detailed discussion is available in
                             the section of this report titled "Tree-Core Collection" in
                             the "Technical Rationale fore Methodology" section.

                         14.  Supportive data to be collected: A variety of data
                             should be collected in conjunction with tree coring to
                             assist in interpretation of the results. The data are summa-
                             rized in table 2.

-------
                                                        Part 1. Methodology for Collection and Analysis of Tree Cores
           Table 2.  Field data to be collected during tree-coring investigation.
            Location information:
                Date/time.
                Site location.
                Unique tree identifier.
                Location of tree in study area.
                Note unusual geographic issues, such as if the tree is in an island in a parking lot where rainfall infiltration
                may be limited, on the edge of a cliff where depth to ground water may be large, on a stream bank where
                bank storage may constitute part of the tree's water supply, or other factors of potential interest.
                Temperature and weather conditions at the time of collection and for approximately 3 days prior to the sam-
                pling, based on the nearest weather station.
            Tree characteristics:
                Species.
                Indications of tree stress (damaged bark, dead branches, etc.).
                Whether or not there are leaves on the tree
                'Free diameter (can be rneasuiod vulh hoc diamcloi tape moasmcl
                Note whether or not the hoc IN oomideied to lopiosont haokgiound conditions
            Core-collection information:
                Height of tree core.
                Side of the tree from which the core was collected.
                Note whether duplicate samples were collected.
                Note whether an air blank was collected at the tree.
                Note any unusual core characteristics, such as a hollow or rotten interior of the tree.
Tree-Core

     Analysis of the tree-core samples involves allowing an
initial equilibration period for the VOCs in the tree core to
partition into the sampling media, typically headspace air. The
media is then analyzed by gas chromatography.

 1. Sample equilibration time: Once the tree cores are
    sealed in the sample vials, the VOCs associated with the
    tree core begin to volatilize into the vial headspace. The
    volatilization is fast enough so that detectable quantities
    of TCE can accumulate within the vial within minutes.
    In a field test for this investigation where the ambient
    air temperature was about 21 °C, analysis of the samples
    about 5 to 6 minutes after capping the vials was adequate
    to detect subsurface VOC contamination. In some cases,
    however, VOC concentrations in the tree-core vials can
    increase substantially by allowing the cores to equilibrate
    at room temperature overnight. Variations in  equilibration
    time can be caused by differences in ambient temperature,
    differences in heating of the core by friction during cor-
    ing, and other unidentified factors. If an important part of
    the investigation is  to compare concentrations among tree
    cores at a particular site, then allow enough equilibration
    time so that the VOC concentrations are less  sensitive to
    factors related to core collection. Overnight equilibra-
    tion at room temperature is a commonly used approach.
    Alternatively, heating the vials can  transfer larger amounts
    of the VOCs into the headspace, resulting in  an increased
   sensitivity. A simple block heater or water bath can be
   used in the field with a power inverter connected to a car
   battery to heat the cores in sealed vials at 60-70 °C for a
   few minutes. An advantage of field analysis of tree cores
   is that it can be used to direct the field sampling effort.

2, Sample analysis: In general, the simplest analyti-
   cal approach for tree cores in vials is to use headspace
   analysis (HSA) gas chromatography. The headspace can
   be analyzed by purge-and trap (PT) or by direct injection
   onto a GC column by syringe. The photoionization detec-
   tors (PID) and the electron-capture detector (ECD) are
   both useful gas chromatographic tools for tree-core analy-
   sis. A protocol for HSA of VOCs is included as Appen-
   dix 2 in this report.  The protocol reported in Appendix 2
   has not been tested  for all analytes reported to have been
   detected in tree cores in this report. Analytes not listed in
   the Appendix should be tested to determine the applicabil-
   ity of the method. A variety of VOC-conccntration report-
   ing units have been used in tree-coring investigations, but
   because semiquantitative data are typically sufficient, a
   simple approach is  to use a consistent sample-vial
   volume, collect a consistent core size, and report the
   results as  the volume of the VOC per billion volumes of
   ambient air.

-------
8     User's Guide to the Collection and Analysis of Tree Cores to       the Distribution of Subsurface Volatile Organic Compounds
                  and

    Four types of quality control and assurance samples
should be collected to maintain the integrity of the data.
The sample types are duplicate samples, air-blank samples,
background samples, and trip blanks.

 1. Duplicate samples: Duplicate samples consist of two
    core samples collected in separate vials. The samples
    should be collected from the same tree with the second
    core collected approximately 1 in. below the first core.
    The cores should be of equal  length. These samples pro-
    vide information on the variability of VOC concentrations
    caused by collection and analysis. The number of dupli-
    cate samples collected should be about 10 percent of the
    total number of trees cored.

 2. Air-blank samples: Air-blank samples are collected
    by rapidly waving an empty vial in the air in the vicinity
    of the target tree. The vial then is capped, transported,
    and analyzed in the same manner as the tree cores. This
    measurement provides information on the influence of
    air pollution on the sample quality. If TCE were present
    in analysis of a tree core, for example, but not in the air
    blank sample collected adjacent to the tree, then ambient
    air can be eliminated as the source of the detected TCE,
    demonstrating  that the TCE is associated with the core.
    Ambient air samples should be collected at various loca-
    tions across the study area and whenever there is a suspi-
    cion that airborne contaminants may provide interference,
    as is sometimes the case near active gasoline stations
    or dry-cleaning facilities. It also is possible that some
    level of VOC concentration may be present in the air by
    virtue of plant cvapotranspiration. In general, the dilution
    factors and wind influences are large enough that VOC
    concentrations in excess of air standards for gas-phase
    contamination  arc unlikely to be caused by phytovola-
    tilization; however, at least one investigation found that
    possible action-level exceedances might occur with highly
    toxic substances, such as VC and carbon tetrachloride, if
    they are present in ground water at levels above kilogram
    amounts in a single plume of a few hectares, and released
    by vigorously growing plants under hot, dry conditions
    (Narayanan and others,  2004).

 3. Background sample:  A background sample for a par-
    ticular tree  species consists of a core of that species col-
    lected from an uncontaminated area. Background samples
    are used because trees can contain natural VOCs, such as
    toluene, that can be detected during gas chromatographic
    analysis of the cores. The background sample ensures that
    target compounds detected in trees from a contaminated
    area arc not a misinterpretation of naturally occurring
    volatile compounds that elute on a gas chromatograph at
    the same time as the target compounds. A background
    sample should be analyzed for each tree species.
 4. Trip blanks: Trip blanks are air-filled vials sealed in a
    contaminant-free environment. The trip blanks are taken
    to the field and are kept with the tree-core samples once
    they are collected. The purpose of the trip blank is to
    determine whether exposure to target compounds during
    sample transportation could have resulted in false detec-
    tions of contaminants.
    Part 2 of this report provides an overview of historical
perspectives related to the use of tree coring with an emphasis
on application of the method to examining uptake of organic
contaminants. In addition, this section of the report provides
technical rationale for various aspects of the methodology
and examines factors influencing VOC concentrations in tree
cores.




    Tree cores have been widely used for dendrochronology
and other environmental applications since the early part of the
20th century when an American astronomer, A.E. Douglass,
related tree-core widths to climatic wet and dry periods
(Douglass, 1919). Examination of plants as a mechanism to
remediate subsurface organic contaminants dates back at least
to the early 1960s (Castelfranco and others, 1961).
    Early interest in the uptake of organic chemicals by
plants had an emphasis on agrochemicals (Shone and Wood,
1972, 1974). These authors used the transpiration stream
concentration factor (TSCF) to normalize the compound
concentration in the transpiration stream with respect to root-
zone bulk compound concentrations (Shone and Wood, 1974).
Briggs and others (1982) showed that TSCFs of nonvolatile
compounds were related to the properties of the chemical
being taken up by the plant, particularly, the degree to which
the compound was hydrophobic. Burken (1996) showed that
the same general factors control plant uptake of VOCs. Thus,
since the 1990s, it was clear mat VOCs could be taken up
into trees through the root system (Burken, 1996; Burken
and Schnoor, 1998; Complon and others, 1998; Davis and
others, 1998a; Newman and others, 1997; Vroblesky, 1998;
Vroblesky and others, 1999).
    The first application of tree-core chemistry to map VOCs
in ground water was at the Savannah River Site in South
Carolina (Vroblesky,  1998; Vroblesky and others, 1999). This
investigation involved tracking a chlorinated solvent plume
beneath a flooded cypress swamp. Headspace analysis of cores
from 97 trees (6 species, predominantly bald cypress [Taxo-
diutn dixtichutn]) growing over ground-water contamination in
a forested flood plain of the Savannah River in South Carolina
showed that rDCE and TCE concentrations in tree cores

-------
                                                       Part 2. Historical Perspectives and Technical Considerations    9
reflected the configuration of the ground-water contamination
plume, despite the fact that most of the trees were growing in a
few feet of uncontaminated standing water from the Savannah
River.
    Landmeyer and others (2000) extended the application
of tree cores from examining subsurface chlorinated solvents
to examining subsurface petroleum hydrocarbons. They
found MTBE and the conventional gasoline compounds
benzene, toluene, ethylbenzene, and the isomers of xylenes
and trimethylbenzene in cores from oak trees (Quercus sp.)
growing above petroleum-hydrocarbon contaminated ground
water. Additional evidence of MTBE uptake by trees was seen
in other investigations by examining biomass of trees (Brown
and others, 2001), transpiration gasses (Parfitt and others,
2000), and bioreactor experiments (Hong and others, 2001;
Ramaswami and Rubin, 2001).
    VOCs in tree cores have been used to delineate ground-
water contamination plumes in a variety of locations. Field
tests have been conducted in Colorado (Vroblesky and
others, 2004), Florida (Doucette and others, 2003), Maryland
(Bui-ken, 2001; Weishaar and Burken, 2005), Missouri (Schu-
macher and others,  2004), South Carolina (Vroblesky and
others, 2004), Texas (Vroblesky and others, 2004), and Utah
(Doucette  and others, 2003; Lewis and others, 2001). Trapp
and others (2007) investigated tree coring as a tool for screen-
ing subsurface pollution in Europe and published a concise
guide to field sampling. At least one study examined chloride
concentrations in tree rings to estimate the onset of chlorinated
hydrocarbon contamination (Yanosky and others, 2001). VOC
analysis of tree cores has been used to detect subsurface VOC
contamination and to direct subsequent drilling efforts in areas
where there were little or no pre-existing characterization data
(Schumacher and others, 2004; Vroblesky and Casey,  2004;
Sorek and others, 2008).
    VOC analysis  of tree cores has been used to monitor
ground-water plumes (Gopalakrishnan and others, 2005).
Tree cores have been used to show that some of the trees
at the leading edge  of a ground-water TCE contamination
plume began to take up TCE in increasing amounts over time
as the plume approached (Vroblesky and others, 2004). The
variety of applications and successful field tests indicate that
tree coring can be a viable reconnaissance tool for examining
subsurface VOCs.
    In some cases, analysis of stems and branches has been
shown to be a less intrusive approach to tracking subsurface
VOCs than collection and analysis of tree cores (Vroblesky
and others, 2004; Gopalakrishnan and others, 2007). In
general, however, the VOC concentrations in stems appear to
be lower than the VOC concentrations in tree cores, sometimes
resulting in stem analyses mat produce false negatives
(Vroblesky and others, 2004). A recent modification of the
tree-coring approach involves inserting activated carbon
into the core hole, followed by recovery  and analysis of the
activated carbon (Sheehan and others, 2007). In most cases,
simple HS A of tree cores is an adequate approach to locating
and mapping subsurface VOCs; however, the activated-carbon
approach has the potential to detect VOCs at lower levels than
HSA in situations where such sensitivity is needed. In areas of
high chlorinated VOC concentrations (greater than 1 pail per
million by volume [ppmvj in tree-core headspace vials) where
relatively little sensitivity is needed, simple field colorimetric
tubes have been used to detect the contaminants in tree cores
(Vroblesky and others, 2007b).


                        for

    In lhis section of the report, technical rationale is
provided for various aspects of the tree coring methodology.
In particular, this section includes discussions of parts of the
trunk to be sampled, maintenance of the core hole after core
collection, limes involved in core transfer to sample vials,
VOC stability in the sample vials, equilibration times, quality-
control issues,  and alternative approaches to collection and
analysis of tree cores for VOC analysis.


Tree-Core Collection

    The depth of coring depends on the length of core
desired. In most cases, il is not necessary to core deeply into
the trunk to obtain VOC concentrations. This is primarily
because most of the water flow during transpiration is in the
outer part of the trunk. In ring-porous trees (table 1), over
90 percent of water transported through the xylem is in the
outermost growth ring (Ellmore and Ewers, 1986). Thus,
inclusion of the outermost growth ring is particularly impor-
tant in  ring-porous trees. In diffuse-porous and nonporous
trees, multiple  growth rings conduct water. The preferential
conductance of water in the outer part of the tree does not
necessarily mean that the highest VOC concentrations are
always in the outermost part of the trunk. Often, the VOC con-
centrations decrease from the inner to outer parts of the trunk
(xylem), possibly related to volatilization loss through the
bark (Ma and Burken. 2003). The inner part, or heartwood, of
some trees is known to provide a waste repository for excess
concentrations of some  constituents (Tout and others, 1977;
Vroblesky and others, 1992), although this issue has not been
investigated for VOCs. Because increased concentrations of
VOCs  sometimes can be found in the inner relative to the outer
part of the trunk, there may be instances where coring more
deeply into the trunk can result in higher detection potential of
VOCs  in the tree cores.  Field investigations, however, indicate
that a core of the outermost 3 in. (not including the bark)
can be sufficient to identify subsurface VOC contamination
(Vroblesky and others,  1999, 2004)
    Research  indicates that sealing the tree core hole prob-
ably is not needed and sometimes can be harmful. Although
il is clear thai open boreholes can allow decay and disease
(Took and Gammage, 1959; Hart and Wargo, 1965; Shigo,
1967) and that tree wounds from increment borers can be
associated with long streaks of discolored and decayed wood
(Shigo, 1983),  plugging the core holes does little to reduce

-------
10     User's Guide to the Collection and Analysis of Tree Coresto Assess the Distribution of Subsurface Volatile Organic Compounds
discoloration or decay (Meyer and Hayward, 1936; Lorenz,
1944; Hepting and others, 1949). In some cases, the swelling
of dowel plugs inserted in the core hole has caused splits in
the trunk near the entry hole (Hepting and others, 1949). In
addition, wounds treated with wound dressing often form large
callus ribs that turn inward to form "ram's horns," and there
are no data to show that wound dressings stop decay (Shigo,
1983). In general, healthy dominant and co-dominant trees
respond well to tree coring both by creating a chemical barrier
to inhibit microbial invasion and by compartmentalizing
infected wood when microorganisms bypass the chemical
barrier (Shigo, 1974). Conifer species appear to be particularly
resilient to coring (Meyer and Hayward, 1936; Shigo,  1985).
A summary of the effects of coring on differing species  can be
found in Grissino-Mayer (2003). Core-damage studies showed
that more than half of the core holes healed within 2-3 years,
with most  of the poorly healing trees being short-lived species
or suppressed individuals (Meyer and Hayward, 1936, Lorenz,
1944; Hepting and others, 1949; Toole and Gammage, 1959).
Further, several studies reported no evidence of tree mortality
after increment coring (Meyer and Hayward, 1936; Lorenz,
1944; Hepting and others, 1949; Toole and Gamage, 1959;
Hart and Wargo, 1965; Eckstein and Dujesiefken, 1999; van
Mantgem and Stephenson, 2004) and little effect on tree
mortality when stem wedge sections were removed using a
chainsaw (Heyerdahl and McKay, 2001). Therefore, a practi-
cal approach is to leave the borehole unsealed. The open hole
allows the borehole to dry, and the water and sap flow from
the hole may cleanse the wound and discourage infection (Neil
Pederson,  Eastern Kentucky University, written commun.,
2007). Some researchers use antiseptic approaches to mini-
mize introduction of microorganisms to the core hole by either
dipping the increment borer in alcohol between trees (Neil
Pederson,  Eastern Kentucky University, written commun.,
2007) or by squirting antiseptic soap into the core hole (Lee
Newman, University of South Carolina, oral commun., 2007).
The effectiveness of these antiseptic approaches, however, has
not been determined. To reduce stress to the tree, excessive
coring of the same tree should be avoided.


Tree-Core Transfer to Sample Vials  and Storage

     Tree cores should be transferred to sealed vials as rapidly
as possible because volatilization loss from the tree cores
is rapid. A field test allowing the cores to remain in open
vials for several minutes prior to sealing showed that TCE
concentrations decreased by about 40 percent over a period of
5 minutes  (fig. 5).
     Other approaches to collect tree cores for analysis also
have been  used. Schumacher and others (2004) collected
replicate cores in vials containing 5 mL of organic-free water
and stored them upside down to limit volatilization loss
through the septa. Rather than preventing VOC loss, however,
the added water reduced the amount of detectable PCE.  Up to
55 percent less PCE was detected in the samples containing
water than in replicate samples not containing water. The
   ®   2,500


   >   2,000
 .c  c
  o.o   1,500
  o =
.y Q.
      _
     CD
     Q.
         1,000
           500
                                  Loblolly pine
                    0    2     4    6     8    10   12
                    Minutes vials containing
                 tree cores were left uncapped

  Figure 5.  Loss of trichloroethene over time from
  uncapped 20-milliliter serum vials, Naval Weapons Station
  Charleston, South Carolina, 2006.
authors attributed the difference to a combination of PCE
partitioning into the water phase and slightly longer time
required to avoid spilling water from the vials while adding
the core samples. Thereafter, the authors discontinued addition
of water to the vials.
    Methanol extraction also has been used for chemical
analysis of tree cores (Landmeyer and others, 2000; Lewis and
others, 2001). In the Landmeyer investigation, the cores were
sealed in vials containing 5 mL of reagent-grade methanol.
Methanol extraction used in combination with PT analysis can
be an  effective approach to confirm the identity of the detected
compound by mass spectrometry.
    An additional approach to collecting tree cores that has
been reported is chilling the cores upon collection, transport-
ing them back to the laboratory, freezing them until analysis,
and then crushing the cores prior to analysis (Sorek and others,
2008). In the laboratory, the crushed cores were heated prior to
analysis.
    Changes in VOC concentrations can take place in the
tree-core sampling vial over time. Sometimes these changes
can be seen as changes in the amplitude of early-eluting
chromatographic peaks typically not associated with ground-
water contamination. To test the length of time that the tree
cores  for TCE analysis can be stored in sample vials prior to
analysis, repeated sampling was done with a series of vials
stored at room temperature, containing cores from a tree
growing above contaminated ground water.  Most vials showed
no TCE volatilization loss from 24 hours to  19 days of storage
(fig. 6). These data indicate that the vials can be stored for
at least a few days prior to analysis. It should be cautioned,
however, that it is prudent to analyze the samples within a
few days of collection rather than waiting 19 days to avoid
potential transformations or volatilization losses not evident in
this investigation.

-------
                                                       Part2. Historical Perspectives and Technical Considerations     11
          CO  ®
          Q.  E
          — "o
          c"  i
          0) -Q
          £  C
          "CD  o
          o =
          o
              0)
                  10,000
                                 123456789
                                                         Vial number

          Figure 6.  Headspace trichloroethene concentrations at 24 hours and 19 days of storage in sealed serum vials
          containing tree cores from a trichloroethene-contaminated site showing generally slight concentration increases
          in most cores and no evidence of volatilization loss overtime.
Tree-Core Equilibration Time

    VOCs begin to de-gas from tree cores immediately upon
removal from the tree. In field investigations conducted for
this user's guide, sufficient TCE concentrations to indicate the
presence of subsurface contamination accumulated in crimp-
capped serum vials containing tree cores within 5 minutes
of collection. In tests where repeated vapor extractions were
taken from the same tree-core vial over time, several vials
showed that the TCE concentration in the vial headspace
after 5 minutes of equilibration was approximately the same
as allowing the vials to equilibrate overnight (fig. 7A, B,
C, D). In other tree-core vials and on other sampling dates,
however, TCE concentrations in the vial headspace increased
by a factor of 2 or more after equilibrating overnight (fig. 7E).
The factors controlling equilibration time probably include
ambient air temperature, differences in heating of the core
barrel by friction during coring, and other unidentified factors.
The influences of these factors are not yet well understood.
Thus, in general, it appears that analysis of tree cores within
5 minutes of core collection can be  a useful indicator of
subsurface TCE contamination. If the intent, however, is to
compare VOC concentrations among several trees at the site,
a prudent approach, until the factors that control equilibrium
time are more thoroughly studied, is to treat all of the cores
the same way and allow them to equilibrate approximately
24 hours or more (Landmeyer and others, 2000; Schumacher
and others, 2004; Vroblesky and others, 2004). A comparison
of cores allowed to equilibrate for 24 hours relative to 19  days
shows that only slight concentration increases took place after
24 hours and there was no evidence of VOC loss (fig. 6).
    An alternate field approach is to analyze the tree cores
in the field after heating the cores in crimp-capped serum
vials for about 5 to!5 minutes. Field tests for development of
this guide showed that field heating the cores produced TCE
concentrations higher than in unheated cores or cores allowed
to equilibrate at ambient temperature for 24 hours (fig. 7D).
A similar relation was seen in field-heated cores with a
reanalysis of the core after about 24 to 28 hours of equilibra-
tion at ambient temperature following heating (fig. 7F, G, H).
Although it is possible that the lower TCE concentrations
in the corresponding unheated sample the next day could
represent volatilization loss during the previous heating,  such
loss probably is minor because similar TCE-concentration
declines between the 5-minute time step and the unheated
analysis the following day were  seen whether the samples
were heated for 40 minutes or for only 5 minutes (fig. 7G, H).
Tests have not yet been done to determine the upper range of
acceptable heating, but it is logical that some level of volatil-
ization loss or thermal destruction could occur with excessive
heating. Therefore, core heating should be maintained within
the ranges cited above or should be tested.
    In additional investigations of field heating conducted
for this guide, duplicate cores were collected 0.75 in. apart
from trees at Solid Waste Management Unit (SWMU) 17,
Naval Weapons Station Charleston, South Carolina,  in June
2006 and at the Durham Meadows Superfund Sites, Durham,
Connecticut, in August 2006. At each site, one of the cores
was heated for 5 minutes at 60-70 °C and analyzed in the
field within 10 minutes of collection. Comparison of VOC
concentrations in vials containing tree cores shows that field
heating the cores, in most cases, produced concentrations that
were in the range of, or higher than, concentrations obtained
after allowing the vials to equilibrate 24 to 30 hours at room
temperature (fig. 8A). Although only two trees were tested for
PCE, the limited data set implies that a similar correspondence
applies to PCE (fig.  8B).

-------
12     User's Guide to the Collection and Analysis of Tree Coresto Assess the Distribution of Subsurface Volatile Organic Compounds
    C
0)  O

5r —  Q)
•= -P  >
500


400





200


100


  0
                                    A. Tree T1 6,

                                     10/15/2007
                                in  in  in  in  in  in  in
                                CD  CD  CD  CD  bi  ±i  ±i
                                4->  4->  4->  4->  =5  =5  =5
                                3  3  3  3  O  O  O
                                C  C  C  C  .C  .C  .C


                                E  E  E  E  f  ••-  °°

                                =  =  =  =  £  "  "

                                   T-  i-  CN   4-> =5
=5 =5 0
                                                                                   500


                                                                                   400


                                                                                   300


                                                                                   200


                                                                                   100


                                                                                     0
                                                                         C. TreeT16,

                                                                          10/15/2007
                                                                               zi
                                                                            =5  O
                                                    E   E
                                                    in  i-
                                                co
                                                CN
                                               D. Tree T100,

                                          Heated at 70°C, 5/3/2007
                      H  C
2,500
2,000
1,500
1,000
500
n
E. TreeTI,
4/16/2006









— I



— |




-
-
-
-
-
                                      ininaiaiininininai
                                      E   E  E  E   E  E

                                      CD  CD  00  10  M-  ^T
                                                                                    =5  =5
                                                                                    o  o
                                                                                    in  in
                                                                                    ,_j  CM
                                                                             =5
                                                                             o
                              F. TreeT16,

                            Heated at 70 °C
                                                G. TreeT16,

                                         Heated at 50 °C, 10/15/2007
                     500
                                                                     H. TreeT16,

                                                              Heated at 50 °C, 10/15/2007

                                                              500
                                                                                       400  -
                                                200  -


                                                100  -
                                                       (/)(/)(/)(/)(/)(/)(/)(/)(/)
                                                       QJQJQJQJQJQJQJQJ1—
                                                       EEEEEEEE°>

                                                      inocD-^-inoino
                                                         T- i- CN CN CO  CO  M-
                                              Time since tree-core collection
                                                         EXPLANATION

                                                Heated prior to analysis

                                                Unheated following the last heated analysis

                                          i     i Unheated

                                           (A)   Core identifier

                                           Each graph represents multiple analyses of the

                                           same core, with the exception of graph  D.


          Figure 7.  Trichloroethene concentrations over time in unheated and heated sealed serum vials containing

          loblolly pine tree cores from a trichloroethene-contaminated site showing little concentration change in cores

          from some trees after 5 to 6 minutes and an increase in sensitivity by field heating and analyzing the cores. Solid

          Waste Management Unit 17, Charleston, South Carolina.

-------
                                                      Part 2. Historical Perspectives and Technical Considerations    13
          "5.2
            Efl)
            t!— Q)
          co a E
                   10,000
               o   1,000
          £ N C
          -t= .>> O
          C CD =
          CD CD w
          •fc= T3 t
          C CD CD
          Sro ^
          c 
-------
14     User's Guide to the Collection and Analysis of Tree Coresto       the Distribution of Subsurface Volatile Organic Compounds
that although the methanol extraction produced higher TCE
concentrations than the HSA, there are costs and analytical
complications associated with the methanol extraction
analyses mat make HSA a preferable alternative.
    Where increased sensitivity is needed relative to HSA,
activated carbon has been inserted into a tree-core hole,
followed by recovery and analysis of the activated carbon
(Sheehan and others, 2007). This approach utilizes the
tree-core hole rather than the tree core.
    In a highly contaminated area where relatively little
sensitivity is needed, field colorimetiic gas-detector tubes
(GasTec 133LL) have been used to rapidly and inexpensively
detect VOCs in tree cores (Vroblesky and others,  2007b).
This approach was used as part of the development of this
field guide and utilized a simple modification of the Color Tec
screening method (Kelso, 2005) to employ vials containing
tree cores in air rather than the recommended vials containing
water or sediment in water. The relative concentrations of total
chlorinated VOCs obtained by the colorimetric method com-
pared favorably to the quantitative analysis of total chlorinated
VOCs obtained by gas chromatography at concentrations
greater than 1 pprnv (r2 = 0.05). Below total chlorinated  VOC
concentrations of 1 ppmv, the colorimetric method sometimes
detected VOCs and sometimes did not. Below total chlorinated
VOC concentrations of 0.7 ppmv, the colorimetric method
failed to detect VOCs.  The gas chromatographic method
remained useful for detecting total chlorinated VOCs down to
at least 10 ppbv.
    Comparison of the colorimetric method to the gas
chromatographic approach of field analyzing tree cores for
chlorinated VOC content shows that both methods can provide
effective detection of subsurface  chlorinated VOCs at high
concentrations (greater than 1 ppmv). At some sites, such as
the Carswell Golf Course in Fort Worth, Texas, this level of
detection is inadequate to detect the ground-water contamina-
tion (Vroblesky and others, 2004). At other sites,  however,
such as  Solid Waste Management Unit 12 (tree-core TCE
concentrations up to 10.19 ppmv) and Solid Waste Manage-
ment Unit 17 (tree-core TCE concentrations up to 85 ppmv) at
the Naval Weapons Station Charleston in South Carolina and
at Air Force Plant PJKS in Colorado (tree-core TCE concen-
trations up to 2.191 ppmv) (Vroblesky and others, 2004), field
analysis by using either gas chromatography or colorimetric
method would be a viable means of locating the plume.
Disadvantages of the colorimetric approach are that (1) it is
substantially less sensitive than the field gas chromatograph
and may miss pails or all of ground-water contamination
plumes that are not reflected by relatively high ttee-core VOC
concentrations, and (2) the colorimetric approach is sensitive
only to chlorinated VOCs with a larger influence  of chlori-
nated alkenes relative to chlorinated alkanes. An advantage
is that the colorimetric approach  is simple to use,  easily  field
portable, and inexpensive (less than about S10 per sample).
At sites with sufficient concentrations of tee-core chlorinated
VOCs, the colorimetric approach can be a viable, inexpensive
reconnaissance tool without the need for a field GC.
Quality Control and Assurance

    A field test conducted for this investigation in October
2005 to determine whether decontamination of core barrels
was required between collecting cores from different trees
showed that the decontamination was unnecessary for
investigating TCE in tree cores. The test involved collecting
a core from the trunk of a loblolly pine known from previous
work to contain high concentrations of TCE. The core was
immediately sealed in an empty 20-mL crimp-top vial. The
core barrel was not cleaned following collection of the core.
Within 6 minutes of collecting the core from the contaminated
tree, a core was collected from a loblolly pine of similar
diameter in a background area. This sequence was repeated
two more times so that a total of three cores were collected
from the contaminated tee and three from the uncontaminated
tree. At no time during the  coring was the core barrel cleaned.
The cores were allowed to equilibrate with the headspace of
the sample vials for approximately 24 hours at approximately
25 °C and then were analyzed by HSA using photoionization
gas chromatography.
    TCE concentrations in the contaminated tree ranged from
5,000 to greater than 10,000 ppbv (table 3). TCE was not
detected in cores from the background tree at a detection limit
of about 15 ppbv. Thus, the core barrel showed no evidence of
TCE carryover from the contaminated to the uncontaminated
cores. The data indicated that decontamination of the core
ban-el prior to sampling the background tee was unnecessary
when investigating TCE. The probable reasons for the lack of
carryover include the low sorption potential of VOCs to metal,
the heat generated during the coring, and the volatility of the
compounds. It should be cautioned, however, that the core
ban-el should be inspected to ensure that mere is no paniculate
carryover, such as sections  of tree core, remaining in the core
 Table 3.   Trichloroethene concentrations in tree cores
 collected within about 5 minutes of each other, using the same
 core barrel with no decontamination between cores, showing
 no carryover between core collections.
 [ND. not detected at 15 parts per billion by volume; >, greater than]
Tree
Background tree
Contaminated tree
Background tree
Contaminated tree
Background tree
Contaminated tree
Background tree
Air sample near contaminated tree
Air sample near background tree
Collection
time,
in hours:
minutes
13:55
14:00
14:04
14:09
14:14
15:10
15:19
15:10
15:19
Trichloro-
ethene,
in parts per
billion by
¥olume
ND
7,649
ND
5,000
ND
> 10,000
ND
ND
ND

-------
                                                       Part 2.  Historical Perspecti¥es and Technical Considerations     15
barrel that could adversely impact subsequent samples. In
addition, TCE was the only compound examined during this
test of potential carryover. Although the results technically
apply only to TCE, they probably are applicable to other
chlorinated solvents, based on their chemical similarity
to TCE.



                   in

     A variety of factors influence the ability of plants to be
useful indictors of ground-water VOC contamination. These
factors include the type of VOC, the tree species, the rooting
depth, aqueous concentrations, the depth to the contaminated
horizon, concentration differences around the trunk related
to different sources of subsurface VOCs, concentration
differences with depth of coring related to volatilization loss
through the bark and possibly other unknown factors, dilution
by rain, seasonal and climatic influences, sorption, and
within-tree VOC degradation. The following sections discuss
these factors in greater detail.


Types of Volatile Organic Compounds

     Various VOCs are known to be taken up by plant roots
into the trunks of trees and are, therefore, probable candidates
for tree-coring investigations. VOCs that have been found
in tree-coring investigations include benzene, toluene,
ethylbenzene, xylene isomers, trimethylbenzene, MTBE,
1,1,-2-2-tetrachloroethane (PCA) (Hirsh and others, 2003),
1,1,1-trichloroethane, 1,1-dichloroethene, carbon tetrachloride
(Sorek and others, 2008), VC (Trapp and others, 2007), TCE,
PCE, and cDCE (Burken and Schnoor, 1998; Nietch and
others, 1999; Vroblesky and others, 1999; Landmeyer and
others, 2000; Burken, 2001; Davis and others, 2003). Sorek
and others (2008) found that 1,1-dichloroethene appeared
to be rarely detected in tree cores despite relatively high
concentrations in the subsurface, possibly due to being lost by
volatilization from the trunk and sampled tree cores.
     Direct uptake of contaminants is controlled by a variety
of factors, but in general, moderately hydrophobic organic
compounds (octanol-water coefficient, log Kow= 0.5-3), such
as TCE and cDCE, readily enter the vegetation transpiration
streams (Briggs and others,  1982, 1983; Schnoor and others,
1995). Hydrophobic chemicals (log Km greater than 3.5) are
too strongly bound to roots and soil to be translocated within
plants (Briggs and others, 1982; Schnoor and others, 1995).
Early work considered very water soluble chemicals (log Kow
less than 0.5) to be neither sufficiently sorbed to roots nor
passively transported through plant membranes (Briggs and
others, 1982; Schnoor and others, 1995); however, a more
recent investigation provides evidence that soluble, highly
polar compounds (such as sulfolene with a log Kmof -0.77)
can be readily taken up by plant root systems (Dettenmaier
and others, 2008). Thus, log Kow (table 4) is an important
factor influencing the ability of a compound to be translocated
up the tree trunk.
           Table 4.  Chemical formula, Henry's Law constant, and log octanol-water partition coefficients for selected
           volatile organic compounds reported in tree-coring investigations,
           [atm nr'/mol, atmosphere cubic meter per mole]
Compound
Tetrachloroethene
Trichloroethene
1 , 1 -Dichlorocthcnc
cis- 1 ,2-Dichloroethene
Vinyl chloride
1.1. 1 -Trichlorethane
1 . 1 -Dichloroe thane
Carbon tetrachloride
Benzene
Toluene
Ethyl benzene
Xylene isomers
/M-xylene
p -xylene
Methyl ferf-butyl ether
Formula
C2C14
C2HCL,
C2H2C12
C,H2CL,
C2H,C1
C2H3C13
C2H,C12
CC14
C6I-I6
C7HS
CSHK,
C8H!0
<'8H10
C8H!0
C5H12°
Henry's Law constant
(atm m3/mol)
1.53xlO-2
9.10x10-3
l.SxlO-2
3.37x10 3
1.22x10-'-°°
1.62xl02
5.42x10-3
2.40xlO-2
5.40x10-3
6.70x10 3
6.60x10-3
5. 27x10- to 7. IxlO-3
7.00x10-3
7.10x10-3
5.4xlO-4
log octanol-water
coefficient
2.60
2.03
1.84
1.86
0.60
2.18
1.79
2.78
2.12
2.65
3.13
2.95 to 3.2
3.20
3.18
1.24

-------
16     User's Guide to the Collection and Analysis of Tree Coresto       the Distribution of Subsurface Volatile Organic Compounds
     In general, the log K^ con-elates well with the root
concentration factor (RCF) and the transpiration stream
concentration factor (TSCF). The RCF describes the efficiency
of solute movement from an external solution into the root
system and is defined by Shone and Wood (1974) as:

     RCF = (Concentration in roots) / (Concentration      (1)
           in external solution).

     The TSCF is defined by Shone and Wood (1974) as:

     TSCF = (Concentration in the transpiration stream) /  (2)
             (Concentration in external solution).

TSCF values have been measured for a number of compounds
(Briggs and others,  1982; Burken and Schnoor,  1998; Davis
and others, 1998b; Inoue and others, 1998). A TSCF of 1.0
indicates unrestricted passive uptake. A TSCF of less than 1.0
indicates some degree of exclusion by the plant, and a TSCF
of greater than 1.0 indicates active uptake by the plant.
     Various TSCF values for TCE have been reported. Davis
and others (1996) reported a TSCF of 0.67 (at 1.5 grams
per liter [g/L]) for TCE. Orchard and others (2000) found
TSCF values of 0.02 to 0.22 with an average of 0.12 for TCE.
Lockheed Martin (2000) used a TSCF value of 0.79 for TCE.
The wide range of TSCF values probably reflects different
experimental setups, such  as hydroponic rather than soil
growth or flowthrough chamber rather than static chambers,
leaks, and other factors.
     Doucettc and others (2003)  found that depending on the
climate, 200 to 1,400 liters per square meter per year (L/m2/yr)
probably represents a reasonable range of annual transpiration
values, and Wullschleger and others (1998) reported that
90 percent of the observations for maximum rates of daily
water use were between 10 and 200 liters per day (L/d)  for
individual trees that averaged 70  ft in height. The authors
concluded that reasonable values for yearly TCE plant uptake
from a ground-water TCE concentration of 1 mg/L are 2.4-
84 milligrams per square meter per year (mg/m2/yr) (TSCF
value of 0.12) to 525 mg/nf/yr (TSCF value of 0.75)  depend-
ing on the choice of TSCF values (Orchard and others, 2000).
     The mass of TCE removed by plant uptake can be
modeled by

Mass = (TSCF) (CTCE) (T)  (f) (Doucette and others, 2003), (3)

where TSCF is assumed to be constant,  CTCE is the average
ground-water concentration of TCE in milligrams per liter, T
is the cumulative volume of water transpired per unit area per
year in liters per square meter per year, and / is the fraction
of the plant-water needs met by contaminated ground water.
It should be cautioned, however,  that TSCF does not include a
vapor-transport term for uptake or loss and, thus, may provide
misleading conclusions in situations where vapor transport is
significant.
     The / factor is difficult to measure at most phytoremedia-
tion sites described in the literature because ground-water
use by plants tends to decrease as the availability of surface
water increases (Nilsen and Orcutt, 1996). Doucette and others
(2003) suggest that a range / from 0.1  to 0.5 is probably
reasonable for climates with more than 16 in. of annual rain.
     Although the equations based on log Kow provide a
generally useful predictive tool for examining uptake of
organic compounds by plants, numerous exceptions exist
(Burken and Schnoor, 1998). There are classes of compounds
(such as nitroaromatics, phenols, and aromatic amines) that are
more tightly bound to roots than predicted by the RCF because
the sorption is related to biochemical bonding rather man to
hydrophobic partitioning behavior. Binding to the roots for
these compounds, such as aniline, nitrobenzene, catechol, and
chlorobenzene, is irreversible (Dietz and Schnoor, 2001).


Subsurface Volatile Organic  Compound
Concentrations

     Field investigations  (Vroblesky and others, 1999, 2004;
Schumacher and others, 2004) showed that the highest
VOC concentrations in tree cores usually were found in
trees growing above the highest ground-water or soil VOC
concentrations, as indicated by samples from ground-water
wells or soil-vapor surveys. Additional evidence for the
correspondence between environmental VOC and tree-core
VOC concentrations was shown in laboratory studies (Ma and
Burken, 2002, 2003). Thus, it is clear that subsurface VOC
concentrations can directly influence VOC concentrations in
tree vascular tissues.
     In one study, comparison of PCE concentrations in a
number of tree-core samples and sediment samples 12 ft deep
showed a linear relation for soil-PCE concentrations greater
than  60 micrograms per kilogram (ug/kg) (Schumacher and
others, 2004). Therefore, the PCE concentration in tree cores
was found to be a good predictor of PCE concentrations in  soil
at 12 ft deep. In general,  however, predictions of subsurface
VOC concentrations based on tree-core results should be
considered as a qualitative rather than quantitative relation and
indicative of minimum concentrations.
     In some cases, VOC concentrations in tree cores appear
to correspond more closely to soil-gas VOC concentrations
than  to ground-water VOC concentrations (Schumacher and
others, 2004). This finding raises the possibility that tree-core
analysis may be useful as a rapid, inexpensive, relatively
low-profile, non-intrusive reconnaissance tool to identify areas
of potential vapor intrusion to be targeted by more definitive
and cumbersome investigative approaches. In a limited study
at the Nyanza Chemical Waste Dump Superfund Site, Mas-
sachusetts, tree cores were examined as a possible indicator
of vapor intrusion (Vroblesky and others, 2006) (fig. 9). Trees
N3, N5, and N8 contained TCE and  were adjacent to buildings
in which vapor intrusion  by TCE had been identified.
     In a separate study at the Durham Meadows Superfund
Site,  Connecticut (Vroblesky and others, 2008), TCA in tree
trunks corresponded to TCA in soil gas, although the two stud-
ies were done 3 years apart (fig. 10). Thus, multiple studies

-------
                                                        Part 2. Historical Perspectives and Technical Considerations     17
                                                         \,
                     PLEASANT STREET JM9
                                         §
MW-302
      EXPLANATION
Monitoring well and label

Tree-coring location and label. Blackfill
indicates positive detection of vapor
intrusion, gray indicates no detection.
                                                             I	1  Building. Blackfill indicates positive
                                                                   detection of vapor intrusion, gray indicates
                                                                   no detection, white indicates no data.

                                                                                           400 FEET
                  Figure 9.  Trees cored at the Nyanza Chemical Waste Dump Superfund Site, Ashland,
                  Massachusetts, August 2006 (modified from Vroblesky and others, 2006), showing proximity
                  to wells and to buildings where vapor-intrusion investigations were conducted in 2004 (ICF
                  Consulting, 2005).
                                                               Existing
                                                    '0—DM7 \  buildin9
         MERRIAM MANUFACTURING
            COMPANY PROPERTY
      25    50    75
0   5   10  15   20   25METERS
                                                         Basemap modified from Metcalf & Eddy, Inc. (2006)
                                      EXPLANATION
           1,1,1-Trichlorethane in soil vapor, in parts per billion by volume .combination
           of soil gas surveys in 2003 and 2006 (Anni Loughlin, U.S.  Environmental Protection
           Agency, written commun., February 2007)
     / \  Tree containing less than 1.6 parts per billion by volume of 1,1,1-trichloroethane

DM5 / \  Tree containing 5.6 parts per billion by volume of 1,1,1-trichloroethane

DM7 ^^  Tree containing 24 parts per billion by volume of 1,1,1-trichloroethane

     I    I  Former areas of liquid storage in tanks and former paint booths
      o14   Soil-gas sampling point. Black indicates 2003 and open indicates 2006 sampling.
                           Figure 10.  1,1,1-Trichloroethane
                           concentrations in tree cores in
                           2006 showing correspondence to
                           the combined results of 2003 and
                           2006 soil-gas investigations at the
                           Merriam Manufacturing Company
                           property, Durham Meadows
                           Superfund Site, Connecticut.

-------
18     User's Guide to the Collection and Analysis of Tree Coresto       the Distribution of Subsurface Volatile Organic Compounds
indicate that tree coring is a potential reconnaissance tool to
identify areas of soil-gas and vapor-intrusion hazard.


Differences Among Tree Species

    Plant utilization of ground water is partly dependent
on plant species (Smith and others, 1991; Busch and others,
1992; Thorburn and Walker, 1994; Kolb  and others, 1997).
Studies also have shown that the degradation and bioavail-
ability of contaminants in soil  systems can vary with plant
species (Shann and Boyle, 1994). In addition, comparisons of
increment cores from trees of differing species growing near
each other sometimes show VOC concentration differences
that appear to be species-related. In a study in South Carolina
(Vroblesky and others, 1999),  oaks consistently contained less
TCK than adjacent bald cypress or loblolly pines. In the same
study, however, adjacent bald cypress and tupelo trees (Nyssa
sp.) showed no significant differences in TCE concentrations
from increment cores, indicating similar uptake of TCE. A
possible contributing factor to  the differences among some
species is that conifers, such as bald cypress and loblolly pine,
conduct water through more than the outermost growth ring,
whereas in ring-porous trees (table 1), nearly all of the water
is conducted through the outermost growth ring (Kozlowski
and others, 1966; Ellmore and Ewers, 1986). Thus, the higher
concentrations detected in conifers relative to the oaks may
be because the cores, being of approximately equal length,
incorporated more of the transpiration stream in conifers than
in ring-porous trees. In another study, similar TCE concentra-
tions were observed in increment cores in 1998 and 2000
from a willow (Salix sp.) and an eastern cottonwood (Populus
delta ides) growing directly adjacent to each other (Vroblesky
and others, 2004). Sorek and others (2008) found order of
magnitude differences in TCE concentrations in a rosewood
(Dalbergia sissoo) and laurel fig (Ficus microcarpa), despite
comparatively small TCE concentration differences in the
subsurface.
    In one set of trees spaced within 30 ft of each other,
Lewis (2001) found that the relative amount of TCE uptake
among tested species appeared to follow the trend cottonwood
(Populus delta ides) > Russian olive (Elaeagnus angustifo-
/za)> poplar species (Populus sp.). There is some uncertainty
with this conclusion, however, because of data limitations,
such as the fact that only one of each species was compared,
there were tree-diameter differences, and there were potential
subsurface influences and spatial variations.


Rooting Depth and Depth to the
Contaminated  Horizon

    Rooting depth, or the proximity of the roots to the
contaminated horizon, is a factor that potentially can influ-
ence VOC uptake  in trees. To some extent, rooting depth is
species-dependent. Trees of different species (Ehleringer and
others, 1991) and even different size trees of the same species
(Dawson and Pate, 1996) can obtain water from different
sources. Some species, such as poplars, and willows, are
genetically predisposed to develop roots extending to the
water table or capillary fringe at depths ranging from 3 to
40 ft (Negri and others, 2003). In general, however, trees
with the capability to root deeply will do so only if mere is
a hydrologic need to do so. In rainy climates where mere is
adequate water available for the plants from the soil zone,
rooting depth will be limited. In some cases, where the depth
to ground water increases, there is a corresponding increase in
rooting depth and decrease in the ground-water contribution to
the plant water use (Sepaskhah and Karimi-Goghari, 2005).
     Few studies have rigorously examined rooting depth
because of the difficulty in uncovering roots. Descriptions of
rooting depths vary widely. A study in the United Kingdom
of five willows and five poplar clones in differing soil types
showed that although the rooting depths were more than 4 ft,
75 to 95 percent of the roots were in the top 14 in. (Keller and
others, 2003). A study in central Texas used DNA sequencing
of roots in caves 16 to 213 ft deep to examine the rooting
depth of various species (Jackson and others, 1999). The
tested species were southern hackberry (Celtsi laevigata), ash
juniper (Junlperus ashei), white shin oak (Quercus sinuata),
Texas live oak (Quercus fusiformis), American elm (Ulmus
americana), and cedar elm (Ulmus crassifolia), and represent
approximately three-fourths of the woody plants comprising
the studied ecosystem. At least six tree species grew roots
deeper than 16 ft, but only Texas live oak was found below
32 ft. The maximum rooting depth for mat ecosystem was
about 82 ft. The oxygen 18 (1SO) isotopic signature for stem
water from a live oak confirmed water uptake from a depth of
59 ft.
     The degree to which the roots are in intimate contact with
the contaminated horizon appears to be an important control
on the amount of contaminant uptake. A study was conducted
in Colorado to examine eastern cottonwood trees that were
about the same diameter (Vroblesky and others, 2004).
Vroblesky and others (2004) found that a core from an eastern
cottonwood where the depth to TCE-contaminated ground
water (200 ug/L TCE) was about 24 ft contained 99 ppbv of
TCE. In contrast, the TCE concentrations were substantially
higher (2,191 and 552 ppbv) in cores from two eastern cot-
tonwoods growing at the bottom of a creek erosional channel
where the ground water was less than 3 ft deep, despite the
presence of only 29 to 39 ug/L TCE in the ground water. The
data strongly  imply that the percentage of transpiration uptake
composed of  contaminated water was higher in trees in the
drainage ditch where the roots were in more close contact with
the contaminated ground water than in the upland trees where
the depth to water was about 24 ft.
     Tree coring has been used in field investigations to
successfully detect subsurface VOCs in areas where the
ground water was 20 to 25 ft deep (Schumacher and others,
2004; Vroblesky and others, 2004) and 59  to 65 ft deep (Sorek
and others, 2008). In some studies, however, poor quantitation
between tree VOCs and ground-water VOCs was attributed to

-------
                                                       Part 2. Historical Perspeeti¥es and Technical Considerations    19
a relatively large depth to water (Cox, 2002; Schumacher and
others, 2004). The probable causes for the poor correlation
in areas of large depth to water include the lack of intimate
contact between the contaminated ground water and the tree
roots, the potential for tree roots to obtain soil water from
shallower horizons when it is available (Mensforth and others,
1994; Thorburn and Walker, 1994; Dawson and Pate,  1996;
Jolly and Walker, 1996), and the diffusion of VOCs out of
the roots during upward transport (Struckhoff, 2003).  The
decrease in TCE concentration with increasing depth to water
is consistent with predictions from ground-water modeling
(Wise, 1997) and from observations that ground water can
provide less of a contribution to plant-water use when it is
deep or when shallower soil moisture is available (Zencich and
others, 2002; Sepaskhah and Karimi-Goghari, 2005).
    Perhaps a more meaningful conceptualization of rooting
depth is the depth of hydraulic influence of the roots. An
example is the engineered phytoremediation site using hybrid
poplar (Populus deltiodes x Populus trichocarpa) at Aberdeen
Proving Ground, Maryland, where the depth to the water
table is 5 to 15 ft (Hirsh and others, 2003). A ground-water
study at the site showed mat water use by the poplar trees
induced upward ground-water hydraulic gradients toward the
roots, with the depth of hydraulic influence extending to 25 ft
(Schneider and others, 2002). Thus, in some environments,
trees can induce movement of water and the associated
dissolved contaminants upward to the roots from areas beyond
the physical extent of the roots.
    It should also be noted that sites that have ground water
at a depth that appears to be beyond the reach of the tree roots
still may be viable candidates for use of tree coring as a tool to
investigate subsurface VOCs. Studies have shown that soil gas
can be an effective transport mechanism of VOCs  to tree roots
(Struckhoff, 2003; Schumacher and others, 2004; Struckhoff
and others 2005a, b) and may explain some of the VOC detec-
tions in tree cores where a relatively large depth to water was
reported. Sorek and others  (2008) found chlorinated solvents
in tree cores where the depth to the contaminated ground water
was 59-65 ft, and where the same chlorinated solvents were
present as soil gas in the vadose zone. In addition, hydraulic
lift has the potential to transport contaminants from deeper
to shallower parts of the soil where the water and solutes can
be accessed by shallower rooted species. Hydraulic lift is a
process by which differences in water potential allow  trees
to derive water from deep,  wet roots and lose water through
shallow dry roots. (Richards and Caldwell, 1987; Caldwell
and others, 1998). Finally,  tree roots tap the ground water
indirectly by extending into the capillary  fringe. Thus, in
areas where the capillary fringe is large, tree roots can derive
water from the ground water even when they appear to be too
shallow to access it.


Subsurface Lithology

    The subsurface lithology can have an influence on the
ability of trees to be useful tools for examining VOCs in
ground-water contamination. In general, tree roots extend to
the depth necessary to maintain a water supply adequate for
growth. If a confined contaminated aquifer is overlain by an
unconfined uncontaminated aquifer, then it is unlikely that the
trees will be useful indicators of the contamination because
sufficient water can be obtained from a shallower source. If
the confining layer is absent, however, men it is possible that
the trees can sample the contaminated ground water, despite
the fact that the contaminated ground water is overlain by a
veneer of uncontaminated ground water. Field investigations
found mat roots from poplar trees induced upward hydraulic
gradients toward the roots (Schneider and others, 2002;  Hirsh
and others, 2003).
     The presence of a confining layer does not neces-
sarily limit the use of tree coring as a tool for examining
subsurface VOCs. A study in South Carolina examined  the
TCE concentrations in tree cores growing above a confined
TCE-contaminated aquifer where a 9- to 10-ft-thick tight
clay confining layer extended to land surface. In mis case,
the availability of shallow ground water for use by the trees
was limited by presence of the clay. The roots of the trees
in this area extended to the aquifer below the clay in order
to maintain adequate water supply, as evidenced by the
presence of live roots in multiple sediment cores below  the
clay (Vroblesky and others, 2007a). Despite the fact  that
the contaminated aquifer was confined beneath about 9  to
10 ft of tight clay, the distribution of trees containing TCE in
tree cores closely matched the distribution of ground-water
contamination (Vroblesky and others, 2004). The concept
of tree roots seeking out sources of adequate water supply
indicates that in fractured-rock environments where there is
little saturated overburden, tree roots will preferentially  sample
the water-bearing fracture zones, potentially optimizing their
use as indicators of shallow ground-water contamination in
this hydraulically complex setting.


Concentration Differences Around the Tree  Trunk

     Studies have reported VOC concentration variations
from cores collected on different sides of the same tree trunk
(Vroblesky and others, 1999, 2004; Lewis, 2001; Schumacher
and others, 2004; Sorek and others, 2008). Vroblesky and oth-
ers (1999) found concentration differences ranging from 44 to
92 percent for TCE and from 6 to 90 percent for cDCE in
cypress trees. The same trees, however, showed relatively good
replication for cores collected 1 in. (approximately 25 mil-
limeters (mm) apart (15.5 percent for TCE and 2.5 percent for
c'DCE), indicating mat the coring  approach did not contribute
significant inconsistencies to the data. Schumacher and  others
(2004) found three- to five-fold differences in PCE concentra-
tions around some tree trunks. Sorek and others (2008) found
VOC concentration variations up to a factor of about 5 from
differing sides of the same tree at the same height.
     A variety of factors potentially can contribute to such
directional variations, including injuries (Scholander and
others, 1957), disease and insect damage (Kozlowski and

-------
20     User's Guide to the Collection and Analysis of Tree Coresto       the Distribution of Subsurface Volatile Organic Compounds
Paiiardy, 1997), gas embolisms (Clark and Gibbs, 1957), and
spiral transport up the trunk (Kozlowski and others, 1967;
Schumacher and others, 2004). In some situations, however,
directional differences in tree-core VOC concentrations are
caused by variations in subsurface VOC concentrations taken
up by root systems on differing sides of the tree (Vroblesky
and others, 1999). A study of loblolly pines (Pinus taeda)
showed that TCE concentrations in cores from a tree near
the edge of a ground-water contamination plume were 61 to
68 percent lower on the sides of the tree facing away from
the axis of contamination than on the sides of the tree facing
toward the axis of contamination (Vroblesky and others,
2004). The data indicate that in some cases, the direction
of the highest VOC concentration in a tree trunk may be
an indicator of the direction toward the greatest subsurface
VOC concentrations. Caution should be exercised with this
approach, however, because of the potential for spiral transport
up the trunk and other influences cited above (Kozlowski and
others, 1967; Schumacher and others, 2004).


Volatilization  Losses from the Tree Trunk

     VOC concentration decreases up the trunk of a tree have
been observed in some studies (Vroblesky and others, 1999;
Schumacher and others, 2004). These changes may be caused
by a variety of factors. In the first investigation using tree
coring as a tool to examine ground-water VOC concentrations,
a decrease in TCE concentration with height up a tree trunk
was  observed and hypothesized to be caused by volatilization
loss  through the bark (Vroblesky and others, 1999). Subse-
quently, laboratory and field investigations confirmed that
volatilization loss through the bark can be a major mechanism
for transfer of VOCs to the atmosphere (Davis and others,
1999; Burken, 2001; Ma and Burken, 2003), and upward
decreases in concentration also were observed in an additional
field investigation (Hirsh and others, 2003). One laboratory
experiment utilized diffusion traps to capture VOCs leaking
from stems and found that TCE leaked through the stems and
in all cases, the amount of TCE in the uppermost traps was
less than in the lowermost traps, indicating TCE loss up the
trunk (Ma and Burken, 2003). VOC loss also can occur from
the roots (Struckhoff, 2003). An additional effect of volatiliza-
tion  loss from the tree trunk is that VOC concentrations in the
trunk sometimes decrease from the inner to the outer part of
the trunk (Ma and Burken, 2003). Thus, in some cases, higher
VOC concentrations may be obtained by coring deep into the
tree rather than by collecting only the outermost rings.
     The rate of diffusive loss from tree trunks may be
influenced by the diameter of the trunk. Struckhoff (2003)
found that a 0.5-in.-diameter poplar cutting planted in con-
taminated soil or water has a 24-percent concentration loss in
5 in. of height, whereas a 6.5-in.-diameter Chinese elm (Ulmus
parvifolia) showed the same percent loss in 5 ft of height. The
author concluded that the greater surface area to volume ratio
in a smaller diameter section of tree will more quickly deplete
the reservoir of PCE in the trunk. In support of this conclusion,
Schumacher and others (2004) found that diffusion loss of
PCE in small (0.5-in. diameter) trees occurred at a rate more
than 10 times higher man in trees 6.5 in. in diameter.
     Decreases in the VOC content of tree cores with increas-
ing height up the trunk have not been observed at all sites
(Vroblesky and others, 2004; Sorek and others, 2008). At a
site in Texas, an upward increase in tree-core TCE concentra-
tions was observed following a heavy rain (Vroblesky
and others, 2004). The concentration increases up the tree
appeared to represent a time series of water movement, with
the lowest pail of the trunk representing TCE-contaminated
ground water diluted with infiltrating rainwater, and the upper-
parts of the tree representing pre-rain undiluted ground water.
Thus, concentration changes up the trunk may be caused by
a variety of factors. Consequently, a prudent approach when
conducting a site survey to examine area! distribution of VOCs
in tree cores is to collect all cores from the same height.


Rainfall  Infiltration as a Dilution Mechanism

     Plant utilization of ground water and uptake  of ground-
water contaminants depends partly on the reliability of
rainfall. In areas where rainfall  is unreliable, riparian trees may
develop roots primarily in the capillary fringe and phreatic
zone rather man throughout the soil profile (Ehleringer and
Dawson,  1992), thus primarily utilizing ground water. Fremont
cottonwood (Populus fretnontii) and Goodding's willow
(Salix gooddingii) growing along streams in western Arizona
used ground water throughout the growing season regardless
of the depth to water (Busch and others, 1992). Plants with
roots disseminated in multiple soil zones may use various
combinations of ground water, rainfall infiltrate, and stream
water, sometimes responding opportunistically to rainfall
events (Mensforth and others, 1994; Thorburn and Walker,
1994; Dawson and Pate, 1996; Jolly and Walker,  1996). Trees
near a perennial stream in California used shallow soil water
early in the growing season and then primarily used ground
water in the later part of the season when the soil  dried (Smith
and others, 1991). Mature box elder (Acer negundo) trees used
only ground water and did not seem to use perennial stream
water or shallow soil water in northern Utah (Dawson and
Ehleringer, 1991), but they did use soil water from precipita-
tion at ephemeral and perennial stream reaches in Arizona
(Kolb and others, 1997).
     The potential for source water to trees to be affected
by rainfall infiltration indicates that the VOC concentrations
detected in cores from trees growing above contaminated
ground water also may be affected by mixing with various
water sources. For example, Doucette and others  (2003)
reported that TCE concentrations in plant tissue at Hill Air
Force Base, Utah, where the climate is arid, were 10 to 100
tunes higher than at Cape Canaveral Air Station, Florida,
where the climate is humid and rainy, despite similar TCE
ground-water concentrations at both sites. The authors
hypothesized that the TCE concentration difference in tree
cores between the two sites was due to greater dilution effects

-------
                                                       Part 2.  Historical Perspectives and Technical Considerations     21
from rainfall at the Florida site than at the Utah site. It also is
possible, however, that the difference could be due to the dif-
ference in sample-collection methods and analytical variabil-
ity. Methanol extraction coupled with PT gas chromatography
was used to analyze the Florida plant tissue, whereas HAS gas
chromatography was used to analyze Utah plant tissue.
    To determine the influence on VOC concentrations in
tree cores of rainfall incorporation into transpiration stream,
a field test was conducted using artificial irrigation of a
mature cottonwood tree in Texas (Clinton and others, 2004;
Vroblesky and others, 2004). The test involved measuring
transpiration and tree-core TCE concentrations before and
after irrigating the tree. The results showed rapid TCE
concentration decreases and maximum transpiration value
increases following the artificial irrigation. These data indicate
that the uptake of irrigation water resulted in a rapid dilution
of TCE concentrations in the trunk. Thus, VOC concentrations
in tree cores collected after a rainfall may be less than before
the rainfall. A possible exception to this may occur if the
rainfall mobilizes contamination in the unsaturated zone.     j

                                                        I
Seasonal Influences

    Clear seasonal trends in VOC concentration of
tree cores has been observed in recent studies (Trapp
and others, 2007;  Sorek and others, 2008). Much
higher TCE concentrations in the trees  were recorded
in the dry hot season than in the wet cold season. Other
tree-coring investigations have shown concentration
changes possibly related to seasonal influences. The TCE
concentrations in bald cypress trees in South Carolina
decreased from July to September to January (Vroblesky
and others, 1999). Lewis (2001) reported that tree cores
from cottonwoods and Russian olive trees contained
significantly higher TCE in June 2000 and June 2001
than in the winter or even in May or late June to July of
the same years. Thus, it appears that in some cases there
are seasonal VOC concentration differences in trees with
higher concentrations during the summer than  the winter.
    Controlled mesocosm experiments have shown
that TCE flux to the atmosphere by transpiration  of bald
cypress seedlings is influenced diurnally and seasonally
(Nietch and others, 1999). The flux decreased from day
to night, probably because the stomata  are closed at
night (fig. 11). TCE flux also decreased from June to
December as transpirative water use seasonally decreased
(fig. 11). Interestingly, the study found that although the
winter TCE flux was reduced relative to the summer flux,
there was still  significant TCE flux in the winter, imply-
ing that trees do not need to be conducting substantial
amounts of water to remove TCE from the ground water.
    Although seasonal variations  in tree-core  VOC con-
centrations sometimes are present, the seasonal variations
do not appear to prevent the use of the tree coring as a tool
to investigate subsurface VOC concentrations during winter
months. The TCE concentrations in cores from a pine and
   an oak tree in Ashland, Massachusetts, were 17 ppbv in late
   August 2006 (Vroblesky and others, 2006), and approximately
   the same amount (14 and 18 ppbv, respectively) at the end of
   November 2006, after the leaves had fallen (Scott Clifford,
   U.S. Environmental Protection Agency, written commun.,
   2006). In addition, winter (February) tree cores provided
   the data that were successfully used in the first study to use
   tree cores as a tool to map subsurface VOCs (Vroblesky and
   others, 1999). Additional work is needed to fully understand
   the seasonal influences on tree-core sampling for VOCs.


   Sorption

        Organic compounds such as TCE have a tendency to sorb
   to plant tissue (Davis and others,  1998a). Newman and others
   (1997) found that a small percentage (3  to 4 percent) of TCE
   remained as an insoluble residue in poplar tree cells. Other
LU I-
CO Z>
55
OS
£<*
"- UJ
X CL
 i <
O^
on z
2-
o
on
       70
       60
       50
       40
       30
       20
       10
            A.
                                                    DAY
                                                    NIGHT
                  AUG
                            OCT
                                       NOV
                                                 DEC
                                MONTH
on
s
      0.08
      0.06
  ±:   0.04
 .
LJJ
I
tJ
O
on
2
o
on
      0.02
      0.00
            B.
                                                    Day
                                                    Night
                       SUMMER
                                         WINTER
Figure 11.   Trichloroethene removal from the rhizosphere of bald
cypress seedlings as (A) nanomoles per minute through the above-
ground part of the plant and (B) fractional trichloroethene loss from
carboy water during the summer and winter (modified from Nietch
and others, 1999).

-------
22     User's Guide to the Collection and Analysis of Tree Coresto       the Distribution of Subsurface Volatile Organic Compounds
investigations found that the predominant mass of VOCs in
tree trunks may reside in the wood tissue (Ma and Burken,
2002, 2003; Struckhoff, 2003; Schumacher and others,
2004) with a greater amount of sorption in more lignified
tissues, such as poplar trees, relative to young sunflower
plants (Helianthus annuus) (Davis and others, 1998b). Thus,
soiption of VOCs to the wood of the tree may be an important
influence on tree-core VOC concentrations (Schumacher and
others, 2004).
     Sorption of VOCs onto wood is related to the compound
Henry's Law constant and vapor pressure (Ma and Burken,
2002) and to the lipophilicity of the compounds, expressed
as log Kow (Trapp and others, 2001). In general, chlorinated
hydrocarbons tend to be partly excluded or sorbed within
plants, ethers are only slightly excluded or sorbed, and less
polar gasoline constituents are more strongly excluded or
sorbed man TCE (Burken,  1996; Makepeace and others, 1996;
Davis and others, 1998a). The soiption of TCE is considerably
stronger than that of TCA despite the greater log Kow of TCA
(Davis and others, 1998b).  Chloroform and dichloromethane
tend to be weakly sorbing compounds (Davis and others,
1998a).
     Approaches to quantifying sorption of VOCs onto wood
include examining the equilibrium distribution of VOCs
between tree tissue and water (KwoiJ. Higher KwooJ values
indicate higher potential for sorption. Mackay and Gschwend
(2000) measured soiption of benzene, toluene, and o-xylene to
wood and found that the Km>od coefficients were between 6.6
and 28 milligrams per gram (mg/g) of dry wood to milligrams
per milliliter of water, indicating a relatively high sorption
capacity of wood. They also found that the  soiption  is linear
and reversible.


Within-Tree Volatile Organic
Compound Degradation

     Phytodegradalion, or the breakdown of organic
contaminants within tree tissue, can take place inside the
plant or within the rhizospere of the plant (Newman and
Reynolds, 2004; Vroblesky and others, 2004). A study of
willow and poplar trees used as a polishing step for mitigating
a chlorinated solvent plume showed the presence in  tree
tissue of parent chlorinated ethenes and chloroacetic acids
(oxidative transformation products), indicating the uptake
and phytodegradalion of the contaminants (Nzengung, 2005).
Intermediate stable metabolites, apparently from the break-
down of chlorinated solvents, have been reported in  various
tree species growing above contaminated soils (Newman and
others, 1997; Complon and others, 1998; Doucetie and  others,
1998; Gordon and others, 1998). The presence of intermediate
degradation products and mass-balance considerations suggest
that MTBE degradation may take place in mature trees  (Rubin,
2007). Newman and others (1997) found that cells from
poplar trees are capable of transforming and mineralizing TCE
without the involvement of microbial metabolism. Vroblesky
and others (2004) found that the tree-core TCE and cDCE
distribution between the inner and outer parts of the trunk of a
narrow-leaf cottonwood (Populus angustifolia James) tree core
from Colorado was consistent with microbial dehalogenation
of TCE within the apparently methanogenic conditions of the
wetwood of the inner trunk. Wetwood conditions are caused
by an infestation in the tree of methanogenic and other bacte-
ria in the tree heartwood (Stankewich and others, 1971; Zeikus
and Ward, 1974). Methanogenic conditions are associated with
efficient dehalogenation of TCE to cDCE (Parsons and others,
1984, 1985; Kloepfer and others, 1985; Wilson and others,
1986).  The data from a variety of investigations indicate that
some level of VOC degradation can take place within the
tree or rhizosphere. Phytodegradation that takes place in the
rhizosphere, roots, or trunks of trees has the potential to reduce
or alter the VOC concentrations detected in tree cores.

    This manual provides a guide to the use of tree coring as
a tool to examine subsurface VOCs. It examines some of the
factors influencing the use of tree coring for that purpose and
summarizes some case studies in which tree coring has been
used to examine subsurface VOCs. Typical VOCs thai have
been detected in tree cores include benzene, toluene, ethylben-
zene, xylene isomers, trimethyl benzene, MTBE, TCE, PCE,
and cDCE. The method is inexpensive, portable, rapid, and
uncomplicated. The presence of VOCs in tree cores is a strong
indicator of subsurface VOC contamination; however, the
lack of VOC detection does not necessarily mean that VOC
contamination is not present.
    Tree cores are obtained by use of a clean and sharp incre-
ment borer. The boring should be started slowly and carefully
to avoid sideways slippage of the bit against the tree trunk. The
core is removed from the borer by means of an extractor. The
core should be transferred to a vial and sealed immediately
upon recovery from the tree. The hcadspacc in the vials should
be given enough time to equilibrate with the VOCs in the tree
core prior to analysis, typically about 24 hours for storage at
room temperature to about 5 to 15 minutes for field-healed
cores. Diffusion of TCE from the tree cores to the hcadspacc
in sample vials is fast enough, however, that analysis of
unhealed cores after 5 minutes can contain sufficient TCE to
indicate the presence of TCE contamination. Core analysis
can be performed using a variety of approaches, but IIS A by
gas chromatography often is the simplest approach. Quality
control and assurance samples should be collected to ensure
the integrity of the data.
    A variety of factors influence the ability of plants to
be useful indictors of ground-water VOC contamination.
These factors include the type of VOC, the tree species, the
rooting depth, aqueous concentrations, the depth to water,
concentration differences around the trunk related to different
sources of subsurface VOCs, concentration differences with

-------
                                                                                             References Cited    23
depth of coring related to volatilization loss through the bark
and possibly other unknown factors, dilution by rain, seasonal
influences, soiption, and within-tree VOC degradation.
Azevedo, J.I,., Maccheroni, Walter. Jr., Pereira, J.O., and de
  Araujo, W.L., 2000, Ednophytic microorganisms—A review
  on insect control and recent advances on tropical plants:
  Environmental Biotechnology, v. 3. no. 1, accessed
  January 12, 2006, ai http://www.ejbiote.chnology.info/
  content/vo13/issuel/fuU/4/index.hlml

Briggs, G.G., Bromilow, R.IL, and Evans, A.A., 1982, Rela-
  tionships between lipophicity and root uptake and transloca-
  lion of non-ionized chemicals by barley: Pesticide Science,
  v. 13, p. 495-504.

Briggs, G.G., Bromilow, R.H., Evans, A.A., and Williams, ML,
  1983, Relationships between lipophilicity and the distribu-
  tion of non-ionized chemicals in barley shoots following
  uptake by the roots: Pesticide Science, v. 14, p. 492-500.

Brown, K., Tyner, L., Grainger, 13., and Perina, T., 2001,
  Impact of natural attenuation and phyloremedialion on
  MTBE and fuel, in Magar, V., Gibbs, J.T., O'Reilly, K.T.,
  Ilyman, M.R., and Leeson, A., eds., Bioremediation of
  MTBE, alcohols, and ether, Sixth International In Situ and
  On-Site Bioremedialion Symposium, June 4-7, 2001, San
  Diego, California: Columbus, Ohio, Battellc Press.

Burken, J.G., 1996, Uptake and fate of organic contaminants
  by poplar trees: Iowa City, Iowa, University of Iowa, Ph. I).
  dissertation, 215 p.

Burken, J.G., 2001, Uptake and volatilization of chlorinated
  solvents by poplars at field scale, in Leeson, A., Foote,
  E.A., Banks, M.K., and Magar, V.S., eds., Phytoremedia-
  tion, wetlands and sediments: Columbus, Ohio, Batelle
  Press, p. 113-120.

Burken, J.G., and Schnoor, J.L., 1998, Predictive relation-
  ships for uptake of organic contaminants by hybrid pop-
  lar trees: Environmental Science and Technology, v. 32,
  p. 3379-3385.

Busch, D.E., Ingraham, N.L., and Smith, S.D.,  1992, Water
  uptake in woody riparian phreatophytes of the southwestern
  United States—A stable isotope study: Ecological Applica-
  tions, v. 2, p. 450-459.

Caldwell, M.M., Dawson, T.E, and Richards, J.H., 1998,
  Hydraulic lift—Consequences of water efflux for the roots
  of plants, Oecologia, v. 113, p. 151-161.
Castelfranco, P., Foy, C.L., and Deutsch, D.B., 1961, Non-
  enzymatic detoxification of 2-chloro-4,6-bis(ethylamino)-
  s-triazine (simazine) by extracts of Zea Mays: Weeds, v. 9,
  p. 580-591.

Chancy, W.R., 2000, Water and chemical movement beneath
  the bark: The Woodland Steward, Spring 2000, v. 9, no. 1,
  downloaded from http://www.fnr.purdue. edu/inwood/
  past%20issues/WATERANDCHEMICAL.htm (accessed on
  November 1,2007).

Clark, J., and Gibbs, R.D., 1957, Studies in tree physiology,
  IV, Further investigations of seasonal changes in moisture
  content of certain Canadian forest trees: Canadian Journal
  of Botany, v. 35, p. 219-253.

Clinton, B.D., Vose, J.M., Vroblesky, D.A., and Harvey, G.J.,
  2004, Determination of the relative uptake of ground vs.
  surface water by Populus deltoides during phytoremedia-
  tion: International Journal of Phytoremediation, v. 6, no. 3,
  p. 239-252.

Compton, II.R., Ilaroski, D.M., Ilirsh, S.R., and Wrobel, J.G.,
  1998, Pilot-scale use of trees to address VOC contamina-
  tion, in Wickramanayake, G.B., and Hinchee, R.E., eds.,
  Bioremediation and Phytoremediation, First International
  Conference on Remediation of Chlorinated and Recalci-
  trant Compounds, May 18-21, 1998, Monterey, California:
  Columbus, Ohio, Battelle Press, p. 245-249.

Cox, S.E., 2002, Preliminary assessment of using tree-tissue
  analysis and passive-diffusion  samplers to evaluate trichlo-
  roethene contamination of ground water at Site SS-34N,
  McChord Air Force Base, Washington, 2001: U.S. Geologi-
  cal Survey Water-Resources Investigations Report 02-4274,
  13 p.

Davis, L.C., Banks, M.K., Schwab, A.P., Muralidharan, N.,
  Erickson, L.E., and Tracy, J.C., 1996, Plant based biore-
  mediation, in Sikdar, S.K., and Irvine, R.L., eds., Bioreme-
  diation: Lancaster, Pennsylvania,  Technomics Publishing
  Company.

Davis, L.C., Erickson, L.E., Narayanan, M., and Zhang,
  Q., 2003, Modeling and design of phytoremediation, in
  McCulheon, S.C., and Schnoor, J.L., eds., Phytoremedia-
  tion—Transformation and control of contaminants: Hobo-
  ken, New Jersey, John Wiley and  Sons, Inc., p. 663-694.

Davis, L.C., Lupher, D., Hu, J., and  Erickson, L.E.,  1999,
  Transport of trichloroethylene through living plant tis-
  sues, in Proceedings of the 1999 Conference on Hazardous
  Waste Research, St. Louis, Missouri, May 24-27, 1999,
  p. 203-209.

-------
24     User's Guide to the Collection and Analysis of Tree Coresto       the Distribution of Subsurface Volatile Organic Compounds
Davis, L.C., Vanderhoof, S., Dana, J., Selk, K., Smith, K.,
  Goplen, B., andErickson, L.E., 1998b, Chlorinated solvent
  movement through plants monitored by Fourier transform
  infrared (FT-IR) spectroscopy: Journal of Hazardous Sub-
  stances Research, v. 1, no. 4, p. 1-26.

Davis, L.C., Vanderhoof, S., and Lupher, D., 1998a, Sorption
  of irichloroelhylene (TCE) and olher halogenaled aliphalics
  on plant materials,  in Ericksoti, L.E., and Rankin, M.M.,
  eds., Proceedings of the 1998 Conference on Hazardous
  Waste Research, Kansas State University, Manhattan, Kan-
  sas,?. 319-325.

Dawson, T.E., and Ehleringer, J.R., 1991, Streamside trees that
  do not use stream water: Nature, v. 350, p. 335-337.

Dawson, T.E., and Pate, J.S., 1996, Seasonal water uptake
  and movement in root systems of phreatophytic plants of
  dimorphic root morphology—A stable isotope investigation:
  Oecologia, v. 107, p. 13-20.

Dcttcnmaicr, E.M., Doucettc, W.J., Bugbcc, B., and Hall, A.,
  2008, Reexamining the relationship between plant uptake of
  organic chemicals and octanol water partition coefficients,
  in Platform Abstracts of the Sixth International Conference
  on Remediation of Chlorinated and Recalcitrant Com-
  pounds, May 19-22, 2008, Monterey, California: Columbus,
  Ohio, Battelle Press, session B6.

Dietz, A.C., and Schnoor, J.L., 2001, Advances in bioremedia-
  tion: Environmental Health Perspectives, v.  109, supple-
  ment 1, p. 163-168.

Doucette, W.J., Bugbee, B., Hayhurst, S., Plaehn, W., Downey,
  D., Taffinder, A., and Edwards, R., 1998, Phytoremediation
  of dissolved-phase trichloroethylene using mature vegeta-
  tion, in Wickramanayake, G.B., and Hinchee, R.E., eds.,
  Bioremediation and Phytoremediation, First International
  Conference on Remediation and Recalcitrant Compounds,
  May 18-21,1998, Monterey, California: Columbus, Ohio,
  Battelle Press, p. 251-256.

Doucette, W.J., Bugbcc, G.B., Smith, S.C., Pajak, C.J., and
  Ginn, J.S., 2003, Uptake, metabolism, and phytovolatiliza-
  tion of trichloroethylene by indigenous vegetation—Impact
  of precipitation, in  McCutcheon, S.C., and Schnoor, J.L.,
  eds., Phytoremediation—Transformation and control of
  contaminants: Iloboken, New Jersey, John Wiley and Sons.
  Inc., p. 561-588.

Douglass, A.E., 1919, Climatic cycles and tree growth: Carn-
  egie Institute of Washington Publication Number 289, v. 1,
  127 p.

Eckstein, D., and Dujesiefken, D., 1999, Long-term effects
  in trees due to increment borings.: Dendrochronologia,
  v. 16-17, p. 205-211.
Ehleringer, J.R., and Dawson, T.E., 1992, Water uptake by
  plants—Perspectives from stable isotope composition:
  Plant, Cell, and Environment, v. 15, p. 1073-1082.

Ehleringer, J.R., Philips, S.L., Schuster, W.F.S., and Sandquist,
  D.R., 1991, Differential utilization of summer rains by des-
  ert plants: Oecologia, v. 88, p. 430-434.

Ellmore, G.S., and Ewers, F.W., 1986, Fluid flow in the
  outermost xylem increment of a ring-porous tree, Ulmus
  atnericana: American Journal of Botany, v. 73, no. 12,
  p. 1771-1774.

Gopalakrishnan, Gayathri, Negri, M.C., Minsker, B.S., and
  Werth, C.J., 2005, Monitoring soil and ground water plumes
  using trees in an existing phytoremediation system: Pre-
  sented at the 3rd International Phytotechnologies Confer-
  ence, April 19-22, 2005, sponsored by The U.S. Environ-
  mental Protection Agency and the Midwest Hazardous
  Substance Research Center, Atlanta, Georgia.

Gopalakrishnan, Gayathri, Negri, M.C., Minsker, B.S., and
  Werth, C.J., 2007, Monitoring subsurface contamination
  using tree branches: Ground Water Monitoring and Reme-
  diation, v. 27, no.  1, p. 65-74.

Gordon, M., Choe, N., Duffy, J., Ekuan, G., Heilman, P.,
  Muiznieks, I., Ruszaj, M., Shurtleff, B., Strand, S., Wilmot,
  J., and Newman, L., 1998, Phytoremediation of trichloroeth-
  ylene with hybrid poplars: Environmental Health Perspec-
  tive, v. 106, no. 4, p. 1001-1004.

Grissino-Mayer, Henri, 2003, A manual and tutorial for the
  proper use of an increment borer: Tree-Ring Research,
  v. 59, no. 2, p. 63-79.

Hart, J.H., and Wargo, P.M., 1965, Increment borer wounds—
  penetration points for Ceralocyslis fagacearum:  Journal of
  Forestry, v. 63, no. 1, p. 38-39.

Hepting, G.H., Roth, E.R., and Sleeth, B., 1949, Discoloration
  and decay from increment borings: Journal of Forestry,
  v. 47, p. 366-370.

Heyerdahl, E.K., and McKay, S.J., 2001, Condition of live
  fire-scarred ponderosa pine trees six years after remov-
  ing partial cross sections: Tree-Ring Research, v. 57,
  p/131-139.

Hirsh, S.R., Compton, H.R., Matey, D.H., Wrobel, J.G., and
  Schneider, W.H., 2003, Five-year pilot study, Aberdeen
  Proving Ground, Maryland, in McCutheon, S.C., and
  Schnoor. J.L., eds., Phytoremediation—Transformation and
  control of contaminants: Hoboken, New Jersey, John Wiley
  and Sons, Inc., p. 635-659.

Hong, M.S., Rarmayan, W.F., Dortch, I.J., and Chiang, C.Y.,
  2001, Phytoremediation of MTBE from a groundwater
  plume: Environmental Science and Technology,  v. 35, no. 6,
  p. 1231-1239.

-------
                                                                                            References Cited     25
ICF Consulting, 2005, Indoor air human health risk assess-
  ment, Nyanza Chemical Waste Dump Superfund Site Oper-
  able Unit II, Ashland, Massachusetts: Consultant's report to
  the U.S. Army Corps of Engineers, New England District,
  October 25, 2005 [variously paged].

Inoue, J., Chamberlain, K., and Vromilow, R.H., 1998,
  Physiochemical factors affecting the uptake by roots and
  translocation to shoots of arnine bases in barley: Pesticide
  Science, v. 54, p. 8-21.

Jackson, R.B., Moore, L.A., Hoffmann, W.A., Pockman, W.T.,
  and Linder, C.R., 1999, Ecosystem rooting depth deter-
  mined with caves and DNA: Proceedings of the National
  Academy of Sciences of the United States of America,
  v. 96, p. 11387-11392.

Jolly,  I.D, and Walker, G.R., 1996, Is the field water use of
  Eucalyptus largiflorens F. Meull affected by short-term
  flooding?: Australian Journal of Ecology, v. 21, p. 173-183.

Keller, C., Hammer, D., Kayscr, A., Richneer, W., Brodbcck,
  M., and Sennhauser, M., 2003, Root development and heavy
  metal phytoextraction efficiency—Comparison of different
  plant species in the field: Plant Soil, v. 249, p. 67.

Kelso, P.W., 2005, An innovative field-based analytical
  method for low-level detection of chlorinated compounds
  in groundwater and soil samples  (The Color-Tec Method);
  accessed June 13, 2008, at http://www.clu-in.org/conf/tio/
  pasifcolortec___description.pdf

Kloepfer, R.D., Easley, B.B., Haas, B.B., Jr., Deihl, T.G.,
  Jackson, D.E., and  Wurrey, C.J.,  1985, Anaerobic degrada-
  tion of trichloroethylene in soil: Environmental Science and
  Technology, v. 19, no. 3, p. 277-280.

Kolb, T.E., Hart, S.C., and Amundson, R., 1997, Boxelder
  water source and physiology at perennial and ephemeral
  stream sites in Arizona: Tree Physiology, v. 17, p. 151-160.

Kozlowski, T.T., Hughes, J.F., and Leyton, L., 1966, Patterns
  of water movement in dormant gymnosperm seedlings:
  Biorheology, v. 3, p. 77-85.

Kozlowski, T.T., Hughes, J.F., and Leyton, L., 1967, Dye
  movement in gynmosperms in relation to tracheid align-
  ment: Forestry, v. 40, p. 209-227.

Kozlowski, T.T., and Pallardy,  S.G., 1997, Physiology of
  woody plants (2d ed.): San Diego, California, Academic
  Press, 411 p.

Landmeyer, J.E., Bradley, P.M., and Vroblesky, D.A., 2000,
  MTBE in trees transpiring gasoline-contaminated ground
  water, presented at  the Second International Conference on
  Remediation of Chlorinated and Recalcitrant Compounds,
  May 22-25, 2000, Monterey, California: Columbus, Ohio,
  Battelle Pi-ess.
Lewis, Kerry, 2001, The relationship between tree-core and
  groundwater trichloroethylene concentrations for ground-
  water plume delineation: Logan, Utah, Utah State Univer-
  sity, Department of Environmental Engineering, M.S. thesis,
  188 p.

Lewis, K., Doucette, W., Bugbee, B., and Ginn, J., 2001,
  Relationship between irichloroelhylene concentrations in
  tree cores and ground water for plume delineation, in Pro-
  ceedings of the Society of Toxicology and Environmental
  Chemistry (SETAC) 22"d Annual Meeting Abstract Book,
  November 11-15, 2001, Baltimore, Maryland.

Lockheed Martin, 2000, J-Field phytoremediation project—
  Field events and activities through July 31, 2000, Aberdeen
  Proving Ground, Edgewood, Maryland, August 31, 2000:
  Consultant's report to the U.S. Environmental Protection
  Agency, 28 p.

Lorenz, R.C., 1944, Discoloration and decay resulting from
  increment borings in hardwoods: Journal of Forestry, v. 42,
  p. 37-43.

Ma, X., and Burken, J.G., 2002, VOCs fate and partitioning
  in vegetation—Use of tree cores in groundwater analysis:
  Environmental Science and Technology, v. 36, no. 21,
  p. 4663-4668.

Ma, X., and Burken, J.G., 2003, TCE diffusion to the atmo-
  sphere in phytoremediation applications: Environmental
  Science and Technology, v. 37, p. 2534-2539.

Mackay, A. A., and Gschwend, P.M., 2000, Soiption of mono-
  aromatic hydrocarbons to wood: Environmental Science and
  Technology, v. 34, no. 5, p. 839-845.

Maeglin, R.R., 1979, Increment cores, how to collect, handle,
  and use them: Madison. Wisconsin, U.S. Department of
  Agriculture, Forest Service General Technical Report
  FPL 25,18 p.

Makepeace, V.D., Davis, L.C., Dana, J., Selk, K, Smith, K.,
  Hammaker, R.M., Fateley, W.G., and Erickson, L.E., 1996,
  Measuring contaminant flux through plants by fourier
  transform infrared (FT-IR) spectrometry, in Proceedings of
  the HSRC/WERC Joint Conference on the Environment,
  May 21-23,1996, Albuquerque, New Mexico: Manhattan
  Kansas, Kansas State University, p. 577-582.

Mensforth, L.J., Thorburn, P.J., Tyerman, S.D., and Walker,
  G.R., 1994, Sources of water used by riparian Eucalyptus
  catnaldulensis overlying highly saline ground water: Oeco-
  logia, v. 100, p. 21-28.

Melcalf & Eddy, Inc.. 2006. Technical memorandum for
  site-wide shallow groundwater investigation, final, Durham
  Meadows Superfund Site, Durham, Connecticut: Consul-
  tant's report to the U.S. Environmental Protection Agency,
  November 2006 [variously paged!.

-------
26     User's Guide to the Collection and Analysis of Tree Coresto       the Distribution of Subsurface Volatile Organic Compounds
Meyer, W.H., and Hayward S.B., 1936, Effect of incre-
  ment coring on Douglas-fir: Journal of Forestry, v. 34,
  p. 867-869.

Narayanan, M.N., Erickson, L.E., and Davis, L.C., 2004, Sim-
  ple plant-based design strategies for volatile organic pollut-
  ants: Environmental Progress, v. 18, no. 4, p. 231-242.

Negri, M.C., Gatliff, E.G., Quinn, J.J., and Hinchman,
  R.R., 2003, Root development and rooting at depths, in
  McCutheon, S.C., and Schnoor, J.L., eds., Phytoremedia-
  tion—Transformation and control of contaminants: Hobo-
  ken,  New Jersey, John Wiley and Sons, Inc., p. 233-262.

Newman, L.A., and Reynolds, C.M., 2004, Phytodegradation
  of organic compounds: Current Opinion in Biotechnology,
  v. 1,  no. 3, p. 225-230.

Newman, L.A.. Strand, S.E., Duffy. J., Choe, Nami,  Ekuan,
  G., Ruszaj, M., Shurtleff, B.B., Wilmoth, J., Heilman, P.,
  and Gordon, M.P., 1997, Uptake and biolransformalion of
  trichlorocthylene by hybrid poplars: Environmental Science
  and Technology, v. 31, no. 4, p. 1062-1067.

Nietch, C.T., Morris, J.T., and Vroblesky, D.A., 1999, Bio-
  physical mechanisms of trichloroethene uptake and loss
  in bald cypress growing in  shallow contaminated ground
  water: Environmental Science and Technology, v.  33,
  no. 17, p. 2899-2904.

Nilsen, E.T., and Orcutt, D.M., 1996, Physiology of  plants
  under stress: New York, New York, John Wiley and Sons,
  704 p.

Nzcngung, V.A., 2005, Case studies of phytoremcdiation
  of petrochemicals and chlorinated solvents in soil  and
  ground water, in Hatcher, K.J., ed., Proceedings of the 2005
  Georgia Water Resources Conference, April 25-27, 2005,
  Athens, Georgia.

Orchard, B.J., Doucette, W.J., Chard, J.K., and Bugbee,
  B., 2000, Uptake of trichloroethylene by hybrid poplar
  trees grown hydroponically in flow-through plant  cham-
  bers: Environmental Toxicology and Chemistry, v. 19,
  p. 895-903.

Parfitt,  D.G., Arnold, C., and Kaltreider, M, 2000, Root
  uptake and transpiration of gasoline oxygenates by Pinus
  radiata  [abst.]: Tenth Annual West Coast Conference on
  Contaminated Soils and Groundwater, March 20-23, 2000,
  San Diego, California.

Parsons, E, Lage, G.B., and Rice, R.. 1985. Biotransformation
  of chlorinated organic solvents in static microcosms: Envi-
  ronmental. Toxicology and Chemistry, v. 4, p. 739-742.

Parsons, E, Wood, P.R., and DeMarco, J., 1984, Transforma-
  tion of tetrachloroethene and trichloroethene in microcosms
  and ground water: Journal of the American Water Works
  Association, v. 72, no. 2, p. 56-59.
Ramaswami, A., and Rubin, A., 2001, Measuring phytoreme-
  diation parameters for volatile organic compounds—Focus
  on MTBE: Practice Periodical of Hazardous, Toxic, and
  Radioactive Waste Management, July 2001, p. 123-129.

Richards, J.IL, and Caldwell, M.M., 1987, Hydraulic lift-
  Substantial nocturnal water transport between soil layers by
  Arlemesia tridentate roots: Oecologia, v. 73, p. 486-489.

Roy F. Weston, Inc., 2000, Electronic map of Nyanza Chemi-
  cal Waste Dump Superfund Site, Ashland, Massachusetts,
  filename MAPPH.DWG, November 2005.

Rubin, E.G., 2007, Evidence for phytodegradation of MTBE
  from coupled bench-scale and intermediate-scale tests:
  Journal of Environmental Engineering, v. 133, no. 4,
  p. 389-396.

Schneider, W.IL, Ilirsh, S.R., Compton, II.R., Burgess,
  A.E., and Wrobel, J.G., 2002, Analysis of hydrologic
  data lo evaluate phyloremedialion system performance, in
  Gavaskar, A.R., and Chen, A.S.C., eds., Proceedings of the
  Conference on Remediation of Chlorinated and Recalci-
  trant Compounds—Third International Conference, March
  20-23, 2002, Monterey, California: Columbus, Ohio, Bat-
  tellc Press, 2B-26.

Schnoor, J.L., Licht, L.A., McCutcheon,  S.C., Wolf, N.L., and
  Carreira, L.H., 1995, Phytoremediation of organic and nutri-
  ent contaminants: Environmental Science and Technology,
  v. 29, no. 7, p. 318A-323A.

Scholander, P.F, Ruud, B., and Leivestad, H.,  1957, The rise of
  sap in a tropical liana: Plant Physiology, v.  32, p. 1-6.

Schumacher, J.G., Struckhoff, G.C., and Burken, J.G., 2004,
  Assessment of subsurface chlorinated solvent contamina-
  tion using tree cores at the Front Street Site and a former
  dry cleaning facility at the Riverfront Superfund Site, New
  Haven, Missouri, 1999-2003: U.S. Geological Survey Sci-
  entific Investigations Report 2004-5049, 35 p.

Sepaskhah, A.R., and Karimi-Goghari, Sh., 2005, Shallow
  ground-water contribution to pistachio water use: Agricul-
  tural water Management, v. 72, p. 69-80.

Shann, J.R., and Boyle, J.J., 1994, Influence of plant spe-
  cies on in situ rhizosphere degradation, in Anderson, T.A.,
  and Coats, J.R., eds., Bioremediation through rhizosphere
  technology: American Chemical Society, Washington DC,
  p. 71-81.

Sheehan, E.M., Burken, J.G., Karlson, II., Mayer, P., and
  Legind, C., 2007, Plant sampling for ground-water investi-
  gations fabsl.l: Fourth International Phylotechnologies Con-
  ference, September 24—26, 2007, Denver, Colorado, p. 43.

Shigo, A.L., 1967, Successions of organisms in discolor-
  ation and decay of wood: International Review of Forestry
  Research: v. 2, p. 237-239.

-------
                                                                                             References Cited    27
Shigo, A.L., 1974, A tree hurts too: Broomall, Pennsylvania,
  Forest Service, Northeastern Forest Experiment Station,
  Report NE-INF-16-73, 28 p.

Shigo, A.L., 1983, Tree defects—A photo guide: Broomall,
  Pennsylvania, Forest Service, Northeastern Forest Experi-
  ment Station, General Technical Rep NE-82, 167 p.

Shigo, A.L., 1985, Compaflmentalization of decay in trees:
  Scientific American, v. 252, no. 4, p. 96-103.

Shone, M.G.T., and Wood, A.V., 1972, Factors affecting
  absorption and translocation of simazine in barley: Journal
  of Exploratory Botany, v. 23, no. 74, p. 141-151.

Shone, M.G.T., and Wood, A.V., 1974, A comparison of the
  uptake and translocalion of some organic herbicides and
  a systemic fungicide by barley: Journal of Experimental
  Botany, v. 25, p. 390-400.

Smith, S.D., Wellington, A.B., Nachlinger, J.A., and Fox,
  C.A.,  1991, Functional responses of riparian vegetation to
  streamflow diversions in the eastern Sierra Nevada: Eco-
  logical Applications, v. 1, p. 89-97.

Sorek, A., Atzmon, N., Dalian, O., Gerstl, Z., Kushisin, L.,
  Laor,  Y., Mingelgrin, U., Nasser, A., Ronen, D., Tsechan-
  sky, L.,  Weisbrod, N., and Graber, E.R., 2008, Phytoscreen-
  ing—The use of trees for discovering subsurface contamina-
  tion by VOCs: Environmental Science and Technology, v.
  42, p. 536-542.

Stankewich, J.P., Cosenza, G.J., and Shigo, A.L., 1971,
  Closlridum quercicolum sp. N., isolated from discolored tis-
  sues in living oak trees: Antonie van Lecuwenhock Journal
  of Microbiology and Serology. v. 37, p. 299-302.

Struckhoff, G.C., 2003, Uptake of PCE vapor by plant-
  Impacts for phytore-mediation: Rolla, Missouri, University
  of Missouri, Department of Environmental Engineering,
  unpublished M.S. thesis, 57 p.

Struckhoff, G.C., Burken, J., and Schumacher, J., 2005a,
  Vadose zone PCE—An important source and fate in planted
  systems: Presented at the 3rd International Phytotechnolo-
  gies Conference, April 19-22, 2005, sponsored by The  U.S.
  Environmental Protection Agency and the Midwest Hazard-
  ous Substance Research Center, Atlanta, Georgia.

Struckhoff, G.C., Burken, J., and Schumacher, J., 2005b,
  Vapor-phase- exchange of perchloroelhene between soil
  and plants: Environmental Science and Technology, v. 39,
  p. 1563-1568.

Thorburn, P.J., and Walker, G.R., 1994, Variations in stream
  water uptake- by Eucalyptus camaldulensis with differ-
  ing access to stream water: Oecologia, v. 100, no. 3,
  p. 293-301.
Took, E.R., and Gammage, J.L., 1959, Damage from incre-
  ment borings in bottomland hardwoods: Journal of Forestry,
  v. 57, no. 12, p. 909-911.

Tout, R.E., Gilboy, W.B., and Spyrou, N.M., 1977, Neutron
  activation studies of trace elements in tree rings: Journal of
  Radioanalytical Chemistry, v. 37, p. 715.

Trapp,  Stefan, Larse-n, Morten, Legind, C.N., Burken, Joel,
  Machackova, Jinna, and Karlson, U.G., 2007, A guide- to
  vegetation sampling for screening of subsurface pollution;
  accessed April 9, 2008, at vww2.er.dtu.dk/Homepages/stt/
  GuidetoVegetationSanipIing.pdf

Trapp,  Stefan, Miglioranza, K.S.B., andMosbak, Hans, 2001,
  Sorption of lipophilic organic compounds to wood and
  implications for their environmental fate: Environmental
  Science and Technology, v. 35, no. 8, p. 1561-1566.

U.S. Environmental Protection Agency, 2003, Method 5021 A,
  Volatile organic compounds in various sample matrices
  using equilibrium headspace analysis; accessed Decem-
  ber 2, 2005, at http://Hrww.epa.gov/epaoswer/hazwaste/test/
  pdfs/502 la _rl.pdf

van Mantgem, P.J., and Stephenson, N.L., 2004, Does coring
  contribute to tree mortality?: Canadian Journal of Forestry
  Research, v. 34, p. 2394-2398.

Vroblesky, D.A., 1998, Chlorinated e-thene uptake- and loss in
  tree- trunks: U.S. Geological Survey Department of Defense
  Environmental Conservation (DODEC) meeting, May 4-8,
  1998, Tampa, Florida.

Vroblesky, D.A., and Casey, C.C., 2004, Tree coring as an
  optimization tool for monitoring-well placement, Fourth
  International Conference-on Remediation of Chlorinated
  and Recalcitrant Compounds, May 24-27, 2004, Monterey,
  California: Columbus, Ohio, Battellc Press.

Vroblesky, D.A., Casey, C.C., Harvey, G., Morris, J.T., and
  Vose, J.M., 2001, TCE and DCE in tree trunks in diverse
  environments, Sixth International In Situ and On-Site Blore-
  mediation Symposium, June 5, 2001, San Die-go, California:
  Columbus, Ohio, Battelle Press.

Vroblesky, D.A., Casey, C.C., Petkewich, M.D., Lowery,
  M.A., Conlon, K.J., and Harrelson, E.G., 2007a, Inves-
  tigation of ground-water contamination at Solid Waste
  Management Unit 12, Naval Weapons Station, Charleston,
  North Charleston, South Carolina: U.S. Geological Survey
  Scientific Investigations Report 2006-5311, 81 p.

Vroblesky, D.A., Clinton, B.D., Vose, J.M., Casey, C.C., Har-
  vey,  G.J., and Bradley, P.M., 2004, Groundwaler chlorinated
  ethcnes in tree trunks—Case studies, influence of recharge,
  and potential degradation mechanism: Ground Water Moni-
  toring and Remediation, v. 24, no. 3, p. 124-138.

-------
28     User's Guide to the Collection and Analysis of Tree Coresto       the Distribution of Subsurface Volatile Organic Compounds
Vroblesky, D.A., Nietch, C.T., and Morris, J.T., 1999, Chlo-
  rinated ethenes from ground water in tree trunks: Environ-
  mental Science and Technology, v. 33, no. 3, p. 510-515.

Vroblesky, D. A., Willey, R.E., Clifford, Scott, and Jenkins,
  David, 2007b, Real-time mapping of ground-water VOC
  contamination with tree cores: Fourth International Phyto-
  lechnologies Conference, September 24-28, 2007, Denver,
  Colorado.

Vroblesky, D.A., Willey, R., Clifford, S., and Murphy,.!.,
  2006, Data from tree-coring investigation near the Nyanza
  Chemical Waste Dump Superfund Site, Ashland, Mas-
  sachusetts, August 30, 2006: U.S. Geological Survey Data
  Series 218, 5 p.

Vroblesky, D.A., Willey, R.E., Clifford, Scott, and Murphy,
  J.J., 2008, Real-time and delayed analysis of tree and shrub
  cores as indicators of subsurface volatile organic compound
  contamination, Durham Meadows Superfund Site, Durham,
  Connecticut, August 29, 2006: U.S. Geological Survey
  Scientific Investigations Report 2007-5212, 12 p.

Vroblesky, D.A., and Yanosky, T.M., 1990, Use of tree-ring
  chemistry to document historical ground-water contamina-
  tion events: Ground Water, v. 28, no. 5, p. 677-684.

Vroblesky, D.A., Yanosky, T.M., and Siegel, F.R., 1992,
  Increased concentrations of potassium in heartwood of trees
  in response to  ground-water contamination: Environmental
  Geology and Water Science, v. 19, no. 2, p.  71-74.

Weishaar,  L, and Burken, J.G., 2005, Phyto-mapping to delin-
  eate a chlorinated VOC plume at Aberdeen  Proving Ground,
  Maryland: Presented at the 3rd International Phytotechnolo-
  gies Conference, April 19-22, 2005, sponsored by The U.S.
  Environmental Protection Agency and the Midwest Hazard-
  ous Substance Research Center, Atlanta, Georgia.
Wilson, B.H, Smith, G.B, and Rees, J.F., 1986, Biotransfor-
  mations of selected alkylbenzenes and halogenated aliphatic
  hydrocarbons in methanogenic aquifer material—A micro-
  cosm study: Environmental Science and Technology, v. 20,
  no. 10, p. 997-1002.

Wise, D.P., 1997, Modeling the uptake and transpiration of
  TCE using phrealophy tic trees: Wright-Patterson Air Force
  Base, Ohio, Air Force Institute of Technology, M.S. thesis,
  31.8 p.

Wullschleger, S., Meinzer, E, and Vertessy, R.A., 1998, A
  review of whole-plant water use studies in trees: Tree Physi-
  ology, v. 18, p. 499-512.

Yanosky, T.M., Hansen, B.P., and Schening, M.R., 2001, Use
  of tree rings to investigate the onset of contamination of a
  shallow aquifer by chlorinated hydrocarbons: Journal of
  Contaminant Hydrology, v. 50, p. 159-173.

Yanosky, T.M., and Vroblesky, D.A., 1992, Relation of nickel
  concentrations in tree rings to ground-water contamination:
  Water Resources Research, v. 28, no. 8, p. 2077-2083.

Yanosky, T.M., and Vroblesky, D.A., 1995, Element analy-
  sis of tree rings in ground-water contamination studies,
  in Lewis, T.E., ed., Tree rings as indicators of ecosystem
  health: Boca Raton, Florida^ CRC Press, Inc., p. 177-208.

Zeikus, J.G., and Ward, J.C., 1974, Methane formation
  in living trees—A microbial origin: Science, v. 184,
  p. 1181-1183.

Zencich, S.J., Froend, R.H., Turner,  J.V., and Gaililis, Vit,
  2002, Influence of ground-water depth on the seasonal
  sources of water accessed by Banksia tree species on a shal-
  low, sandy coastal aquifer: Oecologia, v. 131, no. 1, p. 8-10.

-------
                                                                                    Appendixl. Case Studies     29
Appendix 1.  Case Studies
    This section contains four case studies investigating volatile organic compounds in ground water using tree cores. The case
studies are of varying length and detail because of differences among the copyrights associated with the original publication
sources. Case Study 1 was the first investigation demonstrating that tree-core analysis could be used to delineate shallow
ground-water contamination by chlorinated ethenes. Case Studies 2 and 3 are previously unpublished investigations. Case
Study 2 examines a site where data from tree coring provided a reconnaissance-level understanding of the plume distribution and
allowed optimization of well placement. In Case Study 3, the search for a public-supply-well contaminant source encompassed
much of the city of New Haven, Missouri, but tree coring allowed the investigation to be narrowed to a 1-acre area. Although
much of the work involving tree coring has been directed toward chlorinated solvents, Case Study 4 shows that tree coring also
can be used to detect subsurface petroleum hydrocarbons and methyl tert-butyl ether (MTBE).
Case Study 1:  Chlorinated Ethenes from Ground Water  in Tree Trunks
Summarized from: Vroblesky, DA,1 Nietch, C.T.,2 and Morris, J.T.,21999, Chlorinated ethenes from ground
water in tree trunks: Environmental Science and Technology, v. 33, no. 3, p. 510-515.
    This was the first investigation showing that tree-core
analysis could be used to delineate shallow ground-water
contamination by chlorinated ethenes. Headspace analysis of
cores from 97 trees (6 species, predominantly bald cypress)
growing over ground-water contamination in a forested flood
plain of the Savannah River near the TNX Area, Savannah
River Site, South Carolina (figs. 1.1 and 1.2), showed that
cw-l,2-dichloroethene (cDCE) (fig. 1.2) and trichloroethene
(TCE) (fig.  1.3) concentrations in tree cores reflected the
configuration of the ground-water contamination plume. The
distribution of bald cypress containing TCE was more wide-
spread than the distribution of bald cypress containing cDCE
and was found in trees farther south than the flow path from
the source area at the former seepage basin, indicating the
presence of a second plume of TCE in the aquifer (fig. 1.3).
    Concentration variations around the tree trunks and a
TCE concentration decline of 30 to 70 percent with increasing
tree height up to 56 ft were observed. All tested tree species
were capable of taking up TCE. Some tree species, such as
tupelo and bald cypress, appeared to exhibit similar TCE or
cDCE uptake potential. Oaks, however, appeared to contain
less TCE than adjacent bald cypress or loblolly pines. Sweet-
gum also appeared to contain less TCE than loblolly pines.
Reference

Vroblesky, D.A., Nietch, C.T., and Morris, J.T., 1999, Chlo-
  rinated ethenes from ground water in tree trunks: Environ-
  mental Science and Technology, v. 33, no. 3, p. 510-515.
  'U.S. Geological Survey, Columbia, South Carolina.

  2University of South Carolina, Columbia, South Carolina.
            Figure 1.1.  Location of study area
            [reprinted with permission from
            Vroblesky and others (1999), copyright
            (1999) American Chemical Society].

-------
30     User's Guide to the Collection and Analysis of Tree Coresto Assess the Distribution of Subsurface Volatile Organic Compounds
                                                                                                TNX Area
           o
         " 10  —
         TCM3
          227
           V

          134
           *
               EXPLANATION

BALDCYPRESS TREE SAMPLED FOR THIS INVESTIGATION

LINE OF EQUAL CIS 1,2-DICHLOROETHENE CONCEN-
  TRATION IN TREE CORES, IN NANOMOLES OF GAS
  PER LITER OF WATER IN CORE-Dashed where inferred.
  Countour interval variable as shown.

OBSERVATION WELL AND IDENTIFIER AND CIS
  1,2-DICHLOROETHENE CONCENTRATION IN GROUND
  WATER, IN NANOMOLES PER LITER.

DITCH-BED PORE-WATER SAMPLING POINT AND CIS
  1,2-DICHLOROETHENE PORE-WATER
   CONCENTRATION, IN NANOMOLES PER LITER
V
TNX26D
not done
                                                                                                  •v
0
0
100
200
300 FEET
SO METERS
     Figure 1.2.   c/s-1,2-Dichloroethene concentrations in bald cypress trunks in January and February 1998 and in ground water
     during August 1997 and ground-water flow, TNX flood plain. Savannah River Site, SC [modified and reprinted with permission
     from Vroblesky and others (1999), copyright (1999) American Chemical Society].

-------
                                                                                   Appendixl. Case Studies    31
                                                                                Former
                                                                                             TNX Area
       o
    — 10
     TNX15D
       180
       V
      822
               EXPLANATION

BALDCYPRESS TREE SAMPLED FOR THIS INVESTIGATION

LINE OF EQUALTRICHLOROETHENE CONCENTRATION
  IN TREE CORES, IN NANOMOLES OF GAS PER LITER
  OF WATER IN CORE-Dashed where inferred. Countour
  interval variable as shown.

OBSERVATION WELL AND IDENTIFIER AND
  TRICHLOROETHENE CONCENTRATION IN GROUND
  WATER, IN  NANOMOLES PER LITER.

DITCH-BED PORE-WATER SAMPLING POINT AND
  TRICHLOROETHENE PORE-WATER CONCENTRATION,
   IN NANOMOLES PER LITER.
  TNX26D\
41-84 in 1996
                                                                                            \
                                                                                               \
0
0
100

200 300 FEET
50 METERS
Figure 1.3.  Trichloroethene concentrations in bald cypress trunks in January and February 1998 and in ground water during
August 1997 and ground-water flow, TNX flood plain. Savannah River Site, SC [modified and reprinted with permission from
Vroblesky and others (1999), copyright (1999) American Chemical Society].

-------
32    User's Guide to the Collection and Analysis of Tree Coresto Assess the Distribution of Subsurface Volatile Organic Compounds

Case Study 2: Tree Coring as a Guide to Well Placement, Solid Waste Management Unit 17,
Naval Weapons Station Charleston, South Carolina, 2002
By Don A. Vroblesky1 and Clifton C. Casey2
Abstract

    Several tree cores were collected from Solid Waste Man-
agement Unit (SWMU) 17, Naval Weapons Station Charles-
ton, South Carolina, as part of an effort to characterize the
site and to direct monitoring-well placement. Analysis of the
tree cores showed the presence of two distinct ground-water
plumes. One of the plumes was composed predominantly of
tetrachloroethene and the other was composed predominantly
of trichloroethene. The data allowed optimization of well
placement, and the  subsequent well-sample analysis confirmed
the tree-core data.


Introduction

    Solid Waste Management Unit (SWMU) 17 at the
Naval Weapons Station Charleston, South Carolina, is a
flat-lying forested area in which the dominant tree species
is loblolly pine. Well sampling at the site in 2001 showed
61-190 micrograms per liter (\ig/L)
of tetrachloroethene (PCE) and
very low concentrations (less than
9 (ig/L) of trichloroethene (TCE)
in the shallow ground water (Tetra
Tech NUS, Inc., 2004). Sampling
of temporary wells  in 2002,
however, showed that other parts of
the site contained 31,000 \ig/L of
TCE (Tetra Tech NUS, Inc., 2004)
(fig. 2.1). The data indicated a need
for additional monitoring wells to
map the ground-water contamina-
tion. Initial investigations by the
contractor, however, indicated that
tidal changes caused substantial
variations in ground-water-flow
direction, complicating the plan-
ning of future monitoring-well
placement (Tetra Tech NUS, Inc.,
2004). Naval Facilities Engineering
Command Southeast requested that
                        the U.S. Geological Survey (USGS) conduct a tree-core survey
                        as a reconnaissance tool to direct well placement. As part of
                        the investigation, the USGS and Naval Facilities Engineering
                        Command Southeast cored and analyzed 61 tree cores from
                        the site in September 2002 (fig. 2.2). The cored trees consisted
                        of 28 loblolly pines, 12 Chinese tallow (Sapium sebiferum L.),
                        6 sweet gum, 1 oak, and 12 trees of unknown species.
                             Tree cores were collected by use of an increment borer.
                        The cores were sealed in crimp-cap 20-milliliter serum vials
                        immediately upon recovery and analyzed by headspace
                        analysis gas chromatography the day after sample collection.
                        Comparison of duplicate cores showed a TCE concentration
                        difference of less than 5 percent in a tree containing greater
                        than 8,000 parts per billion by volume (ppbv) of TCE. Most
                        of the tree cores were collected during a single day (Septem-
                        ber 11, 2002) and analyzed the following day. The remaining
                        tree cores were collected 2 days later to expand on results
                        from the first survey.
  'U.S. Geological Survey, Columbia,
South Carolina.

  2Naval Facilities Engineering Com-
mand Southeast, North Charleston, South
Carolina.
O 1OO 2OO FEET
i l
O 25 5O METERS
/ "-• ^
/ SWMU17 -.^ WelM7MW03
Trenches ' \ TCE<5 |jg/L
| Well 17MW03S \ PCE<5 |jg/L
i TCE<9 |jg/L /

EXPLANATION
® Permanent well
sampled in 2001
x-x Temporary well
<> sampled in 2002
~ ~ "i Boundary of
i 	 | SWMU17
ug/L is micrograms
per liter. "<" indicates
less than. "J"
indicates estimated
value

PCE=190ug/L /
; ®
| Well 17TW02S /
' TCE<5 ug/L / ®
/ PCE<5 ug/L ' Well 17MW02S
/ TCE=2ug/L(J)
/ Well17TW01S / PCE=61U9/L
' TCE=31, 000 ug/L /
,' PCE<5 ug/L
n ''
\ Basinl I \
1 ^ \ Well 17MW06S
\ \ TCE<5 ug/L
^ > PCE<5ug/L N
"- — 1 vJ uc
^ 	 <8> ^
Figure 2.1.   Locations of monitoring wells and concentrations of ground-water trichloroethene
(TCE) and tetrachloroethene (PCE), Solid Waste Management Unit (SWMU) 17, Naval Weapons
Station Charleston, South Carolina.

-------
      A. Trichloroethene (TCE)
                   50 METERS
                                    SWMU17
                                      A
  o
 A
 V
 A
ppb/v i
            EXPLANATION
          Countour showing
          ground-water TCE
          concentration, in micro-
          grams per liter (data from
          Tetra Tech NUS, Inc., 2004).
          Dashed where inferred.
          Interval variable as shown.
          Monitoring well
          Temporary well, 2002
          Temporary well, April 2003
          Tree: >500 ppb/v TCE
          Tree: 10-100 ppb/v TCE
          Tree: TCE not detectable
         is parts per billion by volume
                                                                                          Appendixl.  Case Studies    33
                                                          B. Tetrachloroethene (PCE)
                                                                               200 FEET
                                                                             50 METERS
                                                                                        SWMU17
                                                                                           A
      EXPLANATION
 -20- Countour showing
    ground-water PCE
    concentration, in micro-
    grams per liter (data from
    Tetra Tech NUS, Inc., 2004).
    Dashed where inferred.
    Interval variable as shown.
 ®  Monitoring well
 ®  Temporary well, 2002
 o  Temporary well, April 2003
 A Tree: 10-47 ppb/v PCE
 A Tree: PCE not detectable
ppb/v is parts per billion by volume
    Figure 2.2.   (A) Trichloroethene (TCE) and (B) tetrachloroethene (PCE) concentrations in tree cores in 2002 and ground water
    in 2002-2003 at Solid Waste Management Unit (SWMU) 17, Naval Weapons Station Charleston, South Carolina.
     Analysis of the tree cores showed the presence of two
distinct areas of ground-water contamination. In the southern
part of SWMU17, trees contained TCE with no detectable
PCE concentrations. The highest TCE concentrations found
in the tree cores (860 to 85,160 ppbv) were from trees near
a shallow drainage basin (fig. 2.2A). In the northern part of
S WMU17, trees contained 10 to 47 ppbv of PCE, but no
detectable TCE (fig. 2.2). Trees between the north and south
sides of SWMU17 contained no detectable TCE or PCE.
     Using the  tree-coring results as a placement guide,  the
consultant installed and sampled 21 temporary wells in April
2003 (fig. 2.2). Analysis of ground water from the temporary
wells, combined with historical data from earlier wells,
confirmed the presence of the two distinct contamination
plumes identified by the tree coring. As indicated by the tree
coring, the southern plume consisted primarily of TCE,  with
the highest concentrations near the shallow drainage basin
(9,000-95,000  (ig/L) (Tetra Tech NUS, Inc., 2004). PCE
concentrations  were less than 6 \ig/L. In the northern part of
SWMU17 near the sampled trees, PCE concentrations in the
ground water were 29 \ig/L in the 2003 temporary well and 61
to 190 (ig/L in previously tested wells, with TCE concentra-
tions less than 10 |ig/L (Tetra Tech NUS, Inc., 2004).
     The distribution of volatile organic compounds (VOCs)
in trees at the site strongly indicated that the shallow basin
in the southern part of SWMU17 (fig. 2.1) was a source area
for TCE ground-water  contamination. Subsequent drilling
and sampling of temporary wells confirmed that finding
(fig. 2.2A). The lack of VOC detections in trees in the central
part of SWMU17 strongly implied that the PCE  contamination
in the northern  part of SWMU17 was unrelated to the TCE
plume near the shallow basin. Again, subsequent drilling
and sampling of temporary wells confirmed that finding and
                                                        provided additional evidence indicating that the northern PCE
                                                        plume appeared to be coming from the direction of some
                                                        trenches across the road northwest of S WMU17 (fig. 2.2B).
                                                        Thus, tree coring provided a simple, rapid, inexpensive
                                                        reconnaissance tool and guide to optimize monitoring-well
                                                        placement.


                                                        Reference Cited

                                                        Tetra Tech NUS, Inc., 2004, RCRA Facility Investigation
                                                          documentation and data summary 2000-2003, for Old
                                                          Southside Landfill—SWMU 16 and Old Southside Missile
                                                          and Waste Oil Disposal Area—SWMU 17, Naval Weapons
                                                          Station Charleston, Charleston, South Carolina:  Consul-
                                                          tant's report submitted to Southern Division Naval Facilities
                                                          Engineering Command, January 2004 [variously paged].

-------
34     User's Guide to the Collection and Analysis of Tree Coresto       the Distribution of Subsurface Volatile Organic Compounds

                           Unit 4,
By John Schumacher1
Abstract

    Tree coring for volatile organic compounds was used
at a site in New Haven, Missouri, as a reconnaissance tool
to locate the source area for tetrachloroethene-contaminated
ground water that caused the abandonment of two public-
supply wells. The ability to rapidly and inexpensively collect
tree cores from a broad area throughout the town allowed the
tree coring to eliminate some suspected source areas, such
as transport through sanitary sewer lines, and eventually to
narrow the search down to a 1-acre area. Subsequent drilling
confirmed that the site identified by tree coring was a previ-
ously  undetected, shallow source of tetrachloroethene and was
a likely source of the tetrachloroethene-contaminated ground
water.


Introduction

    Ground-water contamination by tetrachloroethene (PCE)
detected in two 800-feet (ft)-deep public-supply wells (wells
Wl and W2) in 1986 in New Haven, Missouri, resulted in
closure of those wells. The  contaminated area is referred to as
the Riverfront site, and consists of six operable units (OUs).
The tree cores indicated that the probable contaminant source
was OU4.  Unlike the other  OUs, however,  there was no known
PCE use or disposal at OU4, and the area was designated as
an OU primarily because it was upgradient from contaminated
city wells Wl and W2. The investigation was conducted in a
scries of iterations,  beginning at the known PCE  contamina-
tion in city wells Wl and W2 and moving upgradient to the
south.


        I,      Tree-Core
Sampling

    In 2001, a reconnaissance sampling of water from
streams (grab samples) and cores from 86 trees in OU4 and
the surrounding area identified 2 stream reaches contaminated
with PCE and 9 trees in the central part of OU4 containing
PCE concentrations ranging from 0.58 to 117 micrograms in
headspace per kilogram (jig-h/kg) of wet core (fig. 3.1). The
largest PCE concentrations detected in trees were in trees
JS106 (117 |ig-/h/kg), a 30-inch (in.)-diameler hedge apple
(Madura pomiferd) tree growing along an old fence row,  and
JS112 (99.1  iig-h/kg), a 36-in.-diameter cottonwood tree about
  HJ.S. Geological Survey, Rolla, Missouri.
80 ft southwest of tree JS106. A large number of trees upslope
from the contaminated stream reach in the east-central part of
OU4 were cored based on a rumor that drums of waste may
have been buried in that area; however, none of the trees cored
along this stream contained detectable concentrations of PCE
and a surface geophysical survey detected no buried metallic
objects.
    In 2001, trace concentrations of PCE (0.24 to 1.3 micro-
grams per liter (|ig/L) in sanitary sewer lines mat drained
suspected source areas south of OU4 (U.S. Environmental
Protection Agency, 2003a) raised concerns that OU2 was
the source area and thai the contamination was  transmitted
through the sanitary sewer and streams. Tree-core evidence,
however, did not support this hypothesis. Although PCE was
detected at concentrations between 5.0 and 14 iig-h/kg in three
trees (JS100, JS104, and JS114) in proximity to the sanitary-
sewer main crossing OU4, most trees cored along the sewer
main had no detectable PCE concentrations. More importantly,
trees JS106 and JS112, which had the highest PCE concentra-
tions, were several hundred feet from the sanitary sewer main.
The 2001 stream and tree-core reconnaissance sampling had
indicated that the PCE source in OU4 was local and shallow.
By 2003, the source for the PCE plume in the bedrock aquifer
was thought to reside within an 80-acre "suspect area" in the
central pal of OU4 (fig. 3.1) In mid-2003, PCE concentrations
as large as 2,300 \igfL were detected at 138 ft deep in monitor-
ing well cluster BW-10 that was installed downgradient (north)
from this suspect area (fig. 3.1).


        2,      Tree-Core

    To refine the  size of the 80-acre suspect area, during
the fall of 2003, an additional 62 trees were cored within
the suspect area, including several trees around the former
corporate guest house (fig. 3.1). PCE was detected above the
0.5 ug-h/kg detection threshold used for this study in only 5
of the 62 trees  cored. Although 2 of the 16 trees cored at the
former corporate guest house contained low (0.51 to
3.9 |ig-h/kg) PCE  concentrations, the largest PCE concentra-
tions detected (21.2 to 100 |ig/kg) were in 3 trees more than a
block south of the  former corporate guest house. Two of these
trees [JS324 (PCE of 100 ug-h/kg) and tree JS340 (PCE of
74.9 (xg-h/kg)]  were within 90 ft of two trees  that previously
showed PCE contamination (trees JS106 and JS112) (fig. 3.2).
Tree JS324 was a small (1.5-in. diameter) mulberry tree
growing along  the same old fence row as tree JS106, and tree
JS340 was a 14-in.-diameter ash tree growing about 4 ft east
of tree JS112. Based on the small size of tree JS324, the high

-------
                                                                             Appendixl.  Case Studies    35
 38° 36'30" —
    38° 36' —
                                  BW-0
                         Former
                         corporate
                         guest house
               AREA1
               Inset shown
               in figure 3.2
                                                                        1,000
                        2,000 FEET
                                                 EXPLANATION
                   STREAM REACH CONTAINING MORE THAN 5.0
                   MICROGRAMS PER LITER PCE
                   SANITARY SEWER LINE

                   PCE CONCENTRATION IN TREE-CORE
                   SAMPLE - Concentrations in micrograms
                   in headspace per kilogram of wet core (|ig-h/kg).
                   Circle indicates 2001 sample,
                   square indicates 2003 sample
                  •  Less than 0.5
                  •  0.5 to 14.9
                  O  15.0 to 49.9
                  O  50.0 to 99.9
                  •  100 to 120
      MONITORING WELL CLUSTER
      AND NUMBER - Color indicates
      maximum PCE concentration
      detected in micrograms
      per liter (|ig/L).

BW-°7A  Less than 0.2
      A  0.2 to 4.9
      A  5.0 to 49
      A  50.0 to  999
      A  1,000 to 9,200

      PUBLIC SUPPLY WELL
Figure 3.1.  Location of tree-core samples and monitoring wells andtetrachloroethene (PCE) concentrations in Operable Unit 4,
Riverfront Superfund Site, Franklin County, Missouri, as of 2003 (modified from U.S. Environmental Protection Agency, 2003a).

-------
36     User's Guide to the Collection and Analysis of Tree Coresto Assess the Distribution of Subsurface Volatile Organic Compounds
                          BW-13about  A T
                          130 feet     A'
                    ESTIMATED EXTENT OF
                    HIGHLY CONTAMINATED
                    SOILS AND PERMANGANATE
                    INJECTION AREA
                                          /JS324
                                         ,t
                                                o
                        D


                        •   D
                        ML204
  GROUND-
  WATER FLOW
                          JS106
               BW11A-S  A
               BW-11 A-D**

                  BW-11 A
JS112  n

   }jS340
                       BW14 about
                       400 feet
                                                          50
                                                                      100 FEET
                                  EXPLANATION
                                          MONITORING WELL CLUSTER AND NUMBER -
                                          Color indicates maximum PCE concentration
                                          detected in micrograms per liter  (ng/L).
                                                    A
                                                    A
                                                    A
                                                    A
                                                    A
                           Less than 0.2
                           0.2 to 4.9
                           5.0 to 49
                           50.0 to 999
                           1,000 to 9,200
    D
    D
 PCE CONCENTRATION IN TREE-CORE
 SAMPLE — Concentrations in micrograms in
 headspace per kilogram of wet core (jig-h/kg).
   • Less than 0.5
   O 0.5 to 14.9
   O 15.0 to 49.9
   O 50.0 to 99.9
   • 100 to 120
MAXIMUM PCE CONCENTRATION IN SOIL BORING
SAMPLE AND BORING NUMBER - Concentrations in
micrograms per kilogram (jig/kg).
  • Less than 0.5
     0.5 to 240
     241 to 483
     484 to 4,999
     5,000 to 499,999
     Greater than 500,000
Figure 3.2.  Locations of soil borings and tree-core samples in area 1 and tetrachloroethene
(PCE) concentrations (modified from U.S. Environmental Protection Agency, 2003a).
PCE concentration in this tree was interpreted to indicate that
the tree was growing in PCE-contaminated soil or shallow
ground water. The PCE concentration of 74.9 |ig-h/kg in tree
JS340 was comparable to the concentration previously
detected in the 2001 core sample from adjacent tree JS112 of
99.1 |ig-h/kg, confirming results of the 2001 reconnaissance.
                          Using data from the 2001 and
                          2003 tree-core samples that
                          contained PCE concentrations
                          higher than 15 |ig-h/kg, a prob-
                          able PCE source area (referred
                          to as area 1) of about 1 acre was
                          delineated (fig. 3.1). The average
                          PCE concentration detected in
                          tree-core samples from area 1 was
                          75 ng-h/kg.
                              Based on delineation of
                          area 1 by the tree-core sampling
                          and the high PCE concentrations
                          detected in well cluster BW-10,
                          a nest of three monitoring wells
                          (BW-11 cluster) was installed
                          during 2004 adjacent to tree
                          JS112 (fig. 3.2). Perched water
                          containing several hundred
                          micrograms per liter of PCE was
                          encountered less than 15 ft deep
                          during drilling at this location,
                          confirming the presence of shal-
                          low PCE contamination initially
                          detected by tree-core sampling.
                          Data from the completed
                          BW-11 cluster indicated PCE
                          concentrations of 210 to 350 (ig/L
                          in perched water within the
                          overburden (11.5 to 15.5 ft deep),
                          PCE concentrations of 190 to
                          440 (ig/L in the shallow bedrock
                          (18 to 30 ft deep), and lower PCE
                          concentrations (33 to 36 (ig/L)
                          deeper in the bedrock (94 to
                          130 ft deep). The measured PCE
                          concentrations of 210 to 240 (ig/L
                          in perched water in well cluster
                          BW-11 compared favorably to
                          concentrations of about 400 (ig/L
                          predicted using the OU1 tree
                          core and ground-water relation.
                          The deeper monitoring interval
                          at cluster BW-11 is of similar
                          altitude to the deep interval at
                        cluster BW-10 that contained much
                        higher PCE concentrations
                        (320 to 2,300 (ig/L). A comparison
of water-level measurements in the two clusters indicated that
ground-water flow is northward from  BW-11 toward BW-10.
The substantially lower PCE concentrations in the deeper
bedrock at cluster BW-11 as compared to those in cluster
BW-10 and the presence of PCE in perched water within the
overburden indicated that cluster BW-11 was slightly south
and upgradient of a PCE source area.

-------
                                                                                      Appendixl. Case Studies    31
Soil             and                 of Tree-Core
Results

    During 2004 and 2005, a total of 41 soil borings were
done within area 1 (20 borings), at the former corporate
guest house (11 borings), and along the sanitary sewer
main (10 borings). These borings were installed to provide
definitive evidence of the extent and magnitude of subsurface
contamination within area 1, to confirm the absence of PCE
contamination at the former corporate guest house, and to
determine if widespread PCE contamination was present in
soils near the  sanitary sewer main. A total of 236 soil samples
were collected and analyzed by a portable gas chromatograph
(GC) for volatile organic compounds (VOCs) (234 samples)
or fixed laboratory (23 samples). Continuous soil cores
were obtained at each boring location and were typically
subsampled every 2 ft of depth for portable GC analysis.
    None of  the soil borings at the former corporate guest
house contained detectable concentrations of PCE  or other
target VOCs,  confirming the tree-core data, which  indicated an
absence of shallow subsurface contamination at this facility.
Four of 10 soil boring locations along the sanitary  sewer line
contained low (less than 240 micrograms per kilogram
(Hg/kg) concentrations of PCE. All concentrations  detected in
soils were less than one-half the U.S. Environmental Protec-
tion Agency (USEPA) Region 9 residential soil preliminary
remediation goal (PRO) of 484 jig/kg. PCE detections in
boreholes along the sewer line generally were found at depths
greater than 12 ft.
    Substantial PCE concentrations were detected in area 1
soil borings. PCE was detected in  18  of the 20 borings with
a maximum concentration of 1,200,000 (ig/kg in a  laboratory
soil sample collected from 13.5 ft deep in boring ML204 near
the center of area 1 (fig. 3.2). Several thin (0.5-in.-thick) bands
of black oily substance suspected to be PCE-rich DNAPL
(dense non-aqueous phase liquid) were present at depths
between 10 and 12 ft in this boring. On the basis of the soil
boring data, the footprint of the PCE-contaminated soils inside
area 1 was estimated to be less than about 5,000 square feet
(ft2) (Rob Blake, Black and Veatch Special Projects Corpora-
tion, oral commun., 2006). Generally, PCE concentrations
in area 1 soil borings  increased with increasing depth,  with
the largest extent of contamination in the 12- to 16-ft-deep
interval. Contamination extended through the soil into the top
of the weathered bedrock estimated at 11 to 18  ft below the
surface. Using the soil boring data, the USEPA estimated the
total volume of contaminated soil/residuum in area 1 is about
2,500 cubic yards (yd3) containing an estimated 760 kilograms
(kg) of PCE or about 125 gallons of pure PCE product.
    To provide additional evidence that area 1  was the likely
source of the bedrock PCE plume in the northern part of the
city, in 2005,  the USEPA installed two additional shallow (less
than 145 ft deep) monitoring well clusters near area 1. One
well cluster was installed upgradient (BW-14) and a second
well cluster (BW-13) was installed about 600 ft downgradient
from area 1 and about one-half the distance between area 1
and existing well cluster BW-10 (fig. 3.1). PCE concentrations
in upgradient well cluster BW-14 were less than 2 |,ig/L,
whereas PCE concentrations in the BW-13 cluster were
9,000 iig/L. Currently (2007), the USEPA is completing a
removal action to address the contaminated soils in area 1
using in situ chemical oxidation (permanganate oxidation) and
continuing with completion of the OU4 Remedial Investiga-
tion/Feasibility Study.




lines. J.I-.., andEmmett, L.E., 1994, Geohydrology of the
  Ozark Plateaus aquifer system it) parts of Missouri, Arkan-
  sas, Oklahoma, and Kansas: U.S. Geological Survey Profes-
  sional Paper 1414-D, 127 p.

Schumacher, J.G., Struckhoff,  G.C., and Burken, J.G., 2004,
  Assessment of subsurface chlorinated solvent contamina-
  tion using ttee cores at the Front Street site and a former
  diy cleaning facility at the Riverfront Superfund site, New
  Haven, Missouri, 1999-2003: U.S. Geological  Survey Sci-
  entific Investigations Report 2004-5049, 35 p.

U.S. Department of Commerce, 2002, Climatological Data,
  accessed March 2002 at http://lwf.ncdc.noaa.gov/oa/ncdc.
  html.

U.S. Environmental Protection Agency, 2001, Expanded
  site investigation/remedial investigation results for the
  Riverfront Superfund site, New Haven, Missouri: U.S.
  Environmental Protection Agency Region 7, Contract
  DW1495212801-2, 56 p.

U.S. Environmental Protection Agency, 2003a, Focused
  remedial investigation of operable units OU1 and OU3,
  Riverfront Superfund site, Franklin County, Missouri:
  U.S. Environmental Protection Agency Region 7, Contract
  DW1495217301-2, 128 p. plus appendixes.

U.S. Environmental Protection Agency, 2003b, Record of
  decision, Operable Unit 1, Front Street site, Franklin
  County, Missouri: U.S. Environmental Protection Agency
  Region 7, MOD981720246, 107 p. including appendixes.

U.S. Environmental Protection Agency, 2003c, Record of
  decision, Operable Unit 3, Old city dump site, Franklin
  County, Missouri: U.S. Environmental Protection Agency
  Region 7, MOD981720246, 53 p. including appendixes.

U.S. Environmental Protection Agency, 2006, Record of deci-
  sion, Operable Unit 5, Old hat factory, Franklin County,
  Missouri: U.S. Environmental Protection Agency Region 7,
  MOD981720246, 53 p. including appendixes.

-------
38     User's Guide to the Collection and Analysis of Tree Coresto       the Distribution of Subsurface Volatile Organic Compounds

              4:         and BIEX in
By James E. Landmeyer,1 Don A. Vroblesky,1 and Paul M. Bradley1


Reprinted with permission from Battelle Press (Landmeyer, I.E., Vroblesky, D.A., and Bradley, P.M., 2000, MTBE in trees
transpiring gasoline-contaminated ground water, Second International Conference on Remediation of Chlorinated and Recalci-
trant Compounds, May 22-25, 2000, Monterey, California: Columbus, Ohio, Battelle Press).
Abstract

    The fuel oxygenate compound methyl fcrf-butyl ether
(MTBE) and the conventional gasoline compounds benzene.
toluene, elhylbenzene, and the isomers of xylene and trimeth-
ylbcnzcne were detected and identified using purge-and-trap
gas chromatography/mass spectrometry methods in core
material of mature live oak trees (Quercus virginiana) located
above a gasoline-contaminated shallow aquifer. Conversely,
these gasoline compounds were not detected in core material
of oaks located outside of the gasoline plume. This detection
of gasoline compounds in trees at a contaminated field
site is important, particularly for the more soluble and less
biodegradable compounds MTBE and benzene, because it-
provides unequivocal field evidence that trees can act as sinks
to remove contaminants from ground-water systems.


Introduction

    Results of laboratory-scale studies have suggested that
herbaceous and woody plants have the potential to take up
a variety of dissolved petroleum-derived compounds during
transpiration.  For example, it has been recognized for some
time that pesticide uptake can occur in a wide variety of
non-woody plants, including bailey (Schone and Wood, 1972;
Donaldson et al, 1973; Briggs et al, 1982), bean (Lichtner,
1983), corn (Darmstadt et al., 1983; Leroux and Gredt, 1977;
Upadhyaya and Nooden, 1980), peanuts (Hawxby et al.,
1972), and soybeans (Moody et al., 1970; McEarlane et al.,
1987; McCrady et al., 1987). For woody plants, Burken and
Schnoor (1997) demonstrated the uptake and metabolism of
the widely used herbicide atrazine by poplar trees (Popultts
deltoides). Additionally, a preliminary report (Newman et
al.,  1999) indicated that poplar (Populus spp.) and eucalyptus
(Eucalyptus spp.) could take up the fuel oxygenate compound
methyl tert-butyl ether (MTBE) under laboratory conditions.
More recently, Burken and Schnoor (1998) reported the
uptake, translocation, and volatilization of the common
ground-water contaminants benzene, toluene, ethylbenzene,
and xylene (BTEX) by poplar cuttings in short-term hydro-
ponic experiments in the lab. Their results confirm that the
  'U.S. Geological Survey, Columbia, South Carolina.
relative ease of compound uptake is related to the logarithm
of the oclanol-water partition coefficient (log KaJ, as slated
initially by Briggs et al. (1982). Essentially, compounds
having a log Kow between 0.5 and 3.0 are preferentially taken
up by roots. Because the log Kow of the ground-water con-
taminants MTBE, benzene, toluene, elhylbenzene, o-xylene,
m-xylene,  andp-xylene arc within this range  (1.20, 2.13, 2.65,
3.13, 2.95, 3.20, and 3.18, respectively), their uptake during
laboratory transpiration studies is not surprising.
    However, the uptake of these gasoline compounds by
mature trees has not been documented under  field conditions.
For example, in the study cited above (Burken and Schnoor,
1998) that indicated uptake of MTBE by poplar (Populus
spp.) and eucalyptus (Eucalyptus spp.) cuttings under
laboratory conditions, no MTBE uptake was measured in
mature trees at an MTBE-conlaminated site. This current
study was  undertaken, therefore, to determine if the soluble
fuel compounds MTBE, benzene, toluene, elhylbenzene. and
o-, m-, andp-xylcnes shown to be taken up under laboratory
conditions are present in mature live oaks growing above
gasoline-contaminated ground water.
    Study Site. The study site is a gasoline  station (fig. 4.1)
near Beaufort, South Carolina (SC). Fuel-oxygenated gasoline
from a leaking underground storage tank was detected in the
shallow, water-table aquifer in late 1991 (Landmeyer et al.,
1996). The water-table aquifer is comprised of well-soiled
sand. The water-table aquifer is underlain by  a regional clay-
rich confining unit at around 45 feet (ft) (13.7 m). There is less
than 0.01 % natural sedimentary organic matter in the sandy
aquifer. The depth to water is about 13 ft (3.9 m) near  the
release area and from 9 to 2 ft (2.7 to 0.6 m) near a drainage
ditch approximately 700 ft (215 m) downgradient (fig. 4.1)
of the release area. Recharge  to the water-table aquifer is by
rainwater infiltration, with precipitation approaching 60 inches
per year (in/yr) (132 cm/yr).
    The study site is characterized by a dense stand of mature
(>40 years old) live oak trees (Quercus virginiana) (fig. 4.1).
Live oaks derive their common name from their ability to
maintain leaves  throughout winter, even though they arc
deciduous. As a result, live oaks  transpire water continually
throughout the year and, therefore, are an excellent genus
to study transpiration-related processes. The trees at the site
have well-developed and extensive networks of horizontal and
vertical roots, as evidenced by conspicuous root material at

-------
                                                                                      Appendixl. Case Studies    39
  LJ
   C


                              10,110(1  1.0110  2(1
       50  100   150 200 FEET
                                      Tree 7
      EXPLANATION

EQUAL MTBE CONCENTRATION (ng/L)


EQUAL BENZENE CONCENTRATION (ng/L)


MONITORING WELL

MULTILEVEL SAMPLER


TREE LOCATION AND NUMBER
Figure 4.1.  Study site near Beaufort, South Carolina, indicating location and
reference number of tree samples, monitoring wells mentioned in text, and
isoconcentration contours of methyl fert-butyl ether (MTBE) and benzene in ground
water (collected January 1998).
land surface some distance from tree trunks and the presence
of observable root material at the water table in boreholes
completed near trees.


Methods

     Ground-Water Geochemistry. The distribution of
gasoline compounds as well as geochemical parameters that
indicate the redox zonations at the site have been documented
over seven sampling events between 1993 and 1998 (Land-
meyer et al., 1996; Landmeyer et al., 1998). However, only
the gasoline compound distribution will be discussed here.
                Conventional polyvinyl chloride (PVC)
                monitoring wells (2-inch [4.4 cm] diameter,
                screened across or below the water table
                with 12.5 ft [3.8 m] screens) and multi-level
                sampling wells (1-inch [2.2 cm] diameter,
                with variably spaced screened intervals)
                were analyzed for MTBE and BTEX at
                each sampling event. Before sampling, each
                well was purged until stable measurements
                of water temperature (in degrees Celsius)
                and pH (in standard units) were obtained.
                MTBE and BTEX samples were collected in
                40-mL glass vials using a peristaltic pump
                at a low flow rate, preserved with 3 drops of
                concentrated hydrochloric acid, and capped
                using Teflon-lined septa. BTEX compounds
                were quantified using purge-and-trap gas
                chromatography with flame-ionization
                detection. MTBE was quantified using
                direct-aqueous injection gas chromatography
                with mass spectrometry (GC/MS) detection
                by the Oregon Graduate Institute (Church et
                al., 1997).
                     Tree-Core Sample Collection and
                Analysis.  Cores of tree tissue were obtained
                from trees located in uncontaminated areas
                upgradient of the ground-water source
                area and plume, and from trees growing
                in the area delineated by dissolved-phase
                ground-water contamination (fig. 4.1) using
                an increment borer in mid-June 1999. Tree
                coring methods have been used previously
                to determine the presence of chlorinated
                solvents (Vroblesky et al., 1999) and metals
                (Forget and Zayed, 1995) in tree rings.
                Cores were collected at a height of 1 ft
                (0.3 m) above ground on the northeast side
                of each tree. Replicate cores about 2 inches
                apart were taken at each tree sampled. The
                average core collected was about 2.0 inches
                by 1/4 inch (volume of 0.09 in3) (4 cm by
                0.5 cm; volume of 0.72 cm3), and consisted
                of the most recent growth rings,  which
                contain the water-conducting xylem in ring-
                porous trees such as oaks. Each core was
immediately placed into a 40-mL glass vial and capped with
a Teflon stopper. At the time of sampling, the  air temperature
was about 85 degrees Fahrenheit (°F), skies were sunny, winds
were from the west at 5 miles per hour (mph) and low relative
humidity (<60%). Because the site is an active gas station, air
samples for gasoline compound detection were also collected
in 40-mL glass vials, after waving an open vial for a few
seconds near the contaminated area downgradient of the fuel
release.
    In the laboratory, the volatile organic compounds in the
tree core were separated and identified using a purge-and-trap

-------
40     User's Guide to the Collection and Analysis of Tree Coresto       the Distribution of Subsurface Volatile Organic Compounds
GC/MS method similar to U.S. EPA method 8260. Prior to
purging of each sample, 5 mL of pesticide-grade methanol
was added to each 40-mL vial containing core material,
and brought to a final volume of 25 mL using organic-free
reagent water. Each vial was then purged with helium, the
volatile compounds trapped in a tube containing sorbent
material, and manually injected onto a 30 in, 0.25 mm inside
diameter capillary column coated with Rtx 502.2 (RESTEK)
at a 1.4 pn film thickness. Identification of target gasoline
compounds was confirmed by comparing sample mass spectra
to the mass spectra of reference material from the National
Institute of Standards and Technology under identical run
conditions. Three internal standards (fluorobenzene, 2-bromo-
1-chloropropane, and 1,4-dichlorobutane) and three surrogate
standards (1,4-difluorobenzene, d8-toluene, 4-bromofluo-
robenzene) were used. Surrogate recoveries ranged between
93 and 100%. Target compound concentrations are reported
as concentrations in micrograms per liter in the headspace of
vials containing core material.


Results and Discussion

    MTBE and BTEX Detection in Tree Cores. MTBE,
benzene, toluene, ethylbenzene, and the xylene isomers
were not detected in the headspace samples of core material
collected from oaks growing hydrologically upgradient of the
release area (table 4.1, Trees  1 and 2; fig. 4.1). Trees 1 and
2 are located about 146 ft (46 m) and 136 ft (45 in) north of
well 5, respectively (fig. 4.1). Well 5 is located about 75 ft
(35 m) upgradient of the release area, screened across the
water table, and MTBE and BTEX concentrations there have
remained below detection limits since monitoring activities
began at the site in 1993 (Landmeyer et al, 1998).
 Table 4.1.  Concentrations (in micrograms per liter) of the gasoline compounds MTBE,
 benzene, toluene, ethylbenzene, and the isomers of xylene and trimethylbenzene (TMB) in
 the headspace of vials containing tree cores collected at a field site near Beaufort, South
 Carolina, June 1999,

 [Non deteclion is represented by nd. nol analyzed by naj
Compound

MTBE
Benzene
Toluene
Ethylbenzene
m,p -Xylene
o-Xylene
1.3,5-TMB
1,2,4-TMB
Uppa client of
Former Source
Area
Tree 1
nd
nd
nd
nd
nd
nd
nd
nd
Tree 2
nd
nd
nd
nd
nd
nd
nd
nd
Former Source
Area
TreeS
nd
nd
5.4
nd
nd
nd
nd
6
Tree 4
nd
nd
nd
nd
5
6.3
nd
19.6
Dissolwed-Phase
Plume
Tree 5
9.4
7.2
26.2
nd
10.1
5.6
7.8
5.3
Tree 6
54
4.8
10
8.5
10.8
7.4
6.1
5.4
Vertical Flow
Area
Well?
5,800
508
674
149
580

na
na
Tree?
nd
nd
nd
nd
nd
nd
nd
nd
     However, MTBE, benzene, toluene, ethylbenzene, and
the xylene isomers were detected in the headspace of core
samples taken from trees growing above the former source
area (table 4.1, Trees 3 and 4; fig. 4.1) and the delineated
plume of ground-water contamination (table 4.1, Trees 5
and 6; fig. 4.1). Headspace samples of core material from
Tree 3 had detections for toluene at 5.4 (ig/L, and Tree 4
had no toluene but m-, p-xylene (5  |,ig/L total), and oxylene
(6.3  (ig/L) were detected. These detections in Trees 3 and 4 are
related to the residual contamination in the former source area
due to incomplete removal of contaminated sediments in 1993
(Landmeyer et al., 1998). This incomplete removal of source-
area material has also caused a "wake" of dissolved-phase
contamination to continue to be observed between the source
area and downgradient wells in the direction of ground-water
flow, even 6 years after source removal activities (fig. 4.1).
     Headspace samples of core material from live oaks
sampled in the area delineated by gasoline compounds down-
gradient of the former source area contained chromatographic
peaks confirmed by MS to be MTBE, benzene, toluene,
ethylbenzene, and the xylene and trimethylbenzene (TMB)
isomers (table 4.1, Trees 5 and 6; fig. 4.1). For example,
headspace samples of core material from Tree 6, located about
11 ft (4 m) east of well 8 (fig. 4.1), had 54.0 [ig/L MTBE,
4.8 |ig/L benzene, 10.1 |ig/L toluene, 8.5 [ig/L ethylbenzene,
10.8 |ig/L m- and/7-xylene, 7.4 |ig/L o-xylene, 6.1 (ig/L
1,3,5-TMB, and 5.4  (.ig/L 1,2,4-TMB. The concentration of
MTBE in Tree 6  was the highest detected in all trees cored.
Tree 5 adjacent to Tree 6 had the highest detection of toluene
(26.2 (ig/L).  In August 1998, samples of ground water from
well 8, which is screened across the water table in the area
where root penetration has been observed, had 5,800 [ig/L
MTBE, 508  (ig/L benzene, 674 fig/L toluene, 149 (ig/L
ethylbenzene, and 580 [ig/L total xylenes. This detection of
                      MTBE and benzene in transpiration-
                      stream water of a mature tree at a
                      contaminated field site is the first
                      known field-scale confirmation of
                      laboratory-scale experimental data.
                      Headspace samples of air collected
                      from this area did not contain peak
                      responses representative of MTBE,
                      benzene, toluene, ethylbenzene, or
                      the isomers of xylene or TMB  (data
                      not shown).
                           No compound detections for
                      MTBE and BTEX were seen in
                      headspace samples of core material
                      from Tree 7 (table 4.1), even though
                      this tree is located downgradient
                      of the release area. This lack of
                      compound detection can be explained
                      by the location of Tree 7 being (1)
                      at the edge of the delineated plume
                      boundary (fig. 4.1), which probably
                      results in the majority of transpiration

-------
                                                                                     Appendixl. Case Studies     41
water being derived from uncontaminated ground water, and
(2) in an area where dissolved-phase contamination originally
near the water-table surface is pushed deeper into the aquifer
by vertical recharge of percolating rainwater (Landmeyer et
al, 1998). This vertical displacement of the dissolved-phase
plume deeper into the aquifer away from root interaction is
why no trees were cored downgradient of Tree 7.
    As stated above, the trimethylbenzene isomers
1,3,5-TMB and 1,2,4-TMB were also identified in the cored
material in Trees 3, 4, 5, and 6 located above the original
source area and delineated dissolved-phase plume (table 4.1).
The TMB isomers are common components of gasoline, and
because their solubility and sorption characteristics are similar
to benzene and toluene, these relatively nonbiodegradable
isomers are routinely used as conservative tracers to estimate
biodegradation rates of aromatic hydrocarbons from field data
(Weidemeier et al., 1997). The detection of the TMB isomers
in transpiration stream water follows from a log Kow of 3.78 for
1,2,4-TMB. The fact that trees can remove TMB isomers from
contaminated ground water needs to be  considered if TMB
isomers are to be used as conservative tracers in ground-water
studies of contaminant transport.
    The chlorinated compounds chloroform and methyl
chloride were detected in tree cores collected in uncontami-
nated and contaminated areas. Chloroform concentrations in
the headspace of vials containing tree cores ranged from 18.7
to 89.1 (ig/L, and methyl chloride concentrations ranged from
20.1 to 63.3 [ig/L (data not shown). Chloroform and methyl
chloride have log Kow's that would suggest uptake by trees
(1.90 and 0.90, respectively). Their detection in tree cores
suggests that the most likely source is chlorinated irrigation
water. A nationwide survey of 1,501 shallow ground-water
samples conducted by the U.S. Geological Survey indicated
that chloroform was the most commonly detected volatile
organic compound in shallow wells (Squillace et al.,  1996).
    The detection of the common ground-water contaminants
MTBE, BTEX, and the TMB isomers in mature trees that
grow above a shallow aquifer characterized  by a fuel spill is
important because it extends laboratory-scale observations to
real field sites. These results suggest the possible use of trees
to remove soluble gasoline-related compounds such as MTBE
and benzene from contaminated ground-water systems. The
transpiration process of trees requires large volumes of water
(up to 53 gal/day for 5-year old trees [Newman et al., 1997J)
to balance transpiration losses. Although trees most commonly
use recent rainfall to meet short-term water demands, ground
water can provide water during times of low precipitation
to meet longer-term needs. Because tree-root systems often
contact the water-table surface, the potential exists for sources
of contaminants containing non-aqueous phase liquids, such as
petroleum hydrocarbon compounds and chlorinated solvents
dissolved in ground water to come into  contact with tree roots,
particularly in discharge zones where ground-water flowlines
converge to bring even the denser chlorinated compounds to
the surface. Results from our study suggest that trees exhibit
the potential to uptake synthetic organic compounds dissolved
in ground water, particularly those gasoline-related compounds
that are accidentally released into the environment. It is not yet
clear whether uptake of soluble ground-water contaminants
by trees may serve to remove substantial amounts of hydro-
carbons from contaminated ground-water systems. However,
these results show that contaminant uptake occurs in measur-
able quantities, and suggest that mis phenomenon may have
important environmental applications.


References

Briggs, G.G., R.H. Bromilow, and A.A. Evans. 1982. Pesticide
  Science. 13, 495-504.

Burken, J.G., and J.L. Schnoor. 1997. Environmental Science
  & Technology. 31,1399-1406.

Burken, J.G., and J.L. Schnoor. 1998. "Predictive relationships
  for uptake of organic contaminants by hybrid poplar ttees."
  Environmental Science & Technology. 32, 3379-3385.

Church, C.D., L.M. Isabelle, J.F. Pankow, D.L. Rose, and P.O.
  Tratnyek.  1997. "Method for determination  of methyl tert-
  butyl ether and its degradation products in water." Environ-
  mental Science & Technology. 31, 3723-3726.

Darmstadt, G.L., N.E. Balke, and L.E. Schrader. 1983. Pesti-
  cide Biochem. Physiology. 19, 172-183.

Donaldson, T.W., D.E. Bayer, and O.A. Leonard. 1973. Plant
  Physiology. 52, 638-645.

Forget, E., and J. Zayed. 1995. "Tree-ring analysis for moni-
  toring pollution by metals," in Lewis, T.E., eds., Tree, rings
  as indicators of ecosystem health: CRC Press, 157-176.

Hawxby,  K., E. Easier, and P.W. Santelmann. 1972. Weed Sci-
  ence. 20, 285-289.

Landmeyer, .I.E., F.H. Chapelle, and P.M. Bradley.  1996.
  "Assessment of intrinsic bioremediation of gasoline con-
  tamination in the shallow aquifer, Laurel Bay Exchange,
  Marine Corps Air Station, Beaufort, South Carolina." U.S.
  Geological Survey Water-Resources Investigations Report
  96-4026, 50 p.

Landmeyer, I.E., F.H. Chapelle, P.M. Bradley, J.F. Pankow,
  C.D. Church, and P.O. Tratnyek. 1998. "Fate of MTBE
  relative to benzene in a gasoline-contaminated aquifer
  (1993-1998)" Ground Water Mon. & Kerned, 18, 93-102.

Landmeyer, J.E., D.A. Vroblesky, and P.M. Bradley. 2000.
  MTBE and BTEX in trees above gasoline-contaminated
  ground water, in Wickramanayake, G.B., and others, eds.,
  Case studies in the remediation of chlorinated and recal-
  citrant  compounds, Proceedings of the 2nd International
  Conference on Remediation of Clorinated and Recalcitrant
  Compounds, Monterey, California, May 22-25, 2000,
  v. C2-7, p. 17-24.

-------
42    User's Guide to the Collection and Analysis of Tree Coresto       the Distribution of Subsurface Volatile Organic Compounds
Leroux, P., and M. Gredt. 1977. Neth. Journal Plant Pathol.
  83, 51-61.

Lid-liner, F.T.  1983. Plant, Physiology, 71, 307-312.

McCrady, J.K., C. McFarlane, and F.T. Lindstrom. 1987. J.
  Experimental Botany. 38, 1875-1890.

McFarlane, J.C., T. Pfleeger, and J. Fletcher. 1987. J. Environ.
  Quality. 16, 372-376.

Moody, K., C.A. Kust, andD.P. Buchholtz. 1970. Weed Sci-
  ence. 18, 642-647.

Newman, L., S.  Strand, J. Duffy, N. Cnoe, G. Ekuan, M.
  Ruszaj, B. Siiurtleff, J. Wilmoth, P. Heilman, and M.
  Gordon. 1997. Environmental Science  & Technology. 31,
  1062-1067.

Newman, L.A., M.P. Gordon, P. Heilman, D.L. Cannon, E.
  Lory, K. Miller, J. Osgood, and S.E. Strand.  1999. "Phy-
  toremediation of MTBE at a California naval site." Soil &
  Groundwater Cleanup. 42-45.
Scfaone, M.G.T., and A.V. Wood. 1972. Weed Research. 12,
  337-347.

Squillace, P.J., J.S. Zogorski, W.G. Wilber, and C.V. Price.
  1996. Environ. Sci. "& Tech. 30, 1721-1730.

Upadhyaya, M.K., and L.D. Nooden. 1980. Plant Physiology.
  66, 1048-1052.

Vroblesky, D.A., C.T. Nietch, and J.T. Morris. 1999. "Chlo-
  rinated ethenes from groundwater in tree trunks." Environ.
  Science & Technol. 33, 510-515.

Wcidcmcicr, T.H., M.A. Swanson, J.T. Wilson, D.H. Kamp-
  bell, R.N. Miller, and I.E. Ilansen. 1997. Intrinsic Bioreme-
  diation: Battelle Press, 3(1), 31-51.

-------
Appendix 2
                                              EIA-FLDGRAB4.WPD
                                              VOCs in  Air Samples
                                              02/12/02
                                              Page 1 of 17
                          AIR SAMPLE ANALYSIS
                                   FOR
                     VOLATILE ORGANIC COMPOUNDS
               By Scott Clifford, U.S. Environmental Protection Agency
              The Office of Environmental Measurement and Evaluation
                           EPA Region New England
                              11 Technology Dr.
                         North Chelmsford, MA 01863

  This reprint contains minor formatting modifications by the U.S. Environmental Protection
                  Agency from document EIA-FLDGRAB4.WPD

-------
                                                   EIA-FLDGRAB4.WPD
                                                   VOCs  in  Air  Samples
                                                   02/12/02
                                                   Page  2 of 17
                                 Table of Contents
      Section
Subject
Page
       1.     Scope and Application	3

       2.     Summary of Method	4

       3.     Definitions	4

       4.     Health and Safety Warnings	5

       5.     Cautions	6

       6.     Interference	6

       7.     Personnel Qualifications	6

       8.     Equipment and Supplies	7

       9.     Instrument Preparation	8

       10.    Sample Analysis	10

       11.    Identification and Quantitation	12

       12.    Data and Records Management	12

       13.    Quality Control	13

       14.    References	14

Tables:

       1.     Quality Control Table	15

Figures:

       1.     Chromatogram - PID	16

       2.     Chromatogram - BCD	17

-------
                                                      EIA-FLDGRAB4.WPD
                                                      VOCs in Air Samples
                                                      02/12/02
                                                      Page 3  of  17
1.0 Scope and Application:
1.1     The procedure contained herein is applicable to all EPA Region I chemists performing
       screening for volatile organic compounds for air grab samples.

1.2     Reporting Levels:

  Reporting levels can vary depending upon instrument performance and settings, as well as
data quality objectives. Typical achievable reporting levels using a photoionization detector
(PID) and an electron capture detector (BCD) are given below.

                                                  Reporting limit in
                   Analyte                        parts per billion by
                  	volume (ppb/v)
                   1,1-Dichloroethene:                    10
                   trans-1,2- Dichloroethene:               10
                   c/5-l,2-Dichloroethene:                 15
                   Benzene:                              10
                   Trichloroethene:                        10
                   Toluene:                              40
                   Tetrachloroethene:                       2
                   Ethylbenzene:                          50
                   Chlorobenzene:                        50
                   w/p-xylenes:                           50
                   o-xylene:                              80
                   1,1,1- Trichloroethane:	6	

1.3     This method may be used when the quality assurance objectives are either QA1 or QA2
       as defined in Interim Final Guidance for the Quality Assurance/Quality Control Guidance
       for Removal Activities, April 1990.  Briefly, QA1 is a screening objective to afford a
       quick preliminary assessment of site contamination. QA2 is a verification objective used
       to verify analytical (field or lab) results. A minimum of 10% of samples screened must
       be analyzed by a full protocol method for qualitative and quantitative confirmation.

-------
                                                     EIA-FLDGRAB4.WPD
                                                     VOCs in Air Samples
                                                     02/12/02
                                                     Page 4  of  17
2.0    Summary of Method:
2.1     Field screening using the portable gas chromatograph is used for tentative identification
       and quantitation of volatile organic compounds in air samples.  This screening technique
       can provide quick and reliable results to assist in important on-site decision making.

2.2     An aliquot of the air sample is injected into a calibrated gas chromatography (GC)
       equipped with a photoionization detector (PID) and electron capture detector (BCD).
       The compounds are separated on a megabore capillary or packed column. Retention
       times are used for compound identification and peak heights are used for quantitation of
       the identified compounds.

2.3     This method can be used to provide analytical data in a timely manner for guidance of
       ongoing work in the field.

2.4     Based on the project=s data quality objectives (DQOs), the operator can modify some
       conditions.  For example, the injection volumes can be  changed depending on the levels
       found at the site.
3.0 Definitions:
3.1     FIELD DUPLICATES (FD1 and FD2):  Two separate samples collected at the same time
       and place under identical circumstances and treated exactly the same throughout field and
       laboratory procedures. Analyses of FD1 and FD2 give a measure of the precision
       associated with sample collection, preservation, and storage, as well as with laboratory
       procedures.

3.2     Headspace: Air above water standard in sample vial.

3.3     Laboratory Duplicate (LD1  and LD2): Two injections from the same sample.  The
       analyses of LD1  and LD2 give a measure of the precision associated with the laboratory
       procedure.

3.4     LABORATORY REAGENT BLANK (LRB) - An aliquot of reagent water or other
       blank matrix that is treated exactly as a sample including exposure to all glassware,
       equipment, solvents, reagents, internal standards, and surrogates that are used with other
       samples. The LRB is used to determine if method analytes or other interferences are
       present in the laboratory environment, the reagents, or the apparatus.

-------
                                                     EIA-FLDGRAB4.WPD
                                                     VOCs  in Air  Samples
                                                     02/12/02
                                                     Page  5 of 17
3.5    STOCK STANDARD SOLUTION - A concentrated solution containing one or more
       method analytes prepared in the laboratory using assayed reference materials or
       purchased from a reputable commercial source.

3.6    WORKING STANDARD SOLUTION - A solution of several analytes prepared in the
       laboratory from stock standard solutions and diluted as needed to prepare calibration
       solutions and other needed analyte solutions.

3.7    SECONDARY STANDARD - A standard from another vender or a different lot number
       that is used to check the primary standard used for quantitation.
4.0    Health and Safety Warnings:
4.1     The toxicity or carcinogenicity of each reagent used in this method has not been precisely
       determined; however, each chemical should be treated as a potential health hazard.
       Exposure to these reagents should be reduced to the lowest possible level. The
       laboratory is responsible for maintaining a current awareness file of OSHA regulations
       regarding the safe handling of the chemicals specified in this method. A reference file of
       data handling sheets should be made available to all personnel involved in these analyses.
       Use these reagents in a fume hood whenever possible and if eye or skin contact occurs
       flush with large volumes of water.

4.2     Always wear safety glasses or a shield for eye protection, protective clothing, and
       observe proper mixing when working with these reagents.

4.3     Some method analytes have been tentatively classified as known or suspected human or
       mammalian carcinogens. Pure standard materials and stock standard solutions of these
       compounds should be handled with suitable protection to skin, eyes, etc.

-------
                                                      EIA-FLDGRAB4.WPD
                                                      VOCs  in Air  Samples
                                                      02/12/02
                                                      Page  6  of 17
5.0    Cautions:
5.1     The stock standard and secondary stock standard are replaced every three months.

5.2     The working and secondary standards are good for 7 days provided these standards are
       stored on ice with no headspace.
6.0    Interferences:
6.1     Method interferences may be caused by contaminants in solvents, reagents, glassware
       and other sample processing hardware that lead to discrete artifacts and/or elevated
       baselines in the chromatograms.  All of these materials must routinely be demonstrated to
       be free from interferences under the conditions of the analysis by running laboratory
       method blanks.

6.2     Matrix interferences may be caused by contaminants that coelute with the target
       compounds.  The extent of matrix interferences will vary considerably from source to
       source. A different column or detector may eliminate this interference.

6.3     Contamination by carry-over can occur whenever high level and low level samples are
       sequentially analyzed.  To reduce carry-over, a VOC free water blank should be analyzed
       following an unusually concentrated sample to assure that the syringe is clean.
7.0    Personnel Qualifications:
7.1     The analyst should have at least a four year degree in a physical science.

7.2     The analyst should be trained at least one week and have a working knowledge of this
       method and quality control before initiating the procedure.

7.3     All personnel shall be responsible for complying with all quality assurance/quality
       control requirements that pertain to their organizational/technical function.

-------
                                                     EIA-FLDGRAB4.WPD
                                                     VOCs  in Air  Samples
                                                     02/12/02
                                                     Page  7  of  17
80    Equipment and Supplies:


8.1    Photovac 10A10 portable gas chromatography equipped with a PID and a 4 ft, 1/8 in, SE-
       30 packed column.

8.2    Shimadzu 14A portable gas chromatography equipped with a PID, BCD, and a 30 m,
       0.53 mm megabore DBFS 624 capillary column, or equivalent.

8.3    Syringes: Hamilton, steel barrel, 250 uL to 500 uL.

8.4    Vial: 40 mL VOA vials with Teflon lined septum caps.

8.5    Air Standard:

8.5.1   Standard Preparation and Use: Standard should be prepared in water at a 10 ug/L
             concentration, and labeled. Standards should be made up fresh weekly from a
             methanol stock solution (Supelco or equivalent vender), and stored with  no head
             space on ice until ready for use. Standard preparation should be recorded in the
             Field Standard Log notebook. After preparation, the standard is placed  into a 40
             mL VOA vial, filling the vial to the top leaving no head space.  The standard is
             then put into a cooler on an ice bath for storage until it is ready to use. When the
             standard is ready to use in connection with air sampling and analysis, 10 mL of
             liquid from the 10 ug/L standard VOA vial are withdrawn to give a head space
             above the liquid standard. The standard is then placed into an ice bath.  It is
             important to realize that the concentration of the volatile organic compounds in
             the head space was calibrated at approximately 0 - 1°C.  Therefore, it is
             mandatory that the working standard be stored in a cooler in an ice bath, septa
             side down.

-------
                                                   EIA-FLDGRAB4.WPD
                                                   VOCs in Air Samples
                                                   02/12/02
                                                   Page 8  of  17

.5.2  The head space above a 10 ug/L aqueous standard at approximately 0 - 1°C (Standard
            must be in an ice bath) is used for an air standard. Through in-house
            experimentation, we have determined the vapor concentration* of various volatile
            organic compounds in the head space of a 10 ug/L aqueous standard at
            approximately 0 - 1°C to be as follows:

                         1,1 - Di chl oroethene:         5 5 4 ppb/v
                         trans 1,2-Dichloroethene:   202 ppb/v
                         cis 1,2- Dichloroethene:     90 ppb/v
                         Benzene:                  151 ppb/v
                         Trichloroethene:            142 ppb/v
                         Toluene:                   159 ppb/v
                         Tetrachl oroethene:          201 ppb/v
                         Ethylbenzene:              145 ppb/v
                         Chlorobenzene:             70 ppb/v
                         m/p-xylenes:               136 ppb/v
                         o-xylene:                  112 ppb/v
                         1,1,1 - Tri chl oroethane:       3 3 0 ppb/v

-------
                                                     EIA-FLDGRAB4.WPD
                                                     VOCs  in Air  Samples
                                                     02/12/02
                                                     Page  9 of 17
9.0 Instrument Preparation:
       The Photovac 10A10 GC and the Shimadzu GC 14A should always have carrier gas
       flowing through their columns. The Photovac uses zero air and the Shimadzu uses zero
       nitrogen as carrier gas.
9.1     The following steps are taken before analysis of samples on the Photovac:

       o  Check detector.  Insure that the detector source is on by observing the "source off"
          lamp (red) on the face of the instrument.  When the source is on, the "source off
          lamp should not be illuminated.  Another method of checking the detector is to
          remove the detector housing with an alien wrench. With the detector on, you will
          observe a purple glow inside the Teflon detector chamber.

       o  Check carrier gas flow.  The gas flow is checked using a  flow meter hooked up to the
          detector out vent port. Flow can be adjusted to the desired rate by using the vernier
          knobs on the left side of the instrument face or by adjusting the delivery pressure on
          the carrier gas cylinder regulator. A desirable flow is from 200 - 600 cc/min,
          depending upon application.

       o  Check injection port septum.  It is a good idea to put in a new septum before
          analyzing a large number of samples.

       o  Check to be sure that signal cable is connected from Photovac output to strip chart
          recorder input.

       o  Set strip chart recorder input to 100 MV full scale and chart speed to 60 cm/hr. for
          Photovac 10A10. (Recorder input for Photovac 10S50 should be set to 1 V full scale).

       o  Adjust needle on Photovac output meter using the offset dial so when instruments
          attenuation is changed, the needle does not deflect. Setting the output somewhere
          between 4-10 Mv DC will usually achieve this.

       o  Set recorder zero to 5% of chart full scale and establish an acceptable base line.

9.2     The following steps are taken before analysis of samples on the Shimadzu GC 14A using
isothermal conditions.

          o  Check injection port septum. It is a good idea to put in a new septum before
             analyzing a large number of samples. The system must be cool before changing

-------
                                                     EIA-FLDGRAB4.WPD
                                                     VOCs  in Air  Samples
                                                     02/12/02
                                                     Page  10 of 17
             the septum.
          o  Check PID detector Temperature. It should be set to 150°C from the external PID
             power source. It can take up to 3 hrs. to warm up the detector from cold. Insure
             that the detector lamp is on by quickly observing the lamp (purple) on the left side
             of the instrument.

          o  Turn on the instrument and the instrument heaters on the face of the instrument.
             On the control keyboard, hit the START button and set the default temperature
             conditions.
                              o  Injector 125°C
                              o  BCD detector 190°C
                              o  Oven 60°C

                       •  After a 30 -60 minute warm-up, monitor actual temperatures using
                          the control keyboard.

          o  Check zero nitrogen carrier gas flow. The gas flow is checked using a flow meter
             hooked up to the detector out vent port. Flow can be adjusted to the desired rate
             by using the vernier knobs on the gas control unit on top of the instrument. A
             desirable flow is from 20 - 60 cc/min, depending upon application.

          o  Check to be sure that signal cables are connected from PID and BCD outputs to
             strip chart recorder inputs.

          o  Set strip chart recorder input to 5 MV full scale for the PID and 50 MV full scale
             for the BCD, and chart speeds to 60 cm/hr.

          o  Set recorder zero to 5% of chart full scale and establish acceptable base lines.


10.0 Sample Analysis:
             Air analysis generally consists of taking a 200 uL volume grab sample of air
             using a 250 uL steel barrel syringe with a 2 inch, 25 gauge needle, and injecting it
             into the GC injection port.

             At the sample collection location, flush the syringe barrel three times using the
             plunger. After flushing, pull the plunger up to the 200 uL point on the barrel and
             place a spare GC septa on the tip of the needle to seal in the sample.  Get the

-------
                                        EIA-FLDGRAB4.WPD
                                        VOCs in Air Samples
                                        02/12/02
                                        Page 11 of 17

sample to the GC as soon as possible for analysis. Put the syringe needle into the
GC injection port, and push the needle through the septum until the barrel comes
up against the injection port and immediately push the plunger with a quick
action. Turn on the strip chart recorder and note on the chart:

             1. start  of run
             2. sample number
             3. sample volume
             4. attenuation or gain
             5. any  other relevant comments

The order in which analyses of a group of samples is performed is as follows:

       1.      Standard - Inject a 200 uL sample of your 10 ug/L standard, at 0 -
             1 °C head space into the GC.  Keep standard peaks at
             approximately 50% scale or more, if possible, by adjusting the
             attenuation or gain.

       2.     Repeat 10 ug/L standard to check for reproducibility. Standard
             chromatograms should have compound peak heights within + 15%
             of each other and identical retention times.

       3.     Inject the secondary standard for confirmation. The acceptance
             criteria is + 20% of the true value.

       4.     Blank - Inject a 200 ul sample of clean air into the gas
             chromatograph with the attenuation set at the same level or lower
             than what your samples will be run on.  Blank clean air is taken
             from the head space above VOA free water in a 40 ml VOA vial.

       5.     Samples - Inject 200 ul sample volumes into the GC at the same
             attenuation or lower, than the standard was run. If contaminant
             levels on the chromatograms are off-scale on the recorder, adjust
             the attenuation or gain to decrease instrument response. If the
             chromatographic peaks are still off-scale rerun the samples using a
             smaller injection volume.

       6.     Repeat 10 ug/L standard every 10 samples and at the  end of the
             sample batch to check the calibration and reproducibility.
             Standard chromatograms should have compound peak heights
             within  + 20% of each other and identical retention times.

-------
                                                    EIA-FLDGRAB4.WPD
                                                    VOCs in Air Samples
                                                    02/12/02
                                                    Page 12 of  17
11.0 Identification and Quantitation:


      Identifications of compounds present in a sample are made by matching retention times
      of peaks in the sample chromatogram to the retention times of standard peaks. After a
      compound is identified, quantitation is done by a peak height comparison.

      Example:     If the 10 ug/L aqueous standard head space had a benzene peak height of
                   32 units from a 200 uL injection with instrument attenuation at 2, an
                   identified benzene peak 12 units high from an 200 uL sample injection
                   with instrument attenuation at 2 would represent a sample concentration of
                   57 ppb benzene.

                            32 units    =   12 units
                            151ppb/v*       Xppb/v

                                X = 57 ppb/v Benzene

      * See Air Standard Section 8.5.2


120 Data and Records Management:


12.1   All work performed for the analyses of samples should be entered into the field screening
      logbook.  The analyses data should be presented to the project manager on  site. This is
      followed up by an Internal Correspondence Report that is reviewed by the Advanced
      Analytical Chemistry  Expert from the Chemistry Section of the EIA Laboratory.
      Chromatograms generated should be saved and filed in the project folder. The samples
      analyzed should also be logged into the laboratory information management system.

12.2   Chromatograms:

12.2.1 Site name, analyst name, and date at the start of the chromatogram strip chart.

12.2.2 Every chromatogram/every sample/standard

                   Sample number or standard
                   Sample volume injected

-------
                                                     EIA-FLDGRAB4.WPD
                                                     VOCs in Air Samples
                                                     02/12/02
                                                     Page 13 of 17
                    Instrument gain or attenuation setting


13 0 Quality Control:
13.1   A blank and a one point standard is used for instrument=s calibration. Initially run 10
      ug/L standard to determine retention times and response factors of instrument.  Repeat a
      second 10 ug/L standard to check the reproducibility. Acceptance criteria: within +
      15% difference from the first standard.

13.2   Blanks are analyzed at the initial calibration and periodically to be sure of no carry over
      from previous injections. Technical judgment is used to determine frequency.
      Acceptance criteria: No target compound peaks greater than one-half the reporting
      level.

13.3   A second source standard containing some compounds of  interest is analyzed daily to
      verify calibration standard. Acceptance criteria: within + 20% agreement of true
      value.

13.4   A standard is run at least every 10 samples and at the end of the sample batch to update
      the instrument=s calibration due to changes from temperature fluctuations with respect to
      retention times and response factors. Acceptance criteria: + 20%D agreement with the
      previous calibration.

13.5   Analyze upwind samples to determine background concentrations during outdoor
      ambient air sample events and report results.

13.6   Run field and laboratory duplicates when possible (i.e., soil gas analysis and passive
      vapor sample analysis)  The acceptance criterion  is agreement within + 20% RPD
      between the two values.

13.7    When possible (i.e., soil gas, ambient air), GC/MS confirmation of 10% of the field
      samples analyzed should be performed. This is done, dependent upon the project data
      quality objective. Summa canisters are used for collecting confirmation samples for
      GC/MS confirmation.
14.0  References:

-------
                                                EIA-FLDGRAB4.WPD
                                                VOCs  in Air Samples
                                                02/12/02
                                                Page  14 of 17
14.1   Interim Final Guidance for the Quality Assurance/Quality Control Guidance for Removal
      Activities, April 1990.

-------
                                                EIA-FLDGRAB4.WPD
                                                VOCs  in Air Samples
                                                02/12/02
                                                Page  15 of  17
Quality Control Table
QC Item
Initial
Calibration
Blank
Second
Source Std
Continue
Calibration
Upwind
Samples
Field
Duplicate
Lab.
Duplicate
Frequency
Daily, before
samples
Daily, every
batch
Daily, every
batch
Every 10
samples and
at the end
Option, if the
situation
warranted
Option,
depends on
DQOs
Option,
depends on
DQOs
Acceptance
Criteria
< 15%D from the
first std1
< 1/2 RL1
< 20%D, from
the true value l
< 20%D, from
the previous std 1
None.
Report results l
<20%RPD1
< 20% RPD1
Corrective
Action
Inject another std, check system
Repeat blank injection, prepare a new
blank, check system, increase RLs
depending on the DQOs
Inject another std, repeat initial
calibration, check system
Inject another std, repeat initial
calibration, check system

Repeat injection, run another duplicate
Repeat injection, run another duplicate
1_
 = Acceptance criteria defined based on technical judgment

-------
                                             EIA-FLDGRAB4.WPD
                                             VOCs in Air Samples
                                             02/12/02
                                             Page 16 of  17
                             Figure 1
                  Volatile Organic Screening Method
                 Target Compound Chromatogram (PID)
1.      1,1 -Di chl oroethene
2.      t-l,2-Dichl oroethene
3.      c-l,2-Dichl oroethene
4.      Benzene
5.      Trichloroethene
6.      Toluene
7.     Tetrachloroethene
8.     Chlorobenzene
9.     Ethyl Benzene
10.    m/p-Xylenes
11.    o-Xylene
    -a
     3
    "a.
     10
     aj
     Q_
    JO
     tu
    cc
                             Relative ElutionTime
Instrument:   Shimadzu gas chromatography 14A
Detector:     Photoionization Detector (PID)
Column:      DBFS 624, 30 m, 0.53 micron
Temperature:  60°C
Carrier Gas:   Zero grade nitrogen
Flow rate:    30-60 cc/min
Chart speed:   1 cm/min

-------
                                            EIA-FLDGRAB4.WPD
                                            VOCs in  Air  Samples
                                            02/12/02
                                            Page 17  of 17
                               Figure 2
                   Volatile Organic Screening Method
                   Target Compound Chromatogram (BCD)
1.     1,1 -Dichloroethene
2.     Trichloroethene
3.     Tetrachloroethene
                              Relative ElutionTime
Instrument:    Shimadzu gas chromatography 14A
Detector:     Electron Capture Detector (BCD)
Column:      DBFS 624, 30 m, 0.53 micron
Temperature:  60°C
Carrier Gas:   Zero grade nitrogen
Flow rate:     30-60 cc/min
Chart speed:   1 cm/min

-------

-------
Prepared by:
   USGS Enterprise Publishing Network
   Raleigh Publishing Service Center
   3916 Sunset Ridge Road
   Raleigh, NC 27607

For additional information regarding this publication, contact:
   Don A. Vroblesky
   USGS South Carolina Water Science Center
   720Gracern Road, Suite  129
   Columbia, SC 29210
   phone: 1-803-750-6100
   email: vroblesk@usgs.gov

Or visit the South Carolina Water Science Center website at:
   http://sc, water, usgs,go¥

This publication is available online at:
   http://pubs.water.usgs.gov/sir2m8-5088

-------
 CD
 00
 00
 CD
 CD

 o
 **

 CD
 O
 o_

 CD"
 a
 o'
 —
•s
 CD
 CD

 O
 O
 5
 o'
 3

 a
 CO
 9)
 o
 CD
 <
 O
 o
 oo
 o
 CD
 00
to
 Q3
 rt-

 o'
 CD
-a
 o
 s
 NO
 CD
 CD
 cn
 CD
 oo
 oo

-------