&EPA
United States
Environmental Protection
Agency
Department of Defense
ENVIRONMENTAL SECURITY
TECHNOLOGY CERTIFICATION PROGRAM
       Phytoremediation of
       Groundwater at
       Air Force Plant 4
       Cars well, Texas

       Innovative Technology
       Evaluation Report
               SUPERFUND INNOVATIVE
               TECHNOLOGY EVALUATION

-------
                                     EPA/540/R-03/506
                                     September 2003
Phytoremediation of Groundwater
         at Air Force Plant 4

           Cars well, Texas
  Innovative Technology Evaluation Report
          National Risk Management Research Laboratory
             Office of Research and Development
            U.S. Environmental Protection Agency
                Cincinnati, Ohio 45268

-------
                                      Notice
The information in this document has been found wholly or in part by the U.S. Environmental
Protection Agency (EPA) in partial fulfillment of Contract No. 68-CO-0048 and Contract No. 68-C5-
0036 to Science Applications International Corporation. It has been subject to the Agency's peer and
administrative review, and it has been approved for publication as an EPA document. Mention of trade
names of commercial products does not constitute an endorsement or recommendation for use.

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

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

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

-------
                                      Abstract
A demonstration of a Phytoremediation  Groundwater Treatment system was conducted at the
Carswell Naval Air Sation (MAS) Golf Club in Fort Worth, Texas to investigate the ability of purposely
planted eastern cottonwood trees, Populus deltoides, to help remediate shallow TCE-contaminated
groundwater in a subhumid climate. Specifically, the study was undertaken to determine the potential
for a planted system to hydraulically control the migration of contaminated groundwater, as well as
biologically enhance  the subsurface environment to optimize  in-situ reductive  dechlorination of
chlorinated ethenes present (trichloroethene and cis-1,2-dichloroethene) in the shallow aquifersystem
beneath a portion of the golf course. Populus deltoides, like other phreatophytes, have long been
recognized as having the ability to tap  into the saturated zone to extract water for metabolic
processes. Based upon this characteristic the species was considered well suited for applications
where shallow aquifers are contaminated with biodegradable organic contaminants. A planted system
of cottonwood  trees is believed to effectuate two processes that aid and accelerate contaminant
attenuation.  First, transpiration of groundwaterthrough the trees is believed to be able to modify and
hopefully control the hydraulic groundwater gradient. This can minimize the rate and  magnitude of
migrating contaminants downgradient of the tree plantation.  Secondly, the establishment of the root
biomass, or rhizosphere, promotes microbial activity and may enhance biodegradative processes in
the subsurface. To assess the performance of the system,  hydrologic and geochemical data were
collected over a three-year period (August 1996 through September 1998). In addition to investigating
changes in groundwater hydrology and chemistry, the trees were studied to determine  important
physiological processes such as rates of water usage, translocation and volatilization of these volatile
organic compounds, and biological transformations  of chlorinated ethenes within the  plant organs.

The demonstration site is situated about one mile  from the southern area of the main assembly
building  at Air Force Plant 4 (Plant 4) at the Carswell MAS. The assembly building is the primary
suspected source of TCE at the demonstration site. The evaluation of this technology application was
a joint effort between the U.S. Air Force (USAF), the U.S. Geological Survey, the U.S. Forest Service,
the  Department of Defense's (DoD's)  Environmental Security  Technology Certification  Program
(ESTCP), and the U.S. EPA's SITE program.
                                           IV

-------
                                     Contents
Notice	ii
Foreword  	iii
Abstract  	  iv
Tables	vii
Figures	viii
Acronyms, Abbreviations and Symbols  	  ix
Acknowledgments  	  xi

Section 1      Introduction	1
              1.1      Background	1
              1.2     Brief Description of SITE Program and Reports	3
              1.3     The SITE Demonstration Program 	3
              1.4     Purpose of the Innovative Technology Evaluation Report (ITER)	4
              1.5     Technology Description	4
              1.6     Key Contacts	6

Section 2      Technology Applications Analysis  	8
              2.1      Key Features	8
              2.2     Operability of the Technology	8
              2.3     Applicable Wastes	11
              2.4     Availability and Transportability of the Equipment	11
              2.5     Materials Handling Requirements	11
              2.6     Site Support Requirements	12
              2.7     Range of Suitable Site Characteristics  	12
              2.8     Limitations of the Technology	12
              2.9     Technology Performance Versus ARARS	13
                      2.9.1   Comprehensive Environmental Response, Compensation,
                            and Liability Act (CERCLA)  	14
                      2.9.2   Resource Conservation and  Recovery Act (RCRA)	14
                      2.9.3   Clean Air Act (CAA)	17
                      2.9.4   Clean Water Act (CWA)	17
                      2.9.5   Safe Drinking Water Act (SDWA)	17
                      2.9.6   Toxic Substances  Control Act (TSCA)	18
                      2.9.7   Occupational Safety and Health Administration
                             (OSHA) Requirements  	18
                      2.9.8   State Requirements	18

Section 3      Economic Analysis	19
              3.1      Introduction  	19
              3.2     Conclusions	22
              3.3     Factors Affecting Estimated Cost 	23
              3.4     Issues and Assumptions	23

-------
                      3.4.1   Site Size and Characteristics	23
                      3.4.2   System Design and Performance Factors	24
                      3.4.3   System Operating Requirements	24
                      3.4.4   Financial Assumptions	25
               3.5     Results of the Economic Analysis	26
                      3.5.1   Site Preparation  	26
                      3.5.2   Permitting and Regulatory Requirements 	28
                      3.5.3   Capital Equipment 	28
                      3.5.4   Startup and Fixed Costs	29
                      3.5.5   Consumable and Supplies	29
                      3.5.6   Labor  	29
                      3.5.7   Utilities 	30
                      3.5.8   Effluent Treatment and  Disposal  	30
                      3.5.9   Residuals & Waste Shipping, Handling and Storage	30
                      3.5.10  Analytical Services	30
                      3.5.11  Maintenance and Modifications  	31
                      3.5.12  Demobilization 	31

Section 4       Treatment Effectiveness	32
               4.1     Background	32
               4.2     Detailed Description of the Short Rotation Woody Crop
                      Groundwater Treatment System	33
                      4.2.1   Site Selection	33
                      4.2.2   Site Characterization 	33
                      4.2.3   Size and Configuration  of the Tree Plantations 	33
                      4.2.4   Planting and Installation of the Irrigation  System	34
                      4.2.5   Irrigation	35
                      4.2.6   Monitoring	35
               4.3     Project Objectives	36
                      4.3.1   Primary Project Objective	36
                      4.3.2   Secondary Project Objectives 	37
               4.4     Performance Data	39
                      4.4.1   Summary of Results - Primary Objective	39
                      4.4.2   Summary of Results - Secondary Objectives	39
               4.5     Discussion	55

Section 5       Other Technology Requirements	58
               5.1     Environmental Regulation Requirements  	58
               5.2     Personnel Issues	58
               5.3     Community Acceptance  	58

Section 6       Technology Status  	59
               6.1     Previous Experience	59
               6.2     Scaling Capabilities	59

               References	60
Appendix A     Data Used to Evaluate Primary Project Objective  	A-1
Appendix B     Air Force Recommendations  	B-1
                                           VI

-------
                                      Tables
2-1     Federal and State Applicable and Relevant and Appropriate Requirements
       (ARARS) for Tree Based Phytoremediation System	15
3-1     Estimated Full-Scale Costs for a 200,000 Square Foot Hypothetical
       Phytoremediation Model Site  	20
4-1     Summary of Primary Objective Results	40
4-2     Average Concentration of Detectable Volatile Compounds In Plant Tissue	49
4-3     Pseudo First Order Disappearance Rate Constants for the Plant-Leaf
       Mediated Transformation of TCE 	51
4-4     Average TCE and DCE Concentrations in Monitoring Wells	51
4-5     TCE to Cis-1,2-DCE Ratio	52
4-6     Selected  chemical data from wells used to define terminal electron accepting
       processes (TEAP) at the demonstration site	53
4-7     Results of microbial population survey  	54
                                          VII

-------
                                     Figures
1-1     SRWCGT Process Mechanisms	5
3-1     Layout of 200,000 Square Foot Hypothetical Model Site for Cost Analysis  	25
4-1     Phytoremediation Groundwater Treatment System site layout	35
4-2     Wells Used to Monitor For Changes in the Volumetric Flux of Groundwater
       Across the Downgradient End of the Short Rotation Woody Crop Groundwater
       Treatment System	37
4-3     Drawdown at the Water Table That can be Attributed to the Trees  	40
4-4     Trunk Diameter Over Time 	42
4-5     Tree Height Over Time 	42
4-6     Canopy diameter Over Time	43
4-7     Caliper-Tree Plantation at the Time of Planting, April  1996 	43
4-8     Caliper-Tree Plantation at the End of Third Growing Season
       October 1998	43
4-9     Root Counts by Depth	44
4-10   Variation in the Mean Hourly SapFlow Rate (a) Expressed on a Per Tree
       Basis and (b) expressed on a Per Unit Basal Area Basis.  Data are Sample
       Period Means for all Months (p<0.05) Differences Between Whips and
       Caliper Trees are Denoted by *. Vertical Lines on all Bars Represent
       Standard errors	45
4-11   Minimum  Predicted Drawdown at the Water Table for Closed-Canopy
       Conditions (year 12 and beyond)  	47
4-12   Maximum Predicted Drawdown at the Water Table for Closed-Canopy
       Conditions (year 12 and beyond)  	47
4-13   Simulated Groundwater Budget (A) Prior to Treatment, (B) Peak of the
       Third Growing Season (1998), © Peak of the Growing Season Once
       Closed Canopy has been Achieved (year 12 and beyond)-Minimum
       Predicted Transpiration, and (D) Peak of the Growing Season Once
       Closed Canopy has been Achieved (year 12 and beyond)-Maximum
       Predicted Transpiration	48
                                         VIM

-------
                Acronyms, Abbreviations and Symbols
A             Cross-Sectional Area of Aquifer
AACE         American Association of Cost Engineers
AFB          Air Force Base
AFCEE        Air Force Center for Environmental Excellence
AQCR         Air Quality Control Regions
AQMD         Air Quality Management District
ARARs        Applicable or Relevant and Appropriate Requirements
ASC/ENV      Aeronautical Systems Center Acquisition, Environmental, Safety and Health
              Division
ATTIC         Alternative Treatment Technology Information Center
BGS          Below Ground Surface
BFDP         Biofuel Feedstock Development Program
BTEX         Benzene, Toluene, Ethylbenzene, and Xylenes
CAA          Clean Air Act
CERCLA      Comprehensive Environmental Response Compensation and Liability Act
CERI          Center for Environmental Research Information
CFR          Code of Federal  Regulations
CGC          Carswell Golf Club
cm/s          centimeters/second
cm            Centimeter
CWA          Clean Water Act
d             day
DCE          Dichloroethene
DO           Dissolved Oxygen
DoD          Department of Defense
DoE          Department of Energy
ESTCP        Environmental Security Technology Certification Program
ft             feet
g             gram
gptpd          Gallons per Tree per Day
ha            Hectare
hr            Hour
I              Hydraulic Gradient
IRP           Installation Restoration Program
ITER          Innovative Technology Evaluation Report
K             Hydraulic Conductivity
Kg            Kilogram
m             Meter
m/d           meters/day
MCLGs        Maximum Contaminant Level Goals
MCLs         Maximum Contaminant Levels
mg/L          milligrams per liter
                                        IX

-------
           Acronyms, Abbreviations and Symbols(Cont'd)
mm           Millimeter
MPN          Most Probable Number
NAAQS        National Ambient Air Quality Standards
NAS          Naval Air Station
NCR          National Oil and Hazardous Substances Pollution Contingency Plan
NPDES        National Pollutant Discharge Elimination System
NPL          National Priorities List
NRMRL        National Risk Management Research Laboratory
O&M          Operation & Maintenance
ORD          Office of Research and Development
ORNL         Oak Ridge National Laboratory
OSHA         Occupational Safety and Health Administration
OSWER       Office of Solid Waste and Emergency Response
PA            Preliminary Assessment
PCBs         Polychlorinated Biphenyls
POTW        Publicly Owned Treatment Works
PPE          Personal Protective Equipment
Q             Volumetric Flux
QA/QC        Quality Assurance/Quality Control
RCRA         Resource Conservation and Recovery Act
RI/FS         Remedial Investigation/Feasibility Study
ROD          Record of Decision
SAIC          Science Applications International Corporation
SARA         Superfund Amendments and Reauthorization Act
SDWA        Safe Drinking Water Act
SITE          Superfund Innovative Technology Evaluation
SWDA        Solid Waste Disposal Act
TCE          Trichloroethene
TEAP         Terminal Electron-Accepting Process
TER          Technology Evaluation Report
TOC          Total Organic Carbon
TSCA         Toxic Substances Control Act
TSD          Treatment Storage and Disposal
USAGE        United States Army Corps of Engineers
USAF         United States Air Force
USDA         United States Department of Agriculture
USEPA        United States Environmental Protection Agency
USGS         United States Geological Survey
VC            Vinyl Chloride
VISITT        Vendor Information System for Innovative Treatment Technologies
VOCs         Volatile Organic Compounds

-------
                              Acknowledgments
This report would like to  acknowledge the  financial support of the Department of Defense's
Environmental Security Office (ESTCP), The United States Environmental Protection Agency, and
the United States Air Force Aeronautical Systems Center Engineering Directorate Environmental
Safety and Health Division at Wright-Patterson Air Force Base.  The authors of this  report  also
acknowledge the technical contributions of urban forester Larry Schaapveld of the Texas State Forest
Service  who was instrumental in making this project a success.
                                         XI

-------
                                              SECTION 1
                                           INTRODUCTION
This section provides a discussion on the fate of fuel and
solvent contaminants in groundwater systems, the limits of
intrinsic remedial mechanisms,  biodegradation of fuel
products and chlorinated compounds, the three chlorinated
solvent  plume behavior types and their  implications on
reductive dechlorination, background information on the
study site and the field test, background information about
the Superfund Innovative Technology Evaluation (SITE)
Program,  the Environmental  Technology  Certification
Program  (ESTCP),  the   purpose  of  this  Innovative
Technology  Evaluation   Report  (ITER),   and  the
Phytoremediation of groundwater process. For additional
information about the SITE Program, this technology, and
the demonstration site, key contacts are listed at the end of
this section.

1.1     Background

Fuels and  chlorinated solvents  are commonly found  in
groundwater. In the last twenty years the persistence and
behavior of fuels  and chlorinated solvents in ground water
have  been  the  subject  of intense investigation  and
vociferous debate. Both fuels and chlorinated solvents can
naturally attenuate if the appropriate conditions exist in the
subsurface.  Natural  attenuation  in groundwater systems
results  from  the   integration   of several subsurface
mechanisms that are classified as either destructive or non
destructive (Wiedemeier, 1996). Biodegradation is the most
important  destructive   mechanism.   Nondestructive
mechanisms include sorption,  dispersion, dilution from
recharge,  and volatilization (Wiedemeier,  1996).  The
behavior of fuels and chlorinated solvents in the subsurface
are different from one another depending on the availability
of  electron  acceptors  and  electron  doners  in  the
subsurface:  The  most significant difference between fuel
products and  chlorinated  solvents  is that  usually fuel
plumes don't move and chlorinated solvent plumes do.

The biodegradation of fuel products is limited by electron
acceptor availability (Wiedemeier, 1996). Fortunately there
is  an adequate  supply  of electron acceptors in  most
hydrologic  settings.  Accordingly,  most fuels plumes
degrade fasterthan they move (Chappelle, 2000). The long
term behavior of chlorinated solvents is more difficult to
predict than fuel plumes. The biodegradation of chlorinated
solvents begins in the saturated subsurface where native
or anthropogenic carbon is used as an election donor, and
dissolved oxygen is  utilized first for the prime electron
acceptor (Wiedemeier,  1996). Once dissolved oxygen is
depleted,  anaerobic microorganisms  most often  use
available electron acceptors in the following order: nitrate,
Fe(lll) hydroxide, sulfate, and carbon dioxide (Chappelle,
2000).  In the absence  of nitrate and dissolved oxygen,
chlorinated solvents compete with otherelectron acceptors
and donors especially sulfate and carbon dioxide. The most
important anaerobic process forthe natural biodegradation
of chlorinated solvents is reductive dechlorination. When a
chlorinated  solvent is used as an electron acceptor, a
chlorine atom is  removed and  replaced with a hydrogen
atom. Electron donors include fuel hydrocarbons,  landfill
leachate or natural organic carbon. If the subsurface is
depleted of electron donors before chlorinated solvents are
removed, microbial  reductive  dechlorination will  cease
(Wiedemeier, 1996).  Plumes of chlorinated solvents can
naturally attenuate but almost 80% of the time they do not
due to the lack of electron donors (Chappelle, 2000).

Chlorinated solvent plumes exhibit three types of behavior
depending  on the amount of solvent, the amount  of
biologically  available organic  carbon in the aquifer, the
distribution and concentration of natural electron acceptors
and types of electron acceptors (Wiedemeier, 1996). Type
1  behavior  occurs  when the   primary  substrate  is
anthropogenic carbon (e.g.  benzene, toluene, xylene,  or
landfill  leachate).  The  microbial  degradation  of  this
anthropogenic carbon drives reductive dechlorination. Type
2 behavior prevails in areas that have high concentrations
of biologically available native organic carbon. Type 3
behavior dominates in areas that are lacking an adequate
amount of native  and  or anthropogenic  carbon  and
concentrations of dissolved  oxygen that are greater than
1.0 mg/L. Reductive dechlorination does not occur under
Type 3 conditions. Type 3 conditions commonly prevail at
Department of Defense  (DoD)  sites resulting in very large

-------
unattenuated plumes.

The  TCE  groundwater plume beneath a  portion  of the
Carswell Golf Club  near Fort Worth, Texas  is a prime
example of a site characterized by Type 3  behavior. This
site was chosen to field test an innovative phytoremediation
process also referred to as the Short Rotation Woody Crop
Groundwater  Treatment   (SRWCGT)   system.  The
SRWCGT system was tested to determine the contribution
of higher  plants in  (1) accelerating  and  enhancing the
bioremediation  and  phytodegradation of  chlorinated
ethenes from a shallow aquifer; and (2)  mitigating the
migration  of the  contaminant plume  through  gradient
control.  The evaluation of this technology application was
a joint effort between the U.S. Air Force (USAF), the U.S.
Geological Survey,  the U.S. Forest  Service,  the  DoD's
Environmental Security Technology Certification Program
(ESTCP),  and the  U.S. EPA's SITE program.

The  system is  an  application  of  phytoremediation
technology designed and implemented by the USAF under
the DoD ESTCP. The ESTCP is a corporate DoD program
that  promotes  innovative,  cost-effective  environmental
technologies through demonstration and validation at DoD
sites. ESTCP's goal  is  to demonstrate  and validate
promising  innovative technologies that target the  DoD's
most  urgent   environmental  needs  through   their
implementation and commercialization. These technologies
provide  a  return on investment through cost savings and
improved  efficiency. ESTCP's strategy is  to  select lab-
proven  technologies  with  broad  DoD   and  market
application.  These technologies are aggressively moved
to the field for rigorous trials that  document their costs,
performance, and  market potential.

The demonstration investigated the use of a phreatophytic
tree, Populus deltoides, as a rapidly growing plant species
that  may  accelerate  natural processes  that  promote
contaminant degradation as  well as  control hydraulic
gradient. Populus deltoides, like any tree or any other living
organism for that matter,  is a complex structure derived
ultimately from enzyme-catalyzed reactions regulated by its
genes (Dickman,  1983). The  study  of the derivative of
these biochemical reactions i.e. the functioning of the tree
or any of its parts as an organized entity is tree physiology.
(Dickmann, 1983)  There are several different approaches
to planting trees currently available. These range from
deep auguring  individual  poles to  the capillary  fringe
employing proprietary planting techniques to employing
short rotation woody  crop techniques. These planting
approaches have their indications, contradictions and their
various  champions  within  the phytoremediation  arena.
Short  rotation  woody/energy  crop  technology was
developed   by  the   Department  of Energy's  Biofuel
Feedstock Development Program  (BFDP)  at Oak Ridge
National Laboratory  (ORNL). The mission of the BFDP is
to develop and demonstrate environmentally acceptable
crops and cropping systems for producing large quantities
of low cost high quality biomass feedstocks The research
strategy  of  the BFDP  is designed to maximize the
economic returns, reduce environmental impacts  and
establish sustainable biomass systems that optimize per
unit area productivity for members of the Populus and Salix
genera over a substantially large portion of the U.S. To
date, the BFDP has screened more than 125 tree and non-
woody species and selected a number of model species for
development as energy crops. Former President William
Clinton issued an executive order calling for increased use
of trees and  crops as environmentally friendly sources of
energy.

This demonstration investigated the use of a phreatophytic
tree  planted  for use  in  phytoremediation  of TCE-
contaminated groundwater. Populus deltoides, commonly
known as the cottonwood, is a  rapidly growing tree that
may  accelerate   natural  processes   that   promote
contaminant degradation as  well as control hydraulic
gradient. Populus deltoides, like other phreatophytes, has
the ability to tap into the saturated zone to extract water for
metabolic processes. Therefore, this species is well suited
for applications where shallow aquifers are contaminated
with  biodegradable organic contaminants.  The  planted
system is believed to effectuate two processes that aid and
accelerate contaminant attenuation.  First, transpiration of
groundwater through the trees is believed to be able to
modify and hopefully control the hydraulic groundwater
gradient. This  can minimize the rate and magnitude of
migrating contaminants downgradient of the tree plantation.
Secondly, the  establishment  of the root  biomass, or
rhizosphere, promotes microbial activity and may enhance
biodegradative processes in the subsurface. A technology
demonstration was designed to determine the effectiveness
of the system to control  hydraulic gradient and enhance
biodegradative processes. As previously mentioned, the
demonstration took place at the Carswell Golf Club (CGC)
at the Naval  Air Station  (NAS) Fort Worth, which is
adjacent to Air Force Plant 4. Specifically, the site is on the
north side of the CGC west of the 8th green about 1 mile
from the southern area of the main  assembly building at
Plant 4. The assembly building is the suspected source of
TCE  at the  demonstration  site.  In  April  of  1996
approximately 660 trees  were  planted in two plots at the
site.

Plant 4 was  constructed  in 1942 and  currently produces
F-16 aircraft, radar units, and various aircraft and missile
components.     General  Dynamics   operated  the
manufacturing facility from 1953 to 1994 when Lockheed
took over operations. Since 1953, Plant 4 has produced
B-36, B-58, and F-111 aircraft.

Historically, the manufacturing  processes at Plant 4 have
generated an estimated 5,500 to 6,000 tons of waste per

-------
year, including waste solvents, oils, fuels, paint residues,
and miscellaneous spent chemicals. Throughout most of
Plant 4's history, the waste oil, solvents, and fuels were
disposed of at onsite landfills orwere burned in fire training
exercises.

Plant  4 is on the  National  Priorities  List and is  being
remediated  in  accordance  with  the Comprehensive
Environmental Response, Compensation, and Liability Act
(CERCLA) as amended  by the Superfund Amendments
and Reauthorization Act (SARA).
The SITE Program's primary purpose is to maximize the
use of alternatives in cleaning hazardous waste sites by
encouraging the development and demonstration of new,
innovative treatment and  monitoring  technologies.   It
consists of three major elements discussed below:

   the Demonstration Program,
   the Monitoring and Measuring Technologies Program,
   and
   the Technology Transfer Program.
Potential contamination at Plant 4 was first noted by a
private citizen in September 1982. TCE may have leaked
from the degreasing tanks in the assembly building at Plant
4 and entered the underlying aquifer over the course of
decades. An Installation Restoration Program (IRP) was
initiated in 1984 with a Phase I Records Search by CH2M
Hill (CH2M Hill 1984). The U.S. Army Corps of Engineers
(USAGE) was retained in June of 1985 to further delineate
groundwater conditions in the  East Parking  Lot area of
Plant 4. The USAGE constructed six monitoring wells (U.S.
Army  Corps of Engineers  1986). Ongoing groundwater
sampling in the East Parking  Lot area  of  Plant 4 has
continued for the purpose of monitoring this plume.

The TCE plume appears to be migrating in an easterly to
southeasterly direction. It appears to have migrated under
the East Parking Lot and towards the MAS Fort Worth. The
plume fingers toward the east with the major branch of the
plume following a paleochannel under the flight lines to the
south of the phytoremediation demonstration site, where it
has undergone remediation with a pump and treat system.
Another  branch  of the  plume  appears  to  follow a
paleochannel to the north of the demonstration site.

Historic activities otherthan the operations at the assembly
building, however, may have contributed to the TCE plume
at the phytoremediation site. Several former landfills have
been identified near the CGC where drums of TCE have
been found . The former landfills appear to be upgradient
and crossgradient from the demonstration site; however,
insufficient groundwater  level  data and aquifer  testing
reports are available to determine whether these former
landfills are actually sources.


1.2     Brief Description of the  SITE Program
       and Reports

The SITE Program is a formal program established by
EPA's Office of Solid Waste and  Emergency Response
(OSWER) and Office of Research and Development (ORD)
in  response   to   the  Superfund  Amendments  and
Reauthorization Act of 1986 (SARA). The SITE Program
promotes the development, demonstration, and use of new
or innovative  technologies to clean up Superfund sites
across the country.
The objective of the Demonstration Program is to develop
reliable  performance  and  cost  data  on  innovative
technologies so  that  potential  users can assess  the
technology's site-specific applicability.    Technologies
evaluated  are either available commercially  or close to
being available  for full-scale remediation  of Superfund
sites.  SITE demonstrations usually are conducted on
hazardous  waste sites  under  conditions that closely
simulate full-scale remediation conditions, thus assuring
the usefulness and reliability of information collected. Data
collected are used to assess: (1) the performance of the
technology,  (2)  the potential  need for  pre- and  post-
treatment  processing of wastes, (3) potential operating
problems,   and   (4)  the  approximate   costs.    The
demonstrations also provide opportunities to evaluate the
long-term risks, capital and O&M costs associated with full-
scale application of the subject technology, and limitations
of the technology.

Existing technologies  and  new technologies  and  test
procedures  that  improve  field  monitoring  and  site
characterizations are identified  in  the  Monitoring  and
Measurement Technologies  Program.  New technologies
that provide faster, more cost-effective contamination and
site assessment data are supported by this program.  The
Monitoring and Measurement Technologies Program also
formulates   the  protocols  and  standard   operating
procedures for demonstrating methods and equipment.

The Technology Transfer Program disseminates technical
information  on   innovative    technologies  in   the
Demonstration,  and the Monitoring  and Measurement
Technologies Programs through various activities.  These
activities increase the awareness and promote the use of
innovative technologies for assessment and remediation at
Superfund sites. The goal of technology transfer activities
is to develop interactive communication among individuals
requiring up-to-date technical information.


1.3    The SITE Demonstration Program

Technologies are selected for the SITE Demonstration
Program through annual requests for proposals. ORD staff
review the proposals to determine which technologies show

-------
the most promise for use at Superfund sites. Technologies
chosen  must be  at the pilot- or full-scale stage, must be
innovative, and must have some advantage over existing
technologies.  Mobile and in-situ  technologies are of
particular interest.

Once  EPA  has  accepted  a  proposal,  cooperative
agreements  between EPA  and the developer establish
responsibilities for conducting the  demonstrations  and
evaluating the technology. The developer is responsible
for demonstrating the technology at the selected site and
is expected to pay any costs for transport, operations, and
removal of the equipment.  EPA is responsible for project
planning, sampling and analysis, quality assurance and
quality  control,   preparing   reports,  disseminating
information,  and transporting  and disposing of treated
waste materials.

The results of this evaluation of the SRWCGT process are
published in this Innovative Technology Evaluation Report.
The  ITER is intended  for  use by remedial managers
making  a  detailed evaluation of the  technology for a
specific site and waste.


1.4     Purpose of the Innovative Technology
        Evaluation Report (ITER)

This ITER provides information on the SRWCGT process
and  includes  a  comprehensive  description  of  the
demonstration and its results. The ITER is intended for use
by  EPA remedial project managers,  EPA  on-scene
coordinators, contractors, and other decision makers in
implementing specific remedial actions.  The ITER is
designed to aid  decision  makers in further evaluating
specific technologies when considering applicable options
for particular cleanup operations. This report represents a
critical step in the development and commercialization of a
treatment technology.

To   encourage  the   general  use  of  demonstrated
technologies, EPA provides  information regarding the
applicability  of each  technology to specific sites  and
wastes.  The  ITER  includes  information on  cost  and
performance,  particularly  as  evaluated  during  the
demonstration.    It   also  discusses  advantages,
disadvantages, and limitations of the technology.

Each SITE demonstration evaluates the performance of a
technology in  treating  a  specific waste.   The  waste
characteristics  of  other  sites  may  differ  from  the
characteristics  of  the  treated waste.   Therefore,  a
successful field demonstration of a technology at one site
does not necessarily  ensure that it will be applicable at
other sites.  Data from the field  demonstration may require
extrapolation for estimating the operating  ranges in which
the technology will  perform satisfactorily.  Only limited
conclusions   can   be   drawn  from   a  single  field
demonstration.
1.5 Technology Description

The SRWCGT process is a phytoremediation technology
that relies on the use of higher plants to augment in situ
biodegradative  reactions as well as  control  hydraulic
gradient to minimize the transport of contaminants.  The
system evaluated at the Carswell Golf Club was designed
to intercept and treat a TCE plume using  strategically
placed plantations  of the Eastern Cottonwood  trees
(Populus deltoides). However, the technology is generally
applicable to most biodegradable organic compounds.
Figure 1-1  depicts the remediation  mechanisms of the
process.

Phytoremediation has  received heightened attention as a
mechanism to augment and accelerate natural degradative
processes. Phytoremediation  is the use of higher plants for
remediating anthropogenically contaminated environments.
Phytoremediation  relies on several plant  physiological
processes to treat contaminants in situ.  These generally
fall into the following categories:

1.     Degradation or the facilitation of degradation of
       organic contaminants  alone  or  via  microbial
       associations within the plant rhizosphere;
2.     Hyperaccumulation or sequestering  of inorganic
       contaminants within plant parts;
3.     Binding of contaminants within plant organs;
4.     Volatilization  of organic  contaminants from the
       rhizosphere and transpiration into the atmosphere.
Plants have evolved biological detoxification mechanisms
over several hundred million  years. Previous work has
indicated that plants such as poplars  and  corn  can
metabolize  TCE to trichloroethanol, trichloroacetic acid,
dichloroacetic  acid, and  carbon  dioxide  (Schnoor  and
Kurimski 1995).   Schnoor (1995b)  suggests  that  a
significant portion of TCE taken up by  such plants  is
transformed and/or bound irreversibly to the biomass. Mass
transfer limitations of organic compounds in soil due to low
solubility and high soil adsorption, however, can limit plant
uptake of many compounds. Highly lipophilic compounds
such as polychlorinated biphenyls (PCBs) are generally so
strongly bound to soil that they do not become bioavailable
to either  plants  or  microbes.   Moderately lipophilic
substances, such as TCE, can move through the soil to the
position of the  rhizosphere  and  are the most  likely
candidates for phytoremediation.

In general, phytoremediation has the potential to mitigate
groundwater contamination in two ways: (1) withdrawal  of
groundwater from an aquifer to minimize  migration  of a

-------
     4. Volatilization
                                                                       3. Enzymatic
                                                                          Degradation /
                                                                          Mineralization
                                                                          Within Vegetation
   1. Hydraulic Control /
      Influence
         2.  Ih-Situ Biodegradation
  Figure 1-1.  Phytoremediation Process Mechanisms
contaminated  plume and to  possibly flush the aquifer,
which  is referred  to  as hydraulic control; and  (2)
remediation of the contaminated water.  In simple terms,
plants are biologically based solar-powered pump and treat
systems.

The consumptive use of water by phreatophytes, deep
rooted plants that can obtain water from a subsurface water
source, has historically been considered a liability in some
arid and semiarid environments. The consumptive use of
water  by vegetation,  however, is  now  being viewed
differently  because of its potential for remediation  of
contaminated groundwater. Instead of employing energy,
capital,  and  maintenance-intensive pump  and  treat
systems, it may be possible to exploit the natural ability of
plants to transpire water. On a hot sunny day the volume
of water loss may exceed the total water content of the
plant.  The success and even the survival of land plants
depend on adequate water moving upward from the roots
to replace that lost from the canopy by transpiration. Water
flow is driven by the difference in free energy of water in
the soil and dry air. Accordingly, plants can pump large

-------
amounts of water soluble contaminants by means of the
transpiration stream.

The amount of transpiration is a function of plant density,
leaf  area index,  radiant  solar energy  flux,  depth to
groundwater, temperature, relative  humidity and  wind
speed (Nichols 1994).  Roots function as water sensors
and grow through the soil following water potentials. When
water becomes limited phreatophytes are more resistant to
wilting than shallow rooted plants. Trees have the most
massive root system of all plants and their root systems are
capable of penetrating several meters below the surface
(Stomp 1993). Examples of phreatophytic trees are willows,
cottonwoods (poplars),  salt cedar and mesquite (Fetter
1988).

Plant roots can increase the biological activity in the soil
adjacent to the roots; this  region in the soil is called the
rhizosphere. The rhizosphere consists of both biotic and
abiotic  parts.    Releases  from   plant roots   into the
rhizosphere may be inorganic or organic.  The carbon in
root exudates is from carbon dioxide fixed in the production
of carbohydrates.  Anywhere from  1 to 40 percent of the
net photosynthate may  be released from the roots to the
soil.  Organic rhizosphere exudates take several different
forms:  simple  sugars, amino acids, organic  acids,
phenolics, and polysaccharides (Shann 1995). The in situ
function of these exudates has not  been fully determined.
Tests show they can act as nutrients, as antibiotics, and
chemoattractants.    Plant  roots  also  affect  the soil
oxidation-reduction potential by transporting oxygen via the
roots or  by  changing soil porosity.   In addition,  plants
moderate  swings  in  soil  water  potential  through
transpiration   and   by  the  continual   addition  of
water-retentive  organic   matter.     In  essence,  the
plant-microbe symbiotic relationship can be thought of as
being the  natural equivalent of  a  bioreactor that is
controlling  the   environmental   conditions   and  the
substances that  are required  by  the microbes for the
metabolism of contaminants in the  subsurface.  By use of
solar energy,  carbon  dioxide, water,  and  inorganic
nutrients,  plants  provide  naturally much  of  what the
bioremediation engineer must supply at a substantial cost
(Stomp 1993).

Phreatophytictreessuch as eastern cottonwoods (poplars)
and willows are rapid growing and  in terms of subsurface
biomass   and  transpiration  capacity,   offer  unique
opportunities for phytoremediation.  Several factors were
considered in the selection of eastern cottonwood trees for
this demonstration. These factors include extent and rate
of root  growth,  rate  of  evapotranspiration,  ability to
assimilate the  contaminant(s) of concern, and  ability to
thrive in the conditions at the site.
1.6    Key Contacts
Additional information on this project and the SITE Program
can be obtained from the following sources:
The Carswell Project

Mr. Gregory Harvey
Technology Implementer
ASC/EMR.
1801 Tenth Street Suite 2
(937) 255-3276
FAX: (937)255-4155
Email: qreqory.harvey@wpafb.af.mil
The Environmental Security Technology Certification
Program (ESTCP)

Dr. Jeff Marqusee
ESTCP Director
ESTCP Program Office
901 North Stuart Street
Suite 303
Arlington, VA 22203
(703)696-2117
FAX: (703) 696-2114
Email: ieffrev.marqusee@osd.mil
The SITE Program
Mr. Steven Rock
EPA Project Manager
National Risk Management Research Laboratory
U.S.  Environmental Protection Agency
26 West Martin Luther King Drive
Cincinnati, OH 45268
(513)569-7149
FAX: (513)569-7879
Email: rock.steven@epa.gov
Information on the SITE Program also is available through
the following on-line information clearinghouses:

   The Alternative Treatment Technology Information
   Center  (ATTIC)   System  is  a  comprehensive
   information  retrieval  system  containing  data  on
   alternative treatment technologies for hazardous waste
   including  thermal, biological, chemical  and physical
   treatment systems. ATTIC contains several databases
   that are accessed through a free, public access bulletin
   board. You may dial into ATTIC via modem at  (513)
   569-7610.  The   FTP   and  Telnet  address   is
   cinbbs.cin.epa.gov. The voice help line number is (513)
   569-7272.

-------
    The  Vendor  Information  System   for  Innovative
    Treatment Technologies   (VISITT)  is an electronic
    yellow pages of innovative treatment technologies and
    vendors.  Offered  by EPA's Technology  Innovation
    Office,  VISITT is  a  user-friendly database providing
    data   on  325  innovative  treatment  technologies
    provided  by 204  vendors. VISITT  is available for
    download at http://www.clu-in.org/. For instructions on
    downloading,   installing,  and  operating  VISITT, or
    submitting information for VISITT, call the help line at
    (800) 245-4505 or (703) 883-8448.

    The Hazardous Waste Clean-up Information Web Site
    provides   information  about  innovative  treatment
    technologies to the  hazardous waste community. It
    describes  programs, organizations,  publications and
    other tools for federal and state personnel, consulting
    engineers,  technology  developers   and  vendors,
    remediation  contractors,  researchers,  community
    groups  and  individual  citizens.  CLU-ln  may  be
    accessed at http://www.clu-in.org/.

Technical   reports  may be obtained by contacting the
Center for Environmental Research Information (CERI), 26
West Martin Luther King Drive in Cincinnati, Ohio, 45268 at
(513)569-7562.

-------
                                             SECTION 2
                        TECHNOLOGY APPLICATIONS ANALYSIS
This  section  of  the  report addresses  the  general
applicability of a phytoremediation system also known as
Short Rotation  Woody  Crop Groundwater Treatment
(SRWCGT) that employs hybrid  Poplar trees to remove
and   breakdown  organic  industrial  contaminants  in
groundwater as well as exert a measure  of  hydraulic
control over the treatment area so as to reduce adverse
contaminant migration. This analysis is based in part upon
the  results  of the  SITE  Program  Phytoremediation
demonstration conducted at the Carswell Naval Air Station
(MAS) Golf Club from April  1996  to September 1998 and
research conducted by others.

2.1     Key Features

Phytoremediation is a system that employs hybrid Poplar
trees to hydraulically control the migration of contaminated
groundwater,   as  well   as  biologically enhance   the
subsurface environment to  optimize  in-situ  reductive
dechlorination of the chlorinated  ethenes. The SRWCGT
system is a low-cost, easy to implement, low-maintenance
system that produces virtually no process residuals and
requires  minimal  maintenance.     The system is an
"evolving" and  adaptive process that  adjusts to site
conditions and increases its effectiveness overtime.

Phytoremediation  systems  represent  a broad  class of
emerging  remediation technologies that use plants and
their associated rhizospheric microorganisms to remove,
degrade, orcontain chemical and radioactive contaminants
in the soil, sediment, groundwater, surface water and even
the atmosphere. Phytoremediation is best described as a
solar-energy driven, passive technique that is applicable for
the remediation of sites  having low to moderate levels of
contaminants at shallow depth.  Phytoremediation takes
advantage of plants' nutrient utilization processes to take
in  water and  nutrients  through  roots,  transpire water
through leaves, and act as a transformational system to
metabolize organic compounds or absorb and accumulate
inorganic compounds.  Research has  found that certain
plants can be used to treat most classes of contaminants,
including petroleum  hydrocarbons, chlorinated  solvents,
pesticides, metals, radionuclides, explosives, and excess
nutrients. In addition, plants have also shown a capacity to
withstand  relatively  high  concentrations   of  organic
chemicals without the types of toxic effects  experienced
with bioremediation systems.  In some cases, plants have
demonstrated the ability to uptake and convert chemicals
quickly to less toxic  metabolites.  Depending upon  the
nature of contamination problems at a site and its particular
hydrogeologic setting, plant species are selected based on
their following characteristics:

       growth rate and yield,
       evapotranspiration potential,
       production of degradative enzymes,
       depth of root zone,
       contaminant tolerance,  and
       bioaccumulation ability.

Despite the fact that most of what is  known about this
technology is derived from laboratory and small scale field
studies,  phytoremediation  approaches have  received
higher public acceptance than most conventional remedial
options. Phytoremediation systems can be used along with
or, in some cases, in place of intrusive mechanical cleanup
methods.  Plant based remediation systems can function
with minimal  maintenance once  established, generate
fewer  air and  water emissions,  generate less secondary
waste, leave soil in place and generally are a fraction of the
cost incurred for a mechanical treatment approach.

2.2    Operability of the Technology

This discussion on technology operability will focus only on
phytoremediation systems that utilize hybrid poplar trees
to reduce the mass flux of chlorinated ethenes in shallow
groundwater systems through a combination of hydraulic
control  and  in-situ   microbially  mediated reductive
dechlorination. The hybrid Poplar tree system differs little
from other phytoremediation approaches in that it basically
involves  the  placement and maintenance  of trees  in
contaminated  regions.  Tree selection  and  preparation,
method of planting,  planting density,  distribution  and
dimensions of tree plots, agronomic inputs, irrigation and
maintenance requirements, are highly site specific and vary

-------
from site  to  site and  amongst  practitioners.   Since a
phytoremediation approach represents a living remediation
system, the planning, installation and maintenance of these
systems  rely  more  on  the  biological and ecological
sciences rather than standard engineering practices.

The  design,  installation,  monitoring  and  maintenance
requirements of a  phytoremdiation system that employs
Poplartrees are highly site-specific, as they are dependent
upon  the  physical,  chemical,  biological,  cultural  and
regulatory aspects of the site.   Factors that affect the
operability  of a  tree-based  phytoremediation  system
include, but are not limited to:

    Hydraulic framework,
    Physical and chemical properties of the soil,
    Distribution and magnitude of contamination,
    Climatic conditions,
    Property characteristics and features, and
    Treatment goals.

A  thorough understanding of each  of these factors  is
required   to   first  enable  a   technology  feasibility
determination, and secondly, to support decisions  on
implementability.

As with most sites with environmental problems, it is likely
that plenty of information has already been compiled on a
site's   features  and   contamination  problems.  This
information, generated by any  number  or  types  of
investigations, can  usually be obtained from the site owner
or operator, the appropriate State or Federal regulatory
agency  overseeing  activities  at the  site, the  local
government   (engineering,   public   works,   health
department),   municipal  or  county   library,  private
consultants and well drilling firms.  Despite the volumes of
information that may already be available on a candidate
phytoremediation site,  it  is still typically necessary  to
perform a series of limited, yet highly specific studies  to
better assist with design decisions, to establish appropriate
site preparation methods and to  determine  maintenance
tasks and schedules.

An understanding of the hydraulic framework of a site relies
on developing and  integrating  the following hydrogeologic
aspects for the site:

    groundwater flow direction,
    hydraulic gradient,
    connectivity of water bearing zones,
    identification of primary groundwater flow pathways,
    principal mechanism of groundwaterflow (intergranular
    or secondary porosity features),
    average depth  to groundwater,
    seasonal and diurnal groundwater level fluctuations,
    aquifer recharge points,
    interrelation  of the contaminated  aquifer with other
    aquifers or surface water features,
    aquifer thickness,
    groundwater velocity,
    volume of groundwaterthat flows through the proposed
    treatment area
    volume of groundwater stored in the aquifer beneath
    the proposed treatment area
    size and shape of the contaminant plume.

An understanding of the hydraulic setting is necessary for
determining whetherthis technology is feasible at a site. It
may be  discovered  after  evaluating certain hydraulic
parameters that the  contaminated aquifer is too deep,
beyond the reach of the hybrid Poplar tree roots. It may
also be discovered that  groundwater flow beneath the
proposed  treatment  area  is in excess  of  what  could
possibly  be attenuated through  some  combination of
hydraulic control and in-situ biologically mediated reductive
dechlorination.

An understanding of the hydrogeologic setting beneath a
site is important.  Many practitioners base most of their
design considerations solely upon the hydraulic constraints
of a tree-based phytoremediation system.  These design
considerations include:

    planting density (i.e., tree spacing),
    plot dimensions and orientation,
    number of plots needed, and
    arrangement of these plots across  the site.

An  effective  tree-based  phytoremdiation   system  is
dependent upon the collective  effort of  numerous  trees
evenly spaced  in a  series of  plots.    A   tree-based
phytoremediation system  is therefore  land  intensive,
requiring plenty of clear space, or at least enough for all the
trees that can be grown in a given area to do the job. It is
therefore importantto identify, and if economically feasible,
eliminate any obstacles or restrictive features on a property
that  might hamper the  effectiveness of a  tree-based
phytoremediation system. In order for the system  to be
effective the site should be cleared of any above or below
ground  obstructions   that  might  interfere  with  the
establishment and health  of the tree plots.

Tree stands or plantations  are oriented  so that the long
sides  of the stands  are  generally perpendicular to the
direction of groundwater flow (See Figure 4-1). The long
sides  of  the   plantations  generally span  the  most
concentrated portion of the contaminant plume. Individual
trees are planted in a series of rows. Tree spacing within
these rows is up to the discretion of the practitioner and is
determined on how quickly the practitioner wants to
achieve maximum stand-level transpiration rates.  Trees
are generally planted between 1.5 m to 2.5 m apart.

Although research has  shown that hydraulic control  is the
principle mechanism responsible for reductions in the mass
flux of contamination transported across the planted area
during the early stages of tree-based treatment,  other
mechanisms, especially  microbially mediated reductive
dechlorination  may become just as prominent after the

-------
third or forth  season. Reductive  dechlorination would
become the most important mechanism operating during
the dormant season. Therefore, applicability and design
decisions should not be based entirely on the ability of the
system to achieve hydraulic capture. Such decisions could
prove to be costly, resulting  in either more and/or larger
tree plots than are necessary, or the disqualification  of a
tree-based phytoremediation system as a viable alternative
for the site.  Hydraulic capture may not be possible or even
practical  at some sites,  yet the  desired  reduction in
contaminant mass flux might still  be  achieved through
some  combination   of   the  other  phytoremediation
mechanisms.  A  discussion  of  these  mechanisms is
presented in Section 4.0 of this report.

Before designing  any  remediation system, and the same
holds true for a tree-based phytoremediation system, it is
important to understand the treatment goals that have been
set for the  site.   Certain  goals may be based upon a
specific soil and/or groundwater cleanup criteria or based
upon a site  receptor risk.  Remediation goals may require
source removal or source control.  Each of these goals
implies  potentially different  design considerations  and
factors into  the overall treatment period.

Another important aspect to remember when designing a
tree-based phytoremediation technology is that the system
is a dynamic one and is capable of changing and adapting
to  particular site  conditions.   In areas characterized by
heterogenous  hydraulic  conditions,  trees  have been
observed to thin themselves  or increase their size based
upon  their  access to groundwater.  This is  especially
evident  with hydrogeologic settings  characterized  by
preferential groundwaterflow pathways (e.g., buried stream
channels).

Prevailing hydraulic conditions at a site generally determine
the time it takes for the trees to begin exerting an influence
on  the  groundwater  system.   Shallower  groundwater
systems would be more readily available to the tree roots,
requiring less time for the  system  to  begin affecting
changes in  the groundwater. Special planting techniques
may be implemented for an application on deeper aquifers
in order to speed up the time it normally takes for the roots
to  reach the contaminated aquifer.

An understanding of the physical and chemical properties
of a site's soil is important in knowing what adjustments
need to be  made  to the soil to foster healthy tree growth,
and in particular, vigorous root growth.  The condition of a
site's  soils  will also  be a factor  in deciding upon the
appropriate  tree  planting procedures.   The soil  in  a
proposed plot area might have to be reworked by plowing
and discing  appropriate  mixtures  of fertilizer   and
amendments (i.e., organic matter, drainage-enhancing
media) into the upper  portions of the soil profile. Special
rooting mixtures of fertilizers, organic-rich soil, native soil
and other amendments may have  to be formulated  and
placed into the tree boreholes or trenches during planting.
Soil moisture retention, soil moisture profiles, drainage and
infiltration  rates  factor  into decisions  regarding  the
necessity of an irrigation system or some type of ground
cover (i.e., grass, legumes).  An irrigation system might be
necessary during the first few growing seasons to provide
the trees with water until the roots reach the groundwater
table. It may also be necessary to install a ground cover to
make the trees less reliant on rainfall infiltration and force
them to seek out the aquifer as a source of water.

Understanding  the   distribution   and   magnitude  of
contamination at  a site is  important  for  the  proper
placement and dimensions of the tree plots and selection
of a tree type that has a natural  tolerance to the levels of
contamination it will encounter  at  the site.  To ensure
optimal positioning of the  plots,  it is important to pinpoint
contaminant source areas, discern historical contamination
patterns and activities that led to  those patterns at the site,
establish concentration gradients in both the soil and
groundwater  and   determine  the  plume  boundaries.
Groundwater contaminants  can be treated significantly
downgradient of the source through tree induced enhanced
bioremediation. Ideally, the phytoremediation plots should
be  positioned perpendicular to the path of migrating
contamination and straddle  an upgradient portion  of the
plume.  This is also the approach that should be taken  for
a treatment strategy intended to limit adverse contaminant
migration away from the site.

If  the  intent  is   to  utilize  the  trees  for  enhanced
bioremediation of the soil contaminants, then care should
be taken to position the tree plots over the  contaminant
source areas.  The trees would then be in position to take
up the contaminants where they would be transpired or
metabolized through enzymatic  reactions  in the tissue of
the  tree, or  broken  down in  the  rhizosphere  as  a
consequence of enhanced  microbial activity due to the
release of exudates and enzymes by the tree roots.

Climatic conditions at a site need  to be  evaluated with
regard to selecting appropriate tree type, determining the
arrangement and size of the plantations, and assessing the
need for an irrigation system. Generally, the trees should
be obtained locally, to ensure that the hybrid variation is
well  adapted to the local climate and less susceptible to
disease. The geographical location of the site dictates the
length of the growing season (i.e., the time when the trees
actively transpire water from the contaminated aquifer).
One can expect  longer growing seasons in the  lower
latitude regions as opposed to higher latitude  regions.
Regardless  of the geographic  location,  each site  will
experience a dormant period when the trees stop pumping
groundwater.  During  these dormant  cycles,  microbial
mediated reductive dechlorination becomes the dominant
remedial mechanism. Regions characterized by hot and dry
summers might need to operate a  drip irrigation system
during the first few growing seasons until the tree roots
extend down to the aquifer.
                                                     10

-------
2.3    Applicable Wastes

Tree-based phytoremediation systems operate through a
process of phytotransformation, which involves the uptake
of organic and  nutrient contaminants from the soil and
groundwater by the tree's roots, followed by the breakdown
of these compounds in the tissue of the tree  (Schnoor,
1997). The direct uptake of organics by trees has been
found to be a surprisingly efficient removal mechanism at
sites  contaminated  at  shallow depth  with moderately
hydrophobic  organic chemicals (octanol-water partition
coefficients, log Kow = 1 to 3.5).

A tree-based phytoremediation system is applicable to sites
where the principal  soil  and groundwater contaminants
consist of benzene, toluene, ethylbenzene and xylenes
(BTEX), chlorinated  organics  and  short-chain aliphatic
compounds.  Given this  list of chemicals,  a tree-based
phytoremediation  system may  be  applicable  at  the
following waste sites: Petrochemical sites, ammunition
waste sites, fuel spills, chlorinated solvent plumes, landfill
leachates  and  agricultural chemicals  (pesticides and
fertilizers).

2.4    Availability and Transportability of the
        Equipment

Unlike a  traditional remediation system,  a tree-based
phytoremediation system is a living remediation technology
that does not have any equipment requirements otherthan
those which are necessary to install, maintain and monitor
such a system. Tree-based phytoremediation systems are
highly site  specific  in-situ  approaches  and are  not
considered transportable. The working components of a
tree-based phytoremdiation system are the roots, stems
and leaves of the trees. The trees for this type of system
can usually be obtained locally from a nursery ortree farm.
Trees would  be delivered to the site via  flat-bed  truck.
Equipment required  to install the system is entirely site
specific, and  to a large extent, dependent upon  the soil
conditions, depth to  which the trees need to be planted,
and the size of the plots. Trenching equipment was used to
install the SRWCGT system at the Carswell  MAS. The
practitioner might choose to out source any ripping (Florida
Forestry Information, accessed September 2001, at URL
http://www.sfrc.ufl.edu/Extension/ffws/home.htm).,
trenching,  or borehole  drilling deemed  necessary to
establish the tree  plots  to  a local agricultural  land
preparation company, construction firm, orwell drilling firm
that  has the  specialized equipment  and  experience to
perform this work.  These construction and well drilling
firms  may also  be called upon to install portions of the
monitoring system, which  may include the installation of
monitoring wells, peizometers, soil moisture sensors, and
soil borings.  Other equipment  that might be  necessary
during any ground preparation activities may include a
backhoe, front-end loader and skid mounted  loader for
moving fertilizer, top soil  and fill around the site, a mixing
unit and a screen for formulating the root mix, a trencher
for burying data cables and irrigation pipe, and discing and
plowing equipment for loosening up the ground and mixing
in fertilizer and soil conditioners.  All of this equipment can
be obtained  locally  and is  usually available  for rent.
Equipment for an irrigation system can usually be obtained
from a local plumbing supplier or home center. There is a
considerable amount of equipment available for monitoring
a tree-based  phytoremediation system,  and  there  is
considerable variation in sophistication and cost. Much of
this  equipment can  be obtained  from companies that
specialize  in  products  (i.e.,  plant bio-sensors,  tree
transpiration  measurements, plant  bio-productivity and
environmental conditions) that support the agricultural
community.

Typical   monitoring   equipment   for   tree-based
phytoremediation systems includes a network of monitoring
wells.  Water levels in  monitoring  wells provide a direct
means for assessing groundwater uptake  by the trees.
These wells can  be  equipped with electronic pressure
transducers connected to data loggers for continuous water
level monitoring.  Soil moisture sensors can be arrayed
across the site and installed at various depths to track
changes  in  soil  moisture as a function  of root mass
development. Soil moisture data can be collected on data
loggers and  used for decisions  on when to irrigate.
Weatherstations are often installed and the data collected
by them is used in conjunction with sap flow measurements
to estimate tree transpiration rates.

2.5    Materials Handling Requirements

A tree-based phytoremediation system does have some
materials handling requirements,  especially during the
installation phase. Depending upon soil conditions, tree
plot  areas might  require  plowing,  tilling, and discing  to
facilitate fertilizer  infiltration, increase soil  porosity, ease
planting and foster vigorous root growth. The equipment
needed to do this can usually be rented locally. Depending
upon the tree planting requirements for a site, the proposed
plots may have to be ripped or trenched, or boreholes may
have to be drilled. Ripping can be contracted out to an
agricultural   land  preparation   company.  Trenching
equipment can usually be obtained locally.  A subcontract
arrangement  is typically  needed for the drilling  of any
boreholes.  Fertilizer and soil conditioners may have to be
mixed into the soil or used to formulate specialized root
mixtures that will be placed in the boreholes or trenches at
the   time the  trees  are planted.   Fertilizer and soil
conditioning components  could include  any variety  of
commercial  fertilizer mixes depending on  the desired
nitrogen/phosphorus/potassium   (N/P/K)   ratios.  Soil
conditioning materials have traditionally included organic
carbon, aged manure, sewage sludge, compost, straw and
mulch. A mix mill/grinder and spreader might be needed
for handling the  fertilizer and various  soil conditioners.
Screening equipment (i.e., subsurface combs,  portable
                                                    11

-------
vibrating screens) may also be necessary to remove debris
and cobbles from the soil and to remove debris from soil
conditioning material.

In addition to a drill rig and  some  of the  agricultural
equipment mentioned,   a  tree-based  phytoremediation
system  normally  requires  an  assortment  of  heavy
equipmentduring the installation phase. Excavators, back-
hoes  and  trenchers are needed to create trenches for
planting tree rows and for laying irrigation piping and  data
cables.  Dump trucks  and front-end loaders would be
required for delivering and/or  moving  soil  and  soil
conditioners around the site.  Flat-bed trucks  might be
required for delivering  trees, seed, fertilizer and other
supplies.  Graders  and scrappers would be used for re-
leveling the ground  surface aftertree installation. Fork lifts
would be used for moving pallets and waste drums around
the site.

Contaminated soil  would  require specialized  handling,
storage and disposal requirements. Soil may have to be
kept damp when being  reworked to limit dust production.
Contaminated drill cuttings usually have to be containerized
(usually in  55-gallon drums) and disposed of at a permitted
disposal facility.  Contaminated soil could  be generated
during any drilling and excavation activities.

As many as 1,000 to 2,000 trees per acre may be initially
planted  to  assure   a  significant   amount   of
evapotranspiration  in the first few years.  The trees will
naturally thin themselves through competition to 600 to 800
trees per acre over the  first six years.  In order to off-set
some of  the costs  associated  with this remediation
technology,  the trees can  be harvested  on a six-year
rotation and sold for fuelwood or pulp and paper. The trees
will grow back from the cut-stump.

2.6     Site Support Requirements

Phytoremediation systems in general have minimal site
support requirements. Typically, these systems require few
utilities to operate. Water is generally needed for irrigation
and possibly decontamination purposes. A drip irrigation
system may be installed and operated periodically overthe
first few growing seasons when the  young trees are most
susceptible to water stress problems.  It may be operated
at times afterwards to  make up  for rainfall deficits that
occur during times  of drought.  Irrigation water would not
necessarily have to be  potable water.  Depending upon
local regulations, water from the contaminated aquifer
might be used at no cost, with the additional benefit  of
enhancing  groundwater treatment  during  the  first few
growing seasons when little remediation is expected.  The
electricity needed to operate well pumps can be provided
by small generators.   Monitoring equipment (e.g., soil
moisture probes,  pressure  transducers,  data loggers,
weatherstation components) can be powered by batteries
or solar panels.

Depending upon site location, security measures might be
required to protect the public from accidental exposures
and prevent accidental and intentional damage to the trees
and monitoring equipment. A fence would also serve the
purpose of discouraging local wildlife from using the trees
as a food source (i.e., deer, beavers).

2.7    Range of Suitable Site Characteristics

Tree-based phytoremediation is best applied to sites with
relatively shallow soil and groundwater contamination. The
contaminants can  be organic  or inorganic, but should
possess certain physical and  chemical properties that
make them amenable to phytotransformation, rhizosphere
bioremediation, and phytoextraction.  This technology is
well suited for use at very large field sites where  other
methods of remediation are not cost-effective or practical.
It is also best utilized at sites with low concentrations of
contaminants where the remediation objectives for the site
are consistent with a long-term contaminant reduction
strategy. Sites should have plenty of open space, and be
clear of man-made structures; existing vegetation can be
left intact.

2.8    Limitations of the  Technology

Research and data from various field demonstrations have
shown  that tree-based phytoremediation systems are a
promising,  cost-effective  and  aesthetically   pleasing
remediation alternative that has been successfully applied
at a number of sites.   Unfortunately, many  of these
applications have been at small  sites, where few funds are
available for long-term compliance monitoring. Long-term
monitoring and evaluation of tree-based phytoremediation
technologies   is  needed   to  demonstrate   system
effectiveness   and  better  define   phytoremediation
mechanisms.  Although  current  research  continues  to
explore and push the  boundaries of phytoremediation
applications, there are some limiting factors that need to be
considered.

Contaminant to root contact, a function of root depth and
mass, is a limiting factor for direct uptake of contaminants
into the tree, but not for enhanced reductive dechlorination
processes. While  most  phytoremediation  systems are
limited  to the upper 3 meters of the soil column, research
and SITE Program experience suggests that hybrid Poplar
systems may be effective to depths greater than 8 meters.
Systems that utilize other tree species may be effective to
even greater depths.  To overcome these depth barriers,
researchers and companies that offer phytoremediation
services have developed and employed specialized (often
proprietary) techniques that train the tree roots to penetrate
to greater depths, or herd them  into deeper contamination
zones through the use of subsurface drip irrigation. Deeper
zones of contamination can possibly be treated through a
process of pumping the contaminated groundwater to the
surface and applying  it  to the  plantations through drip
irrigation.  On   the  other hand, enhanced   reductive
dechlorination  is more dependent on the availability  of
                                                    12

-------
dissolved  organic carbon in the  groundwater, which  is
typically increased in the soil water and groundwater
beneath the tree stands.

Contamination that is too tightly bound to  the  organic
portions of a soil and root  surfaces  may also  pose
limitations on  the effectiveness of this technology. This is
especially true with hydrophobic compounds (log Kow> 3.5),
which due to their  octanol-water partition  coefficients,
cannot be easily translocated within the tree or are simply
unavailable to microorganisms in the rhizosphere. On the
other hand, contaminants that are too watersoluble (logKow
<1.0) are  not sufficiently sorbed to  roots  nor  actively
transported through plant membranes. These contaminants
would simply  pass through the roots unimpeded.

Another limiting factor is that tree-based phytoremediation
may require more  time to achieve  cleanup standards than
other  more  costly  treatment   alternatives,  such  as
excavation, landfilling, or incineration.    A tree-based
phytoremediation  system may take ten  plus years to
completely remediate a site. This type of Phytoremediation
system  is  limited by the  growth  rate  of the trees.
Depending upon the depth to  groundwater, the length of
growing season and tree type, it may take two or more
growing seasons before the trees start to  exert a hydraulic
effect on the contaminated aquifer and even longer before
microbial mediated reductive  dechlorination becomes a
viable mechanism. In addition, removal and degradation of
organics in contaminated matrices is likely limited by mass
transfer. The desorption and mass transport of chemicals
from soil particles to the aqueous phase  may therefore
become a rate limiting  step.

Tree-based phytoremediation systems may not be the most
suitable remediation technique for sites  that pose  acute
risks for human and  other ecological receptors. Although
trees have shown  a remarkable tolerance to contaminant
levels  often  considered  too toxic  for bioremediation
approaches,  very high concentrations of organics may
actually inhibit tree growth, thus limiting the application of
this technology at some sites or portions of sites (Dietz and
Schnoor, 2001).  Sites that possess phytotoxic levels of
organic contamination  and pose acute exposure risks are
best handled by first applying a faster, more expensive ex-
situ technique. A tree-based phytoremediation system can
then serve as a final polishing step to close the site after
other clean-up technologies have been used to treat the
hot spots.

Practitioners of tree-based phytoremediation still  need to
better document the fate of organic contaminants in tree
tissue, establish  whether contaminants can  collect  in
leaves and be released during litter fall, or accumulate in
fuel wood or mulch.

There  has been  some concern over  the  potential of
ecological exposures whenever plants are used to interact
with contaminants. Of course this threat is more obvious
and better understood for plants used for the purpose of
extracting  and  accumulating   heavy  metals   and
radionuclides. Unlike metals, some research has shown
that most organic contaminants do not accumulate in
significant amounts in plant tissue. Nonetheless, if some
organisms (e.g., caterpillars, rodents,  birds, deer, etc.)
seem likely to ingest significant amounts of the vegetation,
and if harmful bioconcentration up the food chain  is a
concern during the life of the remediation effort, appropriate
exposure  control  measures  should  be implemented
including perimeter fencing, overhead  netting, and  pre-
flowering harvesting.

Another issue that  might be  a limiting factor from a
regulatory standpoint is the transfer of the contaminants or
metabolites to the atmosphere.  A number of studies have
been conducted to determine if organic contaminants, such
as TCE, simply pass through the trees and are released to
the  atmosphere  through   leaf  stomata   during
evapotranspiration.  Research in this area has  produced
mixed results and is not close to quantifying the amounts
of organics released.    According to  some  studies,
transpiration of TCE to the atmosphere has been measured
(Newman et al. 1997), but little information is available that
indicates any release of more toxic daughter products (i.e.,
vinyl chloride). The same researcher has shown that a
series of aerobic transformations occur whereby some of
the TCE is transformed to trichloroethanol, trichloroacetic
acid, and dichloroacetic acid by hybrid Poplar trees.


2.9    Technology Performance Versus Arars

Under  the  Comprehensive  Environmental  Response,
Compensation, and Liability  Act (CERCLA), as  amended
by the Superfund Amendments and Reauthorization Act of
1986 (SARA), remedial actions undertaken at Superfund
sites must comply with federal and state (if more stringent)
environmental laws that are determined to be applicable or
relevant and appropriate requirements (ARARs). ARARs
are determined on a site-specific basis by the remedial
project manager. They are  used as a  tool to guide the
remedial project managertoward the most environmentally
safe way to manage remediation activities. The remedial
project manager reviews each federal environmental law
and determines if it  is  applicable. If the  law is not
applicable, then the determination must be made whether
the law is relevant and appropriate.

This subsection discusses specific federal environmental
regulations  pertinent  to  the  operation of tree-based
phytoremediation systems,   including   the  transport,
treatment, storage and disposal of wastes and treatment
residuals.   Federal and state ARARs  are presented in
Table 2-1. These regulations are reviewed with respect to
the demonstration results. State and local requirements,
which  may be more stringent, must also be addressed by
                                                    13

-------
remedial project managers.

2.9.7   Comprehensive  Environmental  Response,
       Compensation, and Liability Act (CERCLA)

The CERCLA  of 1980  as amended by the Superfund
Amendments and Reauthorization Act (SARA) of 1986
provides for federal funding to respond to releases or
potential releases of any  hazardous substance into the
environment, as well as  to releases  of  pollutants or
contaminants that may present an imminent or significant
danger to public health and welfare or to the environment.
As part of the  requirements of CERCLA, the EPA has
prepared  the National Oil and Hazardous Substances
Pollution  Contingency  Plan   (NCP)  for  hazardous
substance response. The NCP is codified in Title 40 Code
of Federal Regulations (CFR) Part 300, and delineates the
methods and criteria used to determine  the appropriate
extent  of removal  and cleanup for hazardous  waste
contamination.

SARA states a strong statutory preference for remedies
that are highly reliable and  provide long-term protection. It
directs EPA to do the following:

    Use  remedial  alternatives that permanently  and
    significantly reduce the volume, toxicity, orthe mobility
    of hazardous substances, pollutants, or contaminants;

    Select remedial actions that protect human health and
    the  environment,  are cost-effective,  and  involve
    permanent solutions  and  alternative  treatment or
    resource recoverytechnologiestothe maximumextent
    possible; and

    Avoid off-site transport and disposal  of untreated
    hazardous substances or contaminated materials when
    practicable  treatment  technologies  exist  [Section
The  Carswell  NAS demonstration site  is  part  of  a
Superfund  site  (Air Force  Plant  No. 4)  ; therefore,
CERCLA/SARA  is relevant  and  appropriate  for the
treatment   technology   occurring  on-site.   The
phytoremediation system at the Carswell site meets most
of the SARA criteria. It is an in situ treatment technology,
thus  the  treatment process  occurred  in place  and the
removal of the contamination is permanent and protective
to human health and the  environment; the volume and
mobility  of  halogenated  organics  in  the soil  and
groundwater is reduced to help prevent the migration of
contamination off-site orto uncontaminated watersupplies;
phytoremediation reduces the toxicity of the treated waste
media (soil or groundwater); and phytoremediation is cost-
effective and an alternative treatment technology.
In general, two types of responses are possible under
CERCLA:  removal and remedial action.   Superfund
removal  actions  are  conducted  in  response to  an
immediate threat caused by  a  release of a hazardous
substance.  Many removals  involve small quantities of
waste of immediate threat requiring quick action to alleviate
the hazard. Remedial actions are governed by the SARA
amendments  to CERCLA.   As  stated  above, these
amendments promote remedies that permanently reduce
the volume, toxicity, and  mobility of hazardous substances
or pollutants. The tree-based  phytoremediation system is
likely to be part of a CERCLA remedial action.  Remedial
actions are governed  by  the  SARA  amendments to
CERCLA.

On-site remedial actions must comply with federal and
more stringent state ARARs. ARARs are determined on a
site-by-site basis and may be waived under six conditions:
(1) the action is an interim measure, and the ARAR will be
met at  completion; (2) compliance with the ARAR would
pose a greater risk to health and the  environment than
noncompliance; (3) it is technically impracticable to  meet
the ARAR; (4)  the standard of performance of an ARAR
can be met by an equivalent method; (5) a state ARAR has
not been  consistently applied elsewhere; and (6) ARAR
compliance would not provide  a  balance  between the
protection achieved at a particularsite and demands on the
Superfund for other sites. These waiver options apply only
to Superfund actions taken on-site, and justification for the
waiver must be clearly demonstrated.

2.9.2   Resource Conservation  and  Recovery Act
       (RCRA)

RCRA, an amendment to the Solid Waste Disposal Act
(SWDA),   is the primary  federal legislation governing
hazardous waste activities.   It  was passed in 1976 to
address the problem of how to  safely  dispose of the
enormous volume of municipal and industrial solid waste
generated  annually.    Subtitle C of  RCRA  contains
requirements forgeneration, transport, treatment, storage,
and disposal of hazardous waste, most of which are also
applicable to CERCLA activities.  The Hazardous and Solid
Waste Amendments (HSWA) of 1984 greatly expanded the
scope and requirements of RCRA.

RCRA regulations define hazardous wastes and regulate
their transport, treatment, storage, and disposal. These
regulations  are only   applicable to  the  tree-based
phytoremediation  system  if  RCRA-defined hazardous
wastes are  present.    If  soils are determined to  be
hazardous  according to RCRA (either because  of a
characteristic or a listing  carried by the waste), essentially
all RCRA  requirements  regarding the  management and
disposal of this hazardous waste will need to be addressed
                                                   14

-------













E

>. w c g 'a 73

* CL -Q $ ro (ARARs)fortheTree-B IS. late Requiremen Q- o Q- Q- c ro c ro 0)

- .P. ro t c o) ra Q_ — 0)8 1 £• o£ Q.-D Q. c < ro CO 3 o T3 8 ro .c ro ±± 0) | 5 'S Id be characterized to d a RCRA-listed waste. =5 (fl o ' treated waste sh ste, and if so, if il c ra 3 3 o ' — ' s CN ^> (V c LJ_ ^ 0 g it • • LJJ ry (D o S o: co *o ^3 "CO 03 03 •C 5 OJ "O O -!-• 2 S •— tu ro ii 6 § ro c .E ^ ^ of = «>" £ '"=• ^^ O^Q- O o QJ ?300) §i „ o -7^ O ^ •— O ^ •— O) JU . _Q . LL. QJ c QJ c — (O r^ tn r^ c c ro c O Iflf l|l| till || ll| ||1 0.0)0-0) i=oa:^ < cnra ra - >; .2 .E T3 ^c -1 73=-^ ^a, 0) ^S, O "-°-S •- E oj Q^CJ) ^ oro.s5 o^ E — ID ^o) •" "^S g" """ro | §1 g| E |ro | ll|-§ ifc ?| 1 f| 1 ||S'| ^WT3 Era '=! -^Q. 0 "^O)""0 .?.= > o.ra o) S° J5^ rai?3~ w^"o ra-^ ^ o)g °-g '"o'cl II- H 1 1| |l sill 111 ll !< =" IS Ei«? lit h tl l« |i illi It: if ;« M fi 111! 2££ roro ^§ g|=feE ^.g;? Q._0 -Ci- ^^ ""^S > .E™^73 111 it 1 11 if 11 ill lit 1^ til !! 111! Q_ Q__Q E H ^ E 0^5^ COQ^ CO^ OCD)(fl R ° ^ ^ ;- rf CD CD • i- ^ CN CN o o g v; „ T~ ^" S (~*\ m rtl m O) CD O — ™ ™ QJ rv rv CN ~C • - iv "*"' rv i • LJ_ u_"D^ C := C u_ | , CO O °ro> ^? ^ <" ° -2"-- O o ^ gj ?"''§• ^CN?"- ^C^ °~>^°~^^ ^ ^ ^ ^ ^^ T~ ^ ^3~ ^ ^ S ^ ^ S I Q^ « -i— i fV s — ' fV fV s — ' Q_ _Q '^ Q_ _Q '^ W O^J5 ^CN1-1- ^CN O^crO^cr O Q^COCO COr-O COr- Q^COLJJQ^COLJJ E ^ D) C (/) C OJ ._ (/) w ro ro o w > ^ ^ g "E u 9 aj w ^3 -2 to m "S w ro ^ ro •oluc °^^>'-^^3 «i- ro(/)-5 Q.m^OQ.-jT m o>.2 ro ^(S 2 o 3 M a) c~-— -i-iiocc-Q ro i^ > ro (/) QJ -^^ _^ ro ro l— Qra.E ^KQ-jioS W 15


-------


E
,
CL
T3
 S "5
il 8
|fl
a "§ 0
c a "S
l§|
£2-2
"O T3 o
-2 S 'c
J^ CO Q_
'E <" x
E D) 0
CO £= Z
SJQ
= ^<
ITS &



CM
c
g
0
 CO
-^ C5

















T3
£= (0
Applicable if undergrou
injection is selected as





^5
ts ts
11
ro c
-E.o
o w
0 <"
"- 'F
o E
—  C
0 fc
Q.T3
.(/) £=
T3 ^3
>— -I-T
O  .«
"O D)
CO °
T3 O
£= £=
JS-S
ic treatment s
ng specific tei
Hazardous wastes must meet specif
land disposal, or must be treated usi




-e
CO
Q_
Cd
LL
O
O
^J"

goo
Cd c\i















16

-------
by the remedial managers.  Wastes defined as hazardous
under RCRA  include  characteristic and listed wastes.
Criteria for identifying characteristic hazardous wastes are
included in 40 CFR Part 261 Subpart C. Listed wastes
from specific  and  nonspecific  industrial sources,  off-
specification products,  spill cleanups, and other industrial
sources are itemized  in 40  CFR Part  261  Subpart D.
RCRA  regulations do  not  apply to sites where RCRA-
defined wastes are not present.

Unless they are  specifically delisted through  delisting
procedures, hazardous wastes listed in 40 CFR  Part 261
Subpart D currently remain  listed wastes  regardless of the
treatment they may undergo and regardless of the final
contamination  levels in the resulting effluent streams and
residues. This implies that even after remediation, treated
wastes are still classified as hazardous  wastes  because
the pre-treatment material was a listed waste.

For  generation  of any  hazardous waste,   the  site
responsible  party  must obtain  an EPA  identification
number. Other applicable RCRA requirements may include
a Uniform  Hazardous Waste Manifest  (if the waste  is
transported off-site), restrictions on placing  the  waste  in
land disposal units, time limits on accumulating waste, and
permits for storing the waste.

Requirements  for corrective action at  RCRA-regulated
facilities are provided  in 40 CFR Part  264,  Subpart  F
(promulgated)  and  Subpart S  (partially promulgated).
These  subparts also generally apply to remediation  at
Superfund sites.  Subparts F and S include requirements
for  initiating and  conducting RCRA corrective action,
remediating groundwater,  and  ensuring that corrective
actions  comply with  other  environmental  regulations.
Subpart S also details conditions under which particular
RCRA  requirements  may be  waived  for  temporary
treatment  units operating at corrective  action sites and
provides information regarding requirements for modifying
permits to adequately describe the subject treatment unit.

2.9.3   Clean Air Act (CAA)

The CAA establishes national primary and  secondary
ambient air quality standards for sulfur oxides, particulate
matter, carbon monoxide,  ozone,  nitrogen  dioxide, and
lead.  It also limits the emission of 189 listed hazardous
pollutants  such as vinyl chloride, arsenic, asbestos and
benzene.  States are responsible for enforcing the CAA.
To assist in this, Air Quality  Control Regions (AQCR) were
established. Allowable emission limits are determined by
the AQCR, or its  sub-unit, the  Air Quality Management
District (AQMD).  These emission limits are based on
whether or not the region is currently within attainment for
National Ambient Air Quality Standards (NAAQS).

The CAA requires that treatment, storage, and  disposal
facilities comply with primary and secondary ambient air
quality standards.  Fugitive  emissions from the tree-based
phytoremediation  system   may  come  from  (1)  soil
conditioning and borehole drilling activities and (2) periodic
sampling activities.  Soil  moisture should be  managed
during system installation to prevent or minimize the impact
from   fugitive   emissions.   Although  rhizospheric
biodegradation  and breakdown  of  chemicals through
metabolic activities within plant tissue are components of
phytoremediation, these processes as they relate to this
technology are not well  understood.   There is  some
concern that organic contaminants are only partially broken
down,  implying  that an unknown portion of the original
contaminants and its daughter products may be released
to the atmosphere during evapotranspiration.

No   air  permits   are  required  for   the  tree-based
phytoremediation system operated at the Carswell NAS
Golf Club.

2.9.4   Clean Water Act (CWA)

The objective of the Clean Water Act  is to restore and
maintain the chemical, physical and biological  integrity of
the nation's waters by establishing federal, state, and local
discharge standards.  If treated water is discharged to
surface water bodies or Publicly Owned Treatment Works
(POTW), CWA regulations will apply.  A facility desiring to
discharge water to a navigable waterway must apply for a
permit  under the National Pollutant Discharge Elimination
System (NPDES).  When  a  NPDES permit is issued, it
includes waste  discharge  requirements.  Discharges to
POTWs also must comply  with general  pretreatment
regulations outlined in 40CFR Part 403, as well as other
applicable state and local administrative and substantive
requirements.

Other  than the tree's  capacity to  pump groundwater,
phytoremediation technologies generally do not involve the
mechanical  pumping,  treatment  and  discharge  of
surface/groundwater.   In  a  few  rare cases  where
contaminated groundwater occurs at depth, mechanical
pumping  might be used to  bring the water to the surface
where  it  would then be applied to  the  plants via  drip
irrigation.   Since this  water technically would not be
discharged to a navigable  waterway, it is unlikely that a
NPDES permit will apply.

2.9.5   Safe Drinking Water Act (SDWA)

TheSDWAof 1974, as most recently amended by the Safe
Drinking Water Amendments of 1986, requires the EPA to
establish  regulations  to   protect  human  health  from
contaminants in drinking water. The legislation  authorized
national drinking water standards and a joint federal-state
system for ensuring compliance with these standards.

The National Primary Drinking Water Standards are found
in 40 CFR Parts 141 through 149.  These drinking water
standards are expressed as maximum contaminant levels
(MCLs) for some constituents, and maximum contaminant
level goals (MCLGs) for others. Under CERCLA (Section
121 (d) (2) (A) (ii)), remedial actions are required to  meet
                                                    17

-------
the standards of the MCLGs when relevant. Since a tree-
based   phytoremediation  system   is   considered   a
groundwater remediation system, it is likely that these
standards would be applicable.

Parts 144 and  145 discuss requirements associated with
the underground  injection of  contaminated  water.   If
processing pumped contaminated groundwaterthrough the
plantation's drip irrigation system is an option, approval
from   EPA   for  constructing  and   operating  this
phytoremediation system in this mode will be required.

2.9.6   Toxic Substances Control Act (TSCA)

The  TSCA of 1976 Grants the U.S. EPA authority to
prohibit orcontrolthe manufacturing, importing, processing,
use, and disposal of any chemical substance that presents
an unreasonable  risk  of  injury to human health or the
environment. These  regulations may be found in 40 CFR
Part  761; Section  6(e) deals specifically with PCBs.
Materials with less than 50 ppm PCB are classified as non-
PCB; those  containing between  50 and  500 ppm are
classified as PCB-contaminated; and those with 500 ppm
PCB or greater are classified as PCB. PCB-contaminated
materials may  be disposed of in TSCA-permitted landfills
or destroyed   by   incineration  at   a TSCA-approved
incinerator; PCBs must be incinerated. Sites where spills of
PCB-contaminated material or PCBs have occurred after
May  4, 1987 must  be addressed under the PCB Spill
Cleanup Policy in 40 CFR Part 761, Subpart G. The policy
establishes cleanup protocols for addressing such releases
based  upon the volume and concentration  of the spilled
material. There is little if any documentation supporting
tree-based  phytoremediation as a  viable option in the
remediation of PCBs. The properties of PCBs do not make
it amenable for direct uptake by the roots of the trees. It is
however   possible  that   enhanced   rhizospheric
bioremediation may be capable of breaking down some
PCB congeners.

2.9.7   Occupational Safety and Health Administration
        (OSHA) Requirements

CERCLA remedial actions and  RCRA corrective actions
must  be  performed  in  accordance with  the  OSHA
requirements detailed in 20 CFR Parts 1900 through 1926,
especially Part 1910.120, which provides forthe health and
safety  of workers at  hazardous waste sites.  On-site
construction activities at Superfund  or RCRA corrective
action  sites must be performed in accordance with Part
1926  of OSHA,  which  describes  safety  and health
regulations  for  construction  sites.    State  OSHA
requirements,  which may be significantly stricter than
federal standards, must also be met.
All technicians involved with the construction and operation
of a tree-based phytoremediation system may be required
to have completed an OSHA training course and be familiar
with all OSHA requirements relevant to hazardous waste
sites.  Workers on hazardous waste sites must also be
enrolled in a medical monitoring program. The elements of
any acceptable program must include: (1) a health history,
(2) an initial exam before hazardous waste work starts to
establish fitness for duty and as a medical baseline, (3)
periodic  examinations  (usually annual)  to  determine
whether changes due to exposure may have occurred and
to ensure continued fitness for the job,  (4) appropriate
medical  examinations  after  a  suspected  or  known
overexposure, and (5) an examination at termination.

For most sites, minimum PPE for workers will include
gloves, hard hats, steel-toe boots, and Tyvek® coveralls.
Depending on  contaminant  types  and  concentrations,
additional PPE may be required, including the  use of air
purifying  respirators or supplied air.  Noise levels are not
expected to be high, except during the ground preparation
and tree planting phase which will involve the operation of
heavy equipment.  During these activities, noise  levels
should be monitored  to  ensure that  workers are not
exposed to noise levels above a time-weighted average of
85 decibels overan eight-hourday. If noise levels increase
above this limit, then workers will be  required to wear
hearing protection.  The levels of noise anticipated are not
expected to adversely affect the community, but this will
depend on proximity to the treatment site.

2.9.8   State Requirements

State and local  regulatory agencies may require permits
prior to the operation of a tree-based  phytoremediation
system.  Most federal  permits  will  be  issued by the
authorized state agency. If, for example, the contaminated
drill cutting waste is considered a RCRA waste, a permit
issued by the state may be required to operate the system
as a treatment,  storage, and disposal  (TSD) facility.  The
state may also require a TSD permit for on-site storage
greater than 90 days of hazardous waste. An  air permit
issued by the state Air Quality Control Region may be
required if air emissions in excess of regulatory criteria, or
of toxic concern, are anticipated.  Local state agencies will
have  direct  regulatory responsibility  for environmental
media issues. If remediation is at a Superfund site, federal
agencies, primarily the U.S. EPA, will  provide regulatory
oversight.  If off-site disposal of contaminated waste is
required, the waste must be taken to the disposal facility by
a licensed transporter.
                                                    18

-------
                                             SECTION 3
                                      ECONOMIC ANALYSIS
3.1    Introduction

The costs associated with applying a Short Rotation Woody
Crop Groundwater Treatment (SRWCGT)System as an
option for the remediation of halogenated hydrocarbons in
shallow groundwater systems and the hydraulic control of
contaminant  migration  have  been  broken  out  and
discussed under the 12 cost categories that reflect typical
cleanup activities performed at Superfund sites:

(1)   Site Preparation;
(2)   Permitting and Regulatory Requirements;
(3)   Capital Equipment;
(4)   Start-up and Fixed Costs;
(5)   Labor;
(6)   Consumables and Supplies;
(7)   Utilities;
(8)   Effluent Treatment and  Disposal;
(9)   Residual Waste Shipping, Handling, and Disposal
     Costs;
(10)  Analytical Services;
(11)  Maintenance and Modifications; and
(12)  Demobilization.

The primary purpose of this economic analysis is to provide
a cost estimate for a commercial application of a SRWCGT
system using Poplar trees. This analysis is based on the
assumptions and costs provided by U.S. Air Force project
personnel, and on the results and experiences gained from
a 3-year  SITE demonstration of the process on a TCE
contaminated shallow aquifer at the Carswell  Naval Air
Station (MAS) Golf Club, Fort Worth, Texas. Table 3-1
presents the costs for an application at a 200,000-ft2 (~4.6
Acres)  hypothetical  model site. When  appropriate  and
relevant, some of the cost figures for the model site were
derived from actual costs and design criteria used for the
Carswell MAS Golf Club system. These costs and design
criteria were then applied to a  hypothetical set of hydraulic
and chemical conditions at the model site.  The costs listed
in each of the  12  categories for the  model  site  are
estimates of the actual costs that might be incurred during
a more typical application, due to the following reasons:
     A larger overall treatment area
     An aquifer system with  a lower flow and  more
     uniform flow regime
     An aquifer system that is somewhat insulated from
     the influences of other features (i.e., tides, streams)
     A treatment  plot width  and spacing  pattern  that
     ensures significant hydraulic control.
     A site monitoring and analytical program that is more
     typical  for   a   commercial  application   of  the
     technology.

This economic  analysis is designed to conform to the
specifications for an Order-of-Magnitude estimate. This is
a  level  of  precision  established  by  the American
Association of Cost Engineers (AACE) forestimates having
an expected accuracy within +50 percent and -30 percent.
In  the  AACE definition, these  estimates are generated
without detailed engineering data.   Suggested uses  of
these estimates are feasibility studies  or as aids in the
selection of alternative processes. The costs derived for
this Phytoremediation application are much more accurate
than these specifications, since actual costs incurred from
the Carswell  MAS  Golf Club SITE  Demonstration  were
used. The applicability of these costs to applications of this
technology at other sites is limited by the highly specific
nature  of each application, regional and climatic issues,
and the differences in regulatory requirements from state
to  state.   Therefore,  labeling  these  cost figures  as
"order-of-magnitude" estimates is appropriate.

When considering the cost for a commercial application of
a SRWCGT system, one should recognize that public and
private landowners establish tree biomass for numerous
reasons.  Some establish tree biomass as a source of profit
from generating fiber, pulp,  timber, and  fuel.  Others
establish tree biomass to restore degraded riparian areas
in rivers and streams.  Still others establish tree biomass to
phytoremediate groundwaterand soil, which is assumed for
the hypothetical model site.

Just as the motives to  establish tree biomass differ, the
prices associated with tree biomass establishment can also
vary markedly from one group to another. To date the
                                                    19

-------
Table 3-1. Estimated Full-Scale Costs for a 200,000 Square Foot Hypothetical Phytoremediation Model Site
Category
(1) Site Preparation
Data Review
Additional Well/Piezometer Installations
Pre-lnstallation Characterization
Ground Preparation
Tree Planting
Irrigation System Installation
Miscellaneous Site Preparation Tasks
Total Subcost
2. Permitting & Regulatory Requirements
Permits
Reporting
Total Subcost
3. Capital Equipment
Central Main Data Logger (1 unit)
Multiplexers (3 units)
Main Telemetry System (1 unit)
Pressure Transducers (10 units plus cabling)
Soil Moisture Probes (18 units)
Sap Flow Probes (32 units) with Data logger and Telemetry System
Weather Station (1unit) with Solar Panel and Batteries
Groundwater Sampling Equipment (Pumps, Water Quality Meters)
Total Subcost
4. Startup and Fixed Costs
Total Subcost
5. Consumables & Supplies
Irrigation System Materials
Fertilizer and Soil Conditioners
Herbicides & Pesticides
Trees (960)
Tool Shed
Ancillary Supplies
Total Subcost
6. Labor
Ground Maintenance
Annual Monitoring and Sampling Activities
Total Subcost
7. Utilities
Cellular Service
Water Usage
Total Subcost
8. Effluent Treatment & Disposal
Total Subcost
9. Residual and Waste Shipping & Handling
Contaminated Soil Disposal
Total Subcost
Subcosts
$2,500
$24,000
$5,000
$3,700
$2,500
$4,250
$1 ,200
$42,650
$5,000
$50,000
$55,000
$2,750
$1 ,500
$1 ,650
$18,000
$6,000
$3,593
$3,000
$1 ,240
$37,833
$3,783
$2,000
$3,000
$2,000
$480
$2,000
$1 0,000 ($1, Odd/year)
$19,480
$28,000
$80,000
$108,000
$12,000
$900
$12,900
$0
$7,500
$7.500
% of Total
Costs
9.1%
11.8%
8.1%
0.8%
4.2%
23.2%
2.8%
0.0%
1.6%
20

-------
Table 3-1. Estimated Full-Scale Costs for a 200,000 Square Foot Hypothetical Phytoremediation Model Site (Cont'd).
Category
10. Analytical Services
Pre-lnstallation Characterization Samples
Annual Monitoring Sampling (10 Years)
Total Subcost
11. Maintenance & Modifications
Irrigation System Repair
Monitoring System Repairer Replacement
Total Subcost
12. Demobilization
Well Abandonment (5 wells)
Total Subcost
Estimated Total Cost for Model Site
Subcosts
$38,455
$134,400
$172,855
$1,000
$4,000
$5,000
$1,050
$1,050
$466,051
% of Total
Costs
37.1%
1.1%
0.2%

prices charged to establish phytoremediation biomass are
significantly  more than the  prices associated with  the
establishment of biomass for profit fiber and fuel or riparian
restoration   biomass.   Factors  influencing  prices  for
establishing  biomass for  phytoremediation are: planting
techniques employed; depth to groundwater; site specific
preparation factors; and, perhaps, the potential customer's
lack of familiarity with forestry and agronomic practices and
techniques.

Prices can vary markedly on a per tree basis.  What one
phytoremediation vendor charges for a single tree may be
equal to what another vendor charges to establish several
hundred trees of the same or similar genus.  One also has
to remember that this price  disparity is  for establishing
biomass. It  doesn't take into consideration  additional
phytoremediation  requirements  such  as  establishing
monitoring   wells,  groundwater  chemical  analysis,
hydrological  studies, or  the  preparation of  reports  to
regulators.

Also,  it should  be kept in mind that the price asked to
perform a given task is often not synonymous with  the
actual cost to perform that task. The true cost to complete
a given task is often closely guarded and not readily shared
with anyone.   Cost  information  on a county  basis
throughout the United Sates is available foranyone wishing
to establish short  rotation woody crop biomass for profit
from the Department  of Energy's  Oak  Ridge National
Laboratory Biomass/Biofuels Program (see Appendix A).
The costs associated with the establishment of a riparian
biomass  can be  found in a chapter written by Berlin
Anderson in a book entitled The Restoration of Rivers and
Streams  - Theories and Experience edited by James A.
Gore (1985).
The phytoremediation system proposed for the model site
was designed with the intent to provide not  only plume
containment but residual contamination source removal.
The upgradient  portion of the model system would be
installed over any pockets of residual contamination. This
economic analysis was performed with the understanding
that an existing  hydraulic control/treatment system (i.e.,
pump and treat, groundwater interception system,  vapor
extraction/airsparging) would coexist and remain operative
on site until the phreatophytes begin to have a substantial
hydraulic affect on the site (i.e., the 3rd or 4th season after
planting). By this time the trees would  begin to exert  a
measure of hydraulic control; thereby, reducing the mass
flux of contamination in the shallow aquifer beneath the
planted   zones.  Costs  associated  with  any  existing
remediation systems were not considered in this economic
analysis. It is also assumed that  these technologies in
combination with other measures have addressed the bulk
of contamination at  the  site,  leaving  only  pockets  of
residual contamination in the vicinity of the former source
area.  The contaminant source area for the hypothetical
model is a former solvent disposal trench. The following
basic  assumptions regarding the hypothetical model site
have been made:

      Groundwater  contamination consists  chiefly  of
      aqueous phase TCE.
      A drip irrigation system would be required forthe first
      few seasons until the tree roots become  established
      in the shallow aquifer.
      The remediation time-frame would be 10-years.

This economic analysis only presents the costs estimated
for the hypothetical model site. A breakdown  of actual
costs  incurred during the 3-year SITE demonstration are
not presented in this economic analysis since these costs
                                                    21

-------
are of little  value  to the end-user given the research
oriented nature of the study.  Many of these costs and cost
categories are  more inflated than would  normally  be
expected due to a greater  amount of people involved,
higher labor rates of  the engineers  and  scientists
performing the installation, maintenance and monitoring,
frequent out of state travel and lodging expenses, and a
more extensive analytical and monitoring program. When
applicable, some of these costs and experience gained
during the SITE demonstration were used  to estimate
categorical  costs for  the  model  site.   Some  of the
assumptions made for the purpose of costing the model
site were based upon experience gained during the SITE
demonstration.  Most of the  costs experienced during the
SRWCGT demonstration were adjusted down forthe model
site to make them more  representative of the  costs
associated with a  commercial  application.  Factors that
influence the costs  associated with a phytoremediation
application   of  a  SRWCGT  system  would   include
contaminant type and concentration, total treatment area
which factors into the numberof trees required, dimensions
of  the  groundwater  contaminant  plume,  hydraulic
framework of the site, treatment goals, climate, and soil
properties,  including dominant lithology,  fertility, soil
moisture, and permeability.

Recent research has suggested the potential of poplar
trees  to exert a substantial hydraulic effect on  shallow
groundwater systems,  induce  reductive dechlorination
processes both  in the rhizosphere and the  tissue of the
tree, and withdraw and evapotranspirate groundwater and
contaminants directly to the  atmosphere. In addition, the
use of higher plants for remediation has gained the  support
of government agencies and the private sector in recent
years  because of  its  low  cost compared  to  that of
conventional technologies.

3.2   Conclusions

      The  cost  to  demonstrate  and  validate  the
      phytoremediation of TCE in the shallow groundwater
      at the Carswell MAS Golf Club over a projected 10
      year period is estimated  to be $1,600,000. Costs
      were  based  upon  two treatment plots oriented
      perpendicular to the direction of groundwater flow
      and measuring 12,500 ft2 each, each plot consisting
      of  two different  types  of Eastern Cottonwoods
      (Populus Deltoides), a  tighterthan normal spacing to
      accelerate hydraulic capture of the shallow aquifer in
      consideration of the abbreviated evaluation period,
      and a total of 660 trees. The costs  were also based
      upon   information  collected  over  the  3-year
      remediation period.   It should be noted that the
      majority of costs with  the Carswell MAS Golf Club
      Phytoremediation Demonstration were for extensive
      technical  support,  reports,  analytical  program,
      posters,  papers,  and presentations  to  validate
various  changes in the  geochemistry, tree  water
usage, and groundwater hydrology.  Costs  at an
actual phytoremediation site would be lower. Under
ideal site conditions the economics of short rotation
woody crops  coupled  with the costs of long term
monitoring  similar  to  that conducted  for natural
attenuation will result in costs well below those at the
Carswell MAS Golf Club.

If a site is conducive to short rotation woody crop
forestry techniques, serious consideration should be
given to the methods and techniques developed over
a  period of  thirty years  by the  Department  of
Energy's  Oak  Ridge  National  Laboratory
Biomass/Biofuel Program.  When large acreage of
tree  biomass  is required  to  accomplish a  given
phytoremediation objective, a cooperative forestry
agreement with a local wood burning power plant or
pulp mill should be explored as a means to offset the
majority  of the cost of establishing the biomass.
Cooperative forestry ventures enable landowners to
let another party grow a short rotation woody crop of
trees on their property in exchange for a portion of
the revenue (typically 40-45%) generated by the sale
of the biomass.

The total cost to remediate residual contamination at
the hypothetical model  site  and  attain hydraulic
influence was estimated to be $466,051. The model
site  also  consisted  of  two  plots  orientated
perpendicular to the shallow aquifer flow direction
and  measured 48,000 ft2 each,  a tree spacing
pattern of 10 feet, a total of 960 trees and a 10 year
remediation period. As one increases the acreage of
biomass established,  the   cost  per  acre   to
phytoremediate shallow  groundwater should also
decrease accordingly.   The  long term technical
support  and  reporting   costs  of   most
phytoremediation projects will exceed the costs to
establish the  necessary biomass.  Small sites  will
have essentially the  same technical support and
reporting  requirements  as  larger  sites.    The
documentation  of   biomass  influences  on
groundwater  chemistry  and  hydrology  and  the
preparation of reports to regulators will be the largest
cost component of a SRWCGT system. Once the
trees mature  and  reach  their operational potential
(hydraulic influences and enhanced biodegradation),
the remedial project manager can petition regulators
for less stringent long term monitoring.

Forthe hypothetical model site analytical (37.1%)
followed  by   Labor  (23.2%)  were  the  most
predominant cost categories.
                                                    22

-------
3.3   Factors Affecting Estimated Costs

The  design,  installation, monitoring and  maintenance
requirements for a tree based phytoremediation system is
highly site specific. As a result, a number of factors could
affect overall cost. These factors might include, but are not
limited to:

      Total Treatment Area
      Distribution and Magnitude of Contamination
      Climate
      Hydraulic Framework  of the Site
      Physical and Chemical Properties of the Soil
      Treatment Goals

The total size of the treatment area would logically  factor
into the  number of trees needed, the amount of time
required to install the system  (ground preparation activities,
installing an irrigation system, planting the trees,  installing
system monitoring stations), the amount of nutrients, soil
conditioners, mulch, pest and disease control substances,
the volume of water consumed  for irrigation purposes, as
well  as  the  man-hours needed  to  perform   periodic
maintenance tasks.

The  distribution of  contamination would  determine the
placement,  alignment   and dimensions  of  the  tree
plantations.  If the objectives of the project are mainly to
reduce  the mass  flux  of  groundwater  contamination
transported across the planted areas through hydraulic
control,  then  it would only be necessary to place the
plantations in a  position   enabling  them  to  intercept
contaminants released from the most downgradient source.
The type of contaminant and magnitude of contamination
(assuming it is a halogenated species as was the case at
the Carswell MAS Golf Club) would factor into the type of
tree chosen and the overall time needed to remediate the
site.  Some species of trees are  known to be more tolerant
to higher  concentrations  or  to  specific  chemicals.
Availability of these trees may factor into cost.

Climatic factors, such as the start and length of the growing
season,  annual precipitation and the amount of solar
radiation would control the amount of time during the year
that the  trees exert  a hydraulic control  on the  aquifer,
biologically enhance subsurface conditions, and remove
contaminants  via  evapotranspiration.   Climatic factors
would also determine the need for an irrigation system
during drought conditions (i.e.,  augment the aquifer and
prevent  the trees from dying).  Shorter growing seasons
could lengthen the time needed to reach remediation goals.

The hydraulic framework of  the site (i.e., aquifer size and
yield, groundwater velocity  and flow direction,  depth  to
groundwater, aquifer thickness, homogeneity and grain
size of aquifer materials) should be used as a guide when
deciding upon tree density, plot size, and  number of plots
needed.  Hydraulic conditions at the site would also control
the time  needed for the  trees to reach full hydraulic and
transpirational potential eithershortening or lengthening the
time the system starts to have a significant hydraulic impact
on the site. Although research has shown that hydraulic
control  is  the  principle  mechanism  responsible  for
reductions in the mass flux of contamination transported
across the planted area during the early stages of tree-
based treatment, other mechanisms, especially microbially
mediated reductive  dechlorination may become just as
prominent afterthe third or forth season.  In fact, reductive
dechlorination  might be the most important mechanism
operating during the dormant season.

The physical and chemical properties of the soil would
include  soil  moisture  retention,  soil moisture  profiles,
drainage, infiltration rates which would determine the need
and design of an irrigation system to help jump start the
trees. These soil properties will also determine the need
for providing some type of groundcoverthat would force the
trees to seek out the aquifer as a source of water rather
than becoming dependent on rainwater infiltrate. Othersoil
properties that have the potential for impacting cost would
be nutrient availability and the organic content of the soil.
This would determine the amounts of fertilizer and soil
conditioners needed overthe course of the project and also
effect the maintenance schedule, possibly increasing the
amount of man-hours needed.

Remediation goals would be site specific. Certain  goals
may be based  upon  specific soil and/or groundwater
cleanup criteria or based upon  a site specific receptor risk.
Remedial goals  at a site may fall into two categories:
source removal or source control. Whatever the remedial
goals might be, certain  design  features and the  time
needed to effect  the necessary changes would ultimately
affect total cost.


3.4  Issues and Assumptions

This  section   summarizes  the  major  issues   and
assumptions used for calculating costs for using a similar
phytoremediation system at a  hypothetical model site.  In
general, assumptions are based on information provided by
the developer and observations made during this and other
SITE demonstrations projects.

3.4.7  Site Size and Characteristics

This economic analysis  assumes that an area wide site
characterization had already been performed as part of a
remedial investigation  or its equivalent, and that only a
series  of  limited but  highly  specific  hydrogeological,
geochemical  and  waste  characterizations  would  be
performed as necessary to  assist with design parameter
decisions, establish appropriate site preparations methods,
and determine maintenance tasks and schedule.

For the purpose of conducting this economic analysis, the
conceptualized model site for a commercial application of
the tree-based phytoremediation system will have a total
                                                    23

-------
treatment area of 200,000 ft2 or roughly 4.6 acres. Surface
topography would generally be flat.  Current vegetation
would consist of several mature deciduous trees. Ground
cover would  consist mostly of grass with  a few  bare
patches.  The model site would be accessible via paved
roads.  Electrical and telephone services and a metered
potable water source would also be available. The source
of contamination at  the site has been linked to a former
trench that the facility once used to  dispose of various
waste solvents. The trench formerly occupied an area of
7,500 ft2  on the  north end of the property.  TCE is the
principal  contaminant of concern at the model  site. The
bulk of solvent-based contamination at the site has already
been  addressed by another  remediation system  (e.g.,
pump and  treat, steam enhanced vacuum extraction).
Residual  amounts of contamination still occur in pockets in
the vicinity  of the  former trench. These pockets  of
contamination continue to be a  source  of groundwater
contamination at the site. Concentrations on the order of
several thousand micrograms per  liter still occur in the
groundwater  in the  vicinity of the former trench.   The
concentrations decrease by an order of magnitude 500 feet
downgradient of the  source area.

The surface soil across the site is assumed to be a very
compact  12 to 18 inch layer of silty  clay to clayey silt.
Infiltration is  generally  poor accept along a network of
widely spaced desiccation cracks that occurthroughout the
site. Depth to groundwater is generally 10 to 15 feet BLS
across the site. Aquifer materials are being assumed to
consist predominantly  of silty fine sands  with  a few
hydraulically  isolated  lenses  of  coarser  material.  A
hypothetical conductivity  (k) value  of 10"2 cm/s is being
assumed for this exercise along with a porosity value of 35
percent.  Shallow aquifer thickness is being set at 5 feet
producing an estimated aquifer water volume of 2,618,000
gallons.  The maximum hydraulic gradient across the site
is 2.20 percent with a principal groundwater flow direction
to  the south.  Groundwater velocities for the model site
have been  estimated at 0.62 feet/day or 226 feet/year.
Groundwater flow across a cross-sectional slice in the
upgradient portion of the site  has been estimated to be
around 9,300 gpd.

3.4.2 System Design and Performance Factors

The goal of the  tree-based phytoremediation approach
designed for the model site is two  fold: remove residual
contamination in the subsurface near the former trench,
and reduce the mass flux of solvent based contamination
in  the upper  aquifer through a combination of hydraulic
control   and   in-situ  microbially  mediated   reductive
dechlorination.  Based  upon  the  type  and  levels  of
contamination persisting  at the  site,  it is assumed that
hybrid poplar trees would be used at the site. The species
selected would be native to the area, possess a tolerance
to the levels of chlorinated ethenes found at the site, have
a fairly long life-span, have some drought tolerance, and
have a natural resistance to pests and disease. As with the
Carswell MAS  Golf  Club site,  poplars  also  have  the
advantage  of fast growth, high  transpiration rates, and
phreatophytic properties.  A sufficient number of these
trees  would need to be planted  in a series of plots to
address a calculated volumetric flux of 9,300 gpd entering
the upgradient portion of the treatment area. The design
should also have enough reserve capacity built into it so as
to be  capable of handling twice the calculated volumetric
flux. Based upon a conservative pertree uptake rate of 20
gallons pertree perday (gptpd) (uptake rates as high as 40
gptpd have been  reported for mature hybrid poplars on
very hot days), a minimum of 466 trees will be needed to
handle the calculated flux of groundwater entering  the
system.

The model site will have 960 trees divided  evenly between
two 120 by 400 foot plots positioned perpendicular to the
direction of groundwater flow and separated by a 100 foot
buffer zone. Figure 3-1 depicts the layout of the model site
used in the economic analysis. Each plot will consist of 12
rows of trees planted 10 feet a apart.  Each row will have
40 trees. The upgradient plot will be positioned over the
former trench  area so as to biologically enhance  the
subsurface environment in a manner that promotes the
reductive dechlorination of residual chlorinated ethenes.

3.4.3  System Operating Requirements

The benefit of using a system like phytoremediation is that
it  only  requires minimal attention once the trees  are
planted, resulting in an O&M cost savings.  The technology
has been described  as a solar  powered  pumping and
filtering system that operates on its own. Phytoremediation
systems also requires minimal capital investment. Capital
expenditures tend to be limited to monitoring instruments.
The purchase cost for some monitoring equipment (e.g.,
Sap Flow Monitoring System, pumps and water quality
meters) can be spread  out over as  many as  10 other
projects. Otherequipment(e.g., data loggers, multiplexors,
weather station) will be dedicated to just  one project and
likely become obsolete at the end of the treatment period.
Periodic maintenance is required to clear and replant dead
trees, remove broken branches, prune healthy trees, apply
pest and  disease  control substances as needed, add
fertilizer and make repairs to the irrigation system and
monitoring system.

The hydraulic influences of the system are limited to the
growing  season  which   can  vary  depending  upon
geography. In most climates the growing season refers to
the period  between April and September.  The period
between October and March represents the dormant period
when  the trees temporarily stop pumping groundwater.
Individual  trees can begin affecting the  shallow  aquifer
systems as early as 1 year after planting.  Special planting
procedures and root training methods using drip irrigation
can be used to encourage young trees to seek out water
                                                    24

-------
                                                     - 400 Feet •
                        Weather
                         Station
                                                    Upgradient Zone
                                              (^For
Former TCE Source Area
                                                    Plantation #1
                                                     480 Trees
rea^>
                                               Buffer
                                                                Zone
                                                     Plantation #2
                                                      480 Trees
                                                  Downgradient Zone
                                          A
                                                                                   500 Feet
                                                                                    Direction of
                                                                                   Groundwater
                                                                                      Flow
                            New Groundwater Monitoring Well

                            New Piezometer

                            Soil Moisture Probe Nest
                    Figure 3-1. Layout of 200,000 square foot hypothetical model site for cost analysis.
from  the aquifer rather than  infiltration from  rainfall;
however, tree  roots will  reach the water table without
special  planting procedures or root training  methods. In
most  cases  it takes  3 to 4  growing  seasons  before
individual trees reach their full transpirational potential.  It
may take up to an additional 10 years, after this milestone
for  the  system to  achieve  final  remediation  goals
established for the site.  Removal of chlorinated ethenes
from the subsurface may be accomplished through several
mechanisms including  enhanced bioremediation in the
rhizosphere due to the release of various plant exudates
through  the  root system  resulting  in a  process called
reductive  dechlorination,  or  direct  uptake   of  the
contaminants through the root system and release to the
          atmosphere via evapotranspiration.

          3.4.4 Financial Assumptions

          All costs  are  presented  in  2001  U.S. dollars  without
          accounting for interest rates, inflation or the time value of
          money. Insurance and taxes are assumed to be fixed costs
          lumped into "Startup and Fixed Costs" (see Section 3.5.4).
          Any licensing fees paid to a developer, for using proprietary
          materials and implementing technology-specific functions,
          would be considered profit. Therefore, these fees are not
          included in the cost estimate.
                                                       25

-------
3.5   Results of the Economic Analysis

Using the general assumptions  already discussed, the
results of the economic analysis of the SRWCGT process
are presented in Table  3-1.  These assumptions are
discussed in more detail by cost category below.  Unless
otherwise specified, information presented in the following
sections focuses on issues and costs related to the model
site.

3.5.7  Site Preparation

Costs associated with Site Preparation have been divided
up  into  six  (6)  subtasks:  Data  Review,  Additional
Monitoring Well/Piezometer Installations, Pre-installation
Characterization   Studies,  Ground  Preparation,  Tree
Planting, Irrigation System Installation, and Miscellaneous
Site Preparation  Tasks.

Data  Review  -  Successful application of a tree-based
phytoremediation system requires  careful planning to
ensure  that  the  contamination  will  be  adequately
remediated  and  hydraulic  control can   be  achieved.
Planning would begin with a thorough review of existing
data sources which would include any number of reports
generated for the site as the result of other environmental
investigations (i.e., Remedial Investigation/Feasibility Study
(RI/FS) Reports,  Record of Decisions (RODs), Preliminary
Assessments (PA), Corrective Action Reports, Remedial
Design Reports,  Environmental Impact Statements). For
the purpose of this economic  analysis, it is assumed that
the model site  has already been extensively investigated.
The purpose of the data review is  to identify potential data
gaps as they pertain to  the design and operation of the
phytoremediation system.  The estimated cost  for data
review is $2,500. This cost was based upon a project
scientist  billing out at $50/hr spending about 50 hours
researching existing literature  and identifying data gaps.

Additional Monitoring Well/Piezometer Installations - For
the model site  it is assumed that the existing well network
is inadequate for providing all the monitoring needs for the
project. It is  assumed that the model site already has 15
existing wells.  An additional  5 monitoring wells  and 10
piezometers will  require  installation to more adequately
define hydraulic  gradient, variations in aquifer thickness,
zones of higher  permeability, depth to groundwater and
hydraulic conductivity. Although some sites may require
fewer wells/piezometers, it would  be a rare case indeed to
have a site that required no additional wells/piezometers.
The cost for  drilling, installing  and  developing these
additional monitoring wells/piezometers at the model site is
estimated at $24,000.  The subcontract cost per 6-inch
diameter well and piezometer is estimated  at $2,800 and
$800  respectively.  It is  assumed  that monitoring wells
would require the use a truck-mounted drill rig for drilling
and installation.  The less expensive GeoProbe® System
could  be  used  to install the piezometers.  The total
subcontract cost  associated with this subtask is estimated
at $22,000.  Labor associated with subcontract oversight
and the collection of 30 soil samples during  drilling  is
estimated  at $2,000. This estimate is based upon a mid-
level geologist billing out at $50/hr working a total of 40
hours (5 days @ 8 hours/day).

Pre-installation Characterization Studies- A numberof pre-
installation characterization  studies may  need to  be
conducted to address data gaps identified  during the data
review subtask.  Data gained from these studies would
contribute  to decisions concerning the type of tree that
should be  used, planting density, the  total number of trees
needed to achieve hydraulic control,  the number, position
and dimensions of the tree plots, the need for specialized
planting procedures, the need for a drip irrigation system,
and  the  types   and  amounts   of  fertilizer  and  soil
conditioners. The types of studies conducted are  highly site
specific and might include:

      Aquifer testing of existing and new wells to better
      define  the  hydraulic  properties  (i.e,  hydraulic
      conductivity, aquifer transmissivity,  hydraulic yield,
      hydraulic connectivity) of the contaminated aquifer
      beneath the treatment plots
      Groundwater and soil  sampling to better define
      certain geochemical and physical properties, such as
      dissolved oxygen, redox potential, macro  and micro
      nutrients, pH, conductivity, particle size distribution,
      soil  moisture, plus evidence of intrinsic  biological
      activity  and   reductive  dechlorination  in  the
      rhizosphere of native trees. Sampling will also better
      define the extend and magnitude of contamination in
      the areas of the proposed tree plots.
      Evapotranspirational   studies   (Sap   Flow
      Measurements and  root biomass studies) can  be
      conducted on several species of existing trees in the
      study area to evaluate current removal of water from
      the aquifer (saturated zone) and provide a means of
      estimating  upper-bound levels of transpiration that
      may be attainable  by the proposed tree-based
      phytoremediation system at the model site.
      Tissue samples (i.e., leaves, stems and roots) can
      be collected from several species of existing trees in
      the  study  area  to analyze contaminant  uptake  in
      plant organ systems and the potential for  metabolic
      transformations.

The  estimated  cost  for the  proposed  pre-installation
characterization  study at the  model site is  $5,000. This
value represents labor costs associated with the  purging of
monitoring well prior to sampling,  the collection of water
level  measurements,  the  collection  of  groundwater
samples from existing and new wells, the collection of tree
tissue samples from existing trees and the recording  of
various field measurements needed to fill some of the data
gaps.  Groundwater sampling associated with this pre-
installation characterization would be limited to just the new
monitoring wells/piezometers and  10 existing monitoring
                                                     26

-------
wells for a total of 25 wells sampled.  It is assumed that
some of the existing wells will be of little value to this
present study as a consequence of either their location or
design. Water level measurements would be obtained from
all site wells (45 total). Tree tissue samples will be obtained
from 12 to 13 existing trees resulting in a total of 25 tissue
samples collected. The $5,000 estimate is based upon two
junior level scientists billing  out at $50/hour working 10
hours per day over a 5 day period. It is assumed that the
pre-installation characterization subtask as with the drilling
oversight  work would be  staffed from a local  office;
therefore, no travel/lodging costs have been included. The
off-site analytical  costs associated with  this subtask  are
presented in Section 3.5.10, Analytical Services.

Ground Preparation - It is assumed that ploughing and
discing will be necessary in the  areas designated for tree
planting  to  facilitate fertilizer  infiltration,  increase soil
porosity, ease planting and foster vigorous root growth.
The appropriate  types  and  amounts  of  nutrients and
conditioners  (i.e., organic matter, drainage-enhancing
media,  etc.) will  be mixed  into  the soil  at  this  time.
Selection and application rate of these materials would be
based  upon the  results of  geochemical and physical
analyses  conducted on  model  site  soils during  the
additional site characterization studies. The plots will also
be ripped and/or trenched to  facilitate the planting of the
trees and setting the piping for the  irrigation system. Costs
associated with this subtask are comprised of labor and
equipment rental  fees.  Based upon the size of each
plantation, and experience gained at the Carswell MAS
Golf Club demonstration,  it is estimated that ground
preparation  activities would  take  around 5 days. Labor
associated with ground preparation  activities  has been
estimated at $1,250. This figure was based upon using a
technician billing out at $25/hour and a work day estimate
of 10 hours.  Discing, ploughing, ripping and trenching will
be   accomplished   using  equipment  rented  locally.
Ploughing and discing will likely be accomplished with at
tractor.  The  tractor and the plough will likely be needed for
5 days at a rate of $1,500 per week. The disc attachment
will probably rent out at $200 dollars a day and will only be
needed one day. The walk behind trencherwill probably be
rented  for a  week at $750/week.  It will  likely be needed
again for installing the irrigation lines. Total rental costs for
ground  preparation activities  at the  model  site  are
estimated at $2,450. Total  cost  for ground preparation
work at the  model site has been  estimated at  $3,700 or
approximately $804/acre.  This  estimate does  not reflect
costs associated with certain consumable items that would
be used during this stage (i.e.,  fertilizers, amendments).
These  consumables are presented  in Section  3.5.5.,
Consumables and Supplies.

Tree Planting - Data obtained from the pre-installation
characterization study would  aid decisions regarding  the
number, size, geometry and orientation of the tree plots as
well as tree planting density.  For the  model site, it is
assumed that 960 trees divided evenly between two 120 by
400 foot plots would be needed. Trees will be placed in
rips or trenches created to the desired  depth.   These
trenches would then be backfilled with a rooting mixture of
fertilizer, organic-rich soil, and other amendments. The cost
to plant the trees at the model site has been estimated at
$2,500. This  cost only reflects the  labor associated with
planting the trees. Is assumed that two technicians billing
out at $25/hour working 10 hours per day for a total of 5
days would be sufficient to complete the job. The costs
related to the purchase of the 960 trees for the model site
are  presented  in  Section  3.5.5., Consumables  and
Supplies.

Irrigation System Installation - A drip irrigation system has
been costed into the model site to jump start the trees. This
subtask involves  the  installation  of  irrigation  system
components  (i.e.,  PVC mainlines  and sub-mains, drip
tubing  arrays,  emitters, valving,  backflow   preventers,
pressure regulators, filters, end caps),  any trenching,
staking and testing  of the system. Costs associated with
the installation of an irrigation system at the model site are
comprised  of  labor   and   equipment  rental  costs.
Components  of the irrigation system are priced separately
in Section 3.5.5., Consumables and  Supplies. Based upon
the layout and size or the tree plots at the model site and
experience  gained  at  the Carswell  MAS  Golf Club
demonstration, it is assumed that installation activities will
take 7 days. Labor costs associated with the installation of
the irrigation  system at the model  site are  estimated  at
$3,500. This cost is based upon two  technicians with a
$25/hour labor rate working 10 hour days throughout the 7
day installation period. The costs associated with rental of
the trencher are based upon a  weekly  rate of $750. Total
costs for the irrigation system installation at the model site
are estimated at $4,250.

Miscellaneous Site Preparation Tasks - Miscellaneous
tasks would include connecting to the facility's water supply
($1,000) and  installing a small lockable tool shed to keep
equipment  and supplies in when no other arrangements
can be made with the site owner ($200). The purchase cost
of the shed is listed in  Section 3.5.5.,  Consumables and
Supplies. Connecting to a facility's electrical power main is
estimated  to  cost in  the  range of $2000,  but  for this
analysis it  is assumed solar  panels  and rechargeable
batteries will be used to power all monitoring equipment (At
the Carswell  MAS Golf Club Site  electrical  power was
supplied thru  solar panels and 12 volt car batteries). Other
possible voluntary expenses, not included  in this analysis,
are renting  an office trailer equipped with a phone and fax,
and rental  of a portable toilet. The office/supply trailer
estimate was based upon a $500/month rental over a 10
year remediation period. Electricity would be needed  to
provide lighting,   air  conditioning and  heat  to the
office/storage trailer so this could be a significant expense
in places, like Texas, with long hot summers. The expense
of an  air-conditioned office trailer was considered at the
                                                     27

-------
Carswell MAS Golf Club Site  and dismissed.  Summer
fieldwork is inherently hot. United States Air Force, United
States Geological Survey, United  Sates  Forest Service,
and other support personnel working at the Carswell MAS
Golf Club during the summers months and record recent
droughts successfully employed simple light loose fitting
clothing, hats, cold drinks, and tarps to minimize heat
stress.

For this analysis, generic site preparation responsibilities
such as site clearing, demolition, grading, road building,
surveying, utility clearance, staging area construction, site
fencing, auxiliary facility construction  (i.e., storage area
building,  decontamination  facility)  and  main  utility
connections were all assumed to have been performed by
the property owner/manager.  None of these  costs have
been included here.

3.5.2  Permitting and Regulatory Requirements

Depending upon the classification of the site, certain RCRA
requirements may  have to be satisfied.  If the site is an
active  Superfund site  it  is possible that  the  technology
could  be  implemented under  the umbrella  of  existing
permits and plans held by the Potentially Responsible Party
(PRP) or site  owner. Otherwise, few permits will likely be
required to operate  a tree-based phytoremediation system
such as the one  proposed for the  model site. No permit
costs were experienced for the Carswell  NAS Golf Club
demonstration system. Certain regions  or states have more
rigorous environmental policies, and a number of permits
might be required. In  addition,  permit requirement and
associated permitting costs can change  rapidly. Certain
municipalities might require permits to construct or operate
the phytoremediation system.   It's possilbe  that  these
requirements might be waived considering the nature  of
this technology.  Permits might also  be required for the
installation and abandonment of monitoring wells. Permit
costs for the model site are being estimated at $5,000.

State and Federal regulatory authorities might require the
preparation and submittal of a series  of reports including
but not limited to a  Corrective Action Report, Conceptual
Design Reports oreven Environmental  Impact Statements.
The cost associated with preparing these reports has been
estimated at $50,000.

3.5.3    Capital Equipment

Capital equipment costs associated with implementation a
tree-based phytoremediation system would be comprised
entirely of field  instrumentation needed  to  monitor the
system. Most of the capital equipment  cost estimates
presented in this  economic analysis are based  upon
present day costs for various monitoring components and
knowledge gained from the Carswell SITE demonstration.
It has been assumed that many of the components of the
field monitoring system will be a one time purchase and will
have no salvage value at the end of the project. Given the
length of the proposed treatment period (10 years) much of
the equipment will either be obsolete (due to advances in
computer  technologies)  or  be  near the  end  of  its
operational usefulness (based on an estimating 10 year
life-span).  Many of the monitoring components will be
dedicated to this project alone (i.e., soil moisture sensors,
weatherstation, some data loggers, multiplexors, pressure
transducers)  and  involve permanent installations  (i.e,
weatherstation). The cost of some othercomponents could
potentially be spread out over 7 other projects (i.e., Sap-
Flow  Probes, Sap-Flow  Data  Logger  and   Telemetry
System, groundwater sampling  pumps, water quality
meters, electronic water level indicator).

As with any  project, monitoring equipment can vary  in
sophistication and cost. The amount invested in equipment
is ultimately a function of the quantity and quality of data
needed to support specific objectives. For purposes of this
analysis, most  of  the  monitoring equipment, with the
exception of the sap flow sensors, will be connected to one
central data logger (approximately $2,750 with software)
through three  multiplexers (approximately $500 each). The
central data logger will be connected to a telemetry system.
The telemetry system will allow the user the capability of
remotely accessing the data, performing  system checks,
and  reprogramming  the  data   logger if   necessary
(approximately $1,650).

For a tree-based phytoremediation system, such as the
one proposed for the model site, equipment would be
needed to monitor changes in water level across the site as
a means of assessing tree root mass development and
transpirational potential of the maturing trees. Continuous
water level data can be obtained through a series  of
pressure transducers placed in a number of wells, in this
case 15 wells and 10 piezometers. It is assumed that 10
pressure transducers would be used for the model site.
These transducers  would  be connected to a central data
logger which would be programmed to collect and record
water level measurements at set times over the course of
treatment. Water  levels  in the other wells  would  be
obtained manually  at regular interval  using  an electronic
water-level  indicator. Costs estimated for  the pressure
transducers, cable, and other related  equipment would be
$18,100 (each pressure transducer is approximately $810
plus approximately  $2.00  per foot of cable).  A lesson
learned at the Carswell NAS Golf Club was that float water
levels should  be avoided because tree roots in well casing
tend to hang water floats up and give erroneous water level
readings. Another reason to avoid water floats is that they
are often made of  carbon steel, which can  interfere with
geochemical measurements.

Soil moisture probes would also be used  to monitor
changes in soils moisture at  various depths.  These
instruments will likely be stacked at six locations to provide
an accurate profile of soil moisture content from surface to
                                                    28

-------
the top of the water table.  Three locations will be selected
in  each plot for soil moisture measurements. Three soil
moisture probes will be installed  at each location in a
shallow, medium and deep configuration. A total of 18 soil
moisture probes will be used for the model site. The soil
moisture probes, as with the pressure transducers, will be
connected to the same central data logger, which can be
remotely accessed and programmed. Total estimated costs
associated  with the soil probe  system would be $6,000
(each probe is approximately  $190 plus approximately
$0.70 per foot of cable).

A Sap Flow/Sap Velocity/Plant Transpiration system will be
used for measuring the transpiration rates and water usage
of the trees through each growing season.  The system
enables the simultaneous monitoring of up to 32-sap flow
sensors. The sap flow equipment cost (data logger, probes,
gauges, multiplexers,  cables, telemetry equipment, and
software) is estimated at $25,150. The cost for this item
can be spread out over 7 other projects. The adjusted cost
for this item at the model site is $3,593. Due to advances
made by the  United  States Forest Service's  Coweeta
Hydrologic  Lab  at the Carswell  MAS Golf Club  Site,
Orlando, Florida,  and Denver,  Colorado  and current
on-going efforts to improve the physiologically based tree
PROSPER  Transpiration Model, routine employment of sap
flow  device at phytoremediation  sites will most  likely
become unnecessary.

An on-site  weather station will  be  used  to  aid with
interpretations of transpiration rates. The weatherstation
will be  capable of measuring temperature,  pressure,
relative humidity, wind  direction, wind velocity, rainfall, soil
temperature, solar radiation.  A complete on-site weather
station would cost approximately $3,000 (including a solar
panel  and  rechargeable  batteries),   assuming  it  is
connected to the central data logger and telemetry system
mentioned earlier.

Periodic groundwater  sampling will require the use of a
groundwater sampling pump and  a water quality  meter
capable  of  providing  continuous  measurements  of
temperature,  conductivity,  dissolved  oxygen,  oxygen
reduction potential and pH. Groundwatersampling will also
require the  use of an electronic water-level indicator. It is
assumed that  groundwater sampling will  employ micro-
purge techniques.  As was the case with the Sap-Flow
equipment,   the cost of  the  groundwater  sampling
equipment  can be spread over 7 other projects.   It is
assumed that a simple  peristaltic pump, capable or running
off a car battery and costing around $1,200, will be used
throughout  the treatment period. The water quality meter,
which  will   have  data  logging  capability,  will   cost
approximately $7,000. An electronic water level indicator
will cost $479. The adjusted costforgroundwatersampling
equipment planned for the model site is $1,240.
3.5.4   Startup and Fixed Costs

From  past experience, the fixed costs for this economic
analysis are assumed to include only insurance and taxes.
They  are  estimated  as  10 percent of the  total capital
equipment costs, or $3,783.

3.5.5   Consumables and Supplies

Consumable  and  supply  items for  the  model  site
application would include plumbing supplies for the drip
irrigation system  (i.e., PVC mainlines and sub-mains, drip
tubing arrays, emitters,  valving, backflow preventers,
pressure regulators, filters, end caps), fertilizer and soil
conditioning  materials, mulch, pest and disease control
materials,  the trees, ancillary  supplies for monitoring
equipment (i.e.,  tubing  for peristaltic pump, tool shed),
miscellaneous expendable landscaping  supplies (i.e.,
rakes, shovels, pruners, garden sprayers, etc.) and health
and safety supplies.  Piping and fittings for the irrigation
system are estimated to cost $2,000 (with a 20% salvage
value). Fertilizerand soil conditionerconsumption is based
upon  a total tree plot area of 96,000 ft2 and 10 years of
treatment.   The  estimated cost for fertilizer  and soil
conditioners is $3,000.  The same assumptions used for
estimating the cost of fertilizer were used for estimating the
cost of pest and disease control materials.  Pest and
disease control materials are estimated to cost $2,000 over
the term  of treatment at the model site.  As previously
discussed, it is estimated that 960 trees will be needed at
the model site. Based upon an estimated purchase price
of $0.50 pertree (assuming volume discounts apply), total
tree cost has been estimated around $480.  Tree cost will
vary based upon geography and tree species. The tool
shed, previously  discussed in  Section  3.5.1,  will cost
around $2,000. Ancillary supplies formonitoring equipment
tubing, gardening supplies and health and safety supplies
are estimated at $1,000/yeartotaling $10,000 overthe term
of project.

3.5.6   Labor

Hourly labor rates include base salary, benefits, overhead,
and general  and  administrative (G&A) expenses. Travel,
per diem, and rental car costs have not been included in
these figures. Local travel  to the site is assumed for the
model site. If a site is located such that extensive travel will
be required, travel related cost would significantly impact
labor  costs.  Labor costs associated  with  a tree-based
phytoremediation system such as the one proposed forthe
model site would be limited to general ground maintenance
tasks  and monitoring and sampling events.

Ground maintenance tasks at the model site would consist
of the periodic  removal  of  dead  branches,  pruning,
replanting and clearing dead trees, weeding, grass mowing
and application of pest and disease control substances as
well   as  fertilizers.   Labor  associated  with ground
maintenance would likely be conducted monthly and occur
                                                    29

-------
primarily during the growing season.  It  is assumed that
ground  maintenance tasks would require a landscaper
working an 8-hour day for 1 day each month. In most
regions, ground maintenance will be required 7 months out
of the year. Assuming a landscaper labor rate of $50/hour,
ground maintenance labor for the term  of treatment (10
years)  is estimated at $28,000.  The amount of ground
maintenance ultimately required will be  a function of the
actual visibility of the site. Sites with higher visibility require
more attention  then remote sites. After the canopy of the
trees has closed, often the growth under the trees rarely
needs  cutting.   Another option  to  reduce long term
landscaping costs is to employ some form of shade tolerant
ground cover that requires little  or no maintenance.

Labor associated with  monitoring and sampling will be
reduced somewhat by the various data logging capabilities
of the  instrumentation installed at the  model  site. This
instrumentation will enable real-time remote access and
monitoring  of  information  pertaining  to tree growth,
hydraulic conditions and soil moisture.   Monitoring and
sampling  events  will  likely   involve  physical  tree
measurements (i.e., tree height, canopy width  and tree
trunk diameter),  additional water level measurements,
calibration  checks on  automated  monitoring  systems,
groundwatersampling and tree sap-flow measurements. It
is assumed that 1 -2 monitoring and sampling events would
be scheduled each year during the growing season. Each
event would require 2 people, working a standard  8-hour
work day, 5 days to complete. It is assumed that the tasks
associated with   monitoring  and sampling  would  be
accomplished by two junior level scientists billing out at
$50/hour.  Total labor costs associated with monitoring
and sampling are estimated at $4000 per sampling event,
or approximately $80,000 over a ten-year period assuming
two  sampling events per year.  Labor associated  with
groundwater,  soil   and tissue  sampling  during  Site
Preparation is presented in Section 3.5.1.

To reduce costs a project manager may want to consider
reducing the number of sampling events in the early years
as the  trees  establish  themselves.    Once anaerobic
groundwaterconditions and maximum hydraulic influences
are  established, the  remedial  project  manager might
consider petitioning the appropriate regulators  for a less
stringent monitoring program to reduce costs.

The  labor associated with the  other tasks, such as site
preparation,   maintenance   and  modification,   and
demobilization  have been assigned to other categories.
Analytical costs  associated  with monitoring/sampling
events  are  presented  in  Section   3.5.10., Analytical
Services.

3.5.7   Utilities

A major utility  cost for the project will be cellular phone
service for each telemetry system at the site. The model
assumes two telemetry systems with a monthly cellular
service  fee  of approximately $100 or  approximately
$12,000 over a ten-year period.

Another utility required forthis project would be water used
by the drip irrigation system.  The drip irrigation system
would  only  be  required   until  the roots reach  the
groundwater. It is assumed that the irrigation system would
only be  required for 2 years, but would be available to
augment the aquifer in situations of severe  drought.  Cost
associated with water consumption for the  model site are
estimated  at $900

No costs forelectrical usage is included, since solar panels
and  rechargeable batteries will  be  used  to power the
monitoring systems.

3.5.8   Effluent Treatment and Disposal

No  costs  were  assigned to this  category because the
transpirate from the trees is not regulated.

3.5.9 Residuals &  Waste  Shipping,  Handling, and
      Storage
It is assumed that as many as 15 drums will be needed to
dispose  of waste soil, drill cuttings and contaminated water
generated by  purging  and  drilling.  Based upon the
classification and disposal requirements for the types of
contaminants found in the subsurface at the  model site, the
cost to manifest, transport,  and dispose of these drums
was estimated at $500/drum. The total cost to dispose of
these drums is estimated to  be $7,500. Additional drums
might be generated for the disposal of contaminated PPE
items. It is assumed  that no more than 2  PPE drums of
PPE  contaminated  enough to  require  special  waste
handling and disposal would  be generated over the course
of treatment.  Disposal of these drums would be nominal
and therefore have not been included here.

3.5.10  Analytical Services

It was assumed that off-site analytical support would  be
needed  during any  sampling associated  with  the pre-
installation  characterization  study  and   during  each
monitoring and sampling event conducted  at the model
site.  As discussed previously, the purpose of  samples
collected during the pre-installation characterization stage
is to support decisions on tree type, plot placement and
dimensions,  number  of trees, planting density, fertilizer
schedule,  the  types  and amounts of soil conditioners
needed, and irrigation system design. Twenty-five (25)
groundwater samples would be analyzed for volatile
organic  compounds (VOCs),  ICP metals,  total  organic
carbon  (TOC), common ions, and  pH.  Thirty  (30) soil
samples would be analyzed  for VOCs, ICP metals, TOC,
pH, percent  moisture, porosity, particle size distribution,
nitrate-nitrites,  and phosphates.  Twenty-five  (25) tree
tissue samples  would  be   analyzed  for  VOCs.  Pre-
installation characterization analytical costs for the model
site are  estimated to be $38,455.
                                                    30

-------
Samples collected  for  off-site analyses  during  each
monitoring  and  sampling  event would  consist  of  15
groundwater samples per event. These samples would be
collected to  monitor changes  in  VOC  contaminant
concentrations  and  the spatial  distribution  of  VOC
contaminants  in  the groundwater.   Analytical   costs
associated  with  monitoring  and sampling  events are
estimated at $13,440 per year, assuming two sampling
events per year.  Total analytical costs for monitoring and
sampling events conducted over the  10  year treatment
period  are  estimated at $134,400.  Additional analytical
costs might be  incurred if the regulators  require soil
verification samples to be collected.

3.5.11  Maintenance and Modification

It is assumed that repairs will have to be made periodically
to the drip irrigation system.  The irrigation system may
have to be drained during the winter months to prevent ice
damage.  Estimated repair  costs for the  model  site's
irrigation system are assumed to be around $1,000. It is
also  possible that the weatherstation, soil moisture probes,
and data logger may  get damaged over the course of
treatment  due to grounds  keeping activities,  lightning
strikes, etc., therefore,  it is assumed that $4,000 would be
needed for replacement parts (see Section 3.5.6 Labor for
associated costs).

3.5.72 Demobilization

Demobilization of a plant-based phytoremediation system
would basically involve the proper abandonment of all
wells. Trees can most likely be left in  place unless and
arrangement has been made to harvest and sell the wood.
Well abandonment requirements vary from state to state,
as a result abandonment costs can vary as well.  Use of a
drill rig to abandon the 5 additional wells at the model site
would be approximately $250, and the charge for well
abandonment would be approximately $8 per foot.  This
price includes labor, materials, insurance,  and taxes.  The
five additional wells at the model site represent 100 linear
feet that  will  require  abandonment. The total cost for
demobilization is estimated at $1,050.
                                                    31

-------
                                             SECTION 4
                                TREATMENT EFFECTIVENESS
This  section  describes  the  effectiveness  of  the
phytoremediation system in controlling the migration of a
trichloroethene  (TCE)-groundwater  plume during  a
field-scale demonstration of the technology at a site in
Fort Worth, Texas.  Information provided in this section
includes:  (1)  site conditions  prior to treatment, (2)
implementation, and monitoring, (3) objectives, including
the methodologies  implemented  to achieve  these
objectives, and  (4)  results and performance, including
system reliability and process residuals.

4.1 Background

This field-scale demonstration was a cooperative effort
between the U.S. Air Force Aeronautical Systems Center
Acquisition, Environmental, Safety and Health Division
(ASC/ENV),  the  U.S.  Department  of  Defense
Environmental Security Technology Certification Program
(ESTCP),  the U.S. Environmental Protection Agency
(USEPA) Superfund Innovative Technology Evaluation
(SITE) Program,  and  the   U.S.  Geological Survey
(USGS).   The  overall purpose of this effort was to
demonstrate the  feasibility  of  purposefully  planting
eastern  cottonwood trees to  help remediate shallow
TCE-contaminated groundwater in a subhumid climate.
Specifically, the study was undertaken to determine the
potential for a planted system to hydraulically control the
migration  of contaminated groundwater,  as  well as
biologically enhance  the  subsurface environment to
optimize   in-situ  reductive   dechlorination   of  the
chlorinated  ethenes  present  (trichloroethene  and
cis-1,2-dichloroethene).  To assess the performance of
the system, hydrologic and  geochemical data were
collected over  a  three-year  period.   In  addition to
investigating changes in groundwater  hydrology and
chemistry, the trees were studied to determine important
physiological processes  such as water usage  rates,
translocation and volatilization of these volatile organic
compounds, and biological transformations of chlorinated
ethenes within the plant organs. Since planted systems
may require many years to reach their full  remediation
potential, the study also made use of predictive models
to  extrapolate  current  transpirational  hydrologic
conditions to future years. In addition, a section  of the
aquifer that underlies a mature cottonwood tree (~20 years
old) was investigated to  provide evidence of transpiration
rates and geochemical conditions that may be achieved at
the site when the planted trees reach full maturity.

The selected site is on the north side of the  Carswell Golf
Course (CGC) at the Naval Air Station Fort Worth (MAS Fort
Worth) about one mile from the southern  area of the main
assembly  building  at Air Force Plant 4 (Plant 4).  The
assembly building is the primary suspected source of TCE at
the demonstration  site.   Historically,  the  manufacturing
processes at Plant 4 have generated an estimated 5,500 to
6,000 tons of waste per year, including waste solvents, oils,
fuels,  paint residues, and miscellaneous spent chemicals.
Plant  4 is on the  National  Priorities List  and  is being
remediated  in   accordance  with   the  Comprehensive
Environmental Response, Compensation, and Liability Act
(CERCLA) as amended by the Superfund Amendments and
Reauthorization Act (SARA). TCE is believed to have leaked
from degreasing tanks in the assembly building at Plant 4 and
entered the underlying  alluvial  aquifer.   An  Installation
Restoration Program (IRP) was initiated in 1984 with a Phase
I Records Search by CH2M Hill (CH2M Hill, 1984). The U.S.
Army Corps of Engineers (USAGE) was retained in June of
1985 to further delineate groundwater conditions in the East
Parking Lot area of Plant 4; the Corps installed six monitoring
wells  as part of this  investigation  (U.S. Army Corps of
Engineers, 1986). Groundwatersampling in the East Parking
Lot area of Plant 4  continues for the purpose of monitoring
the TCE plume.  The plume has migrated in  an easterly to
southeasterly direction  under the East Parking Lot towards
the MAS Fort Worth.  The plume extends toward the east with
the major branch of the  plume  following a  paleochannel
under the  flight  lines  to the south  of the  Tree  system
demonstration site.   This finger of the plume  is being
remediated with  a pump and treat system.  Another branch
of the plume appears to follow a paleochannel to the north of
the demonstration site.  Data indicate that the TCE may have
entered the area of the demonstration site along an additional
finger of the plume.

Under  the  USEPA  SITE Program,  the Phytoremediation
system was evaluated  for its ability to reduce the mass of
                                                   32

-------
TCE that is transported across the downgradient end of
the site (mass flux). Specifically, the following primary
performance  objectives were established:  (1)  there
would be a 30 percent reduction in the mass of TCE in
the aquifer that is transported across the downgradient
end of the site during the second  growing season, as
compared to baseline TCE mass flux calculations, and
(2) there would be a 50 percent reduction in the mass of
TCE  in  the  aquifer that  is  transported  across the
downgradient end of the site during the third  growing
season,  as  compared to  baseline TCE  mass flux
calculations.  In  order to evaluate the primary claim,
groundwater levels were monitored and samples were
collected and analyzed for TCE concentrations over the
course of the study.

In addition to the primary performance objectives, several
secondary objectives were evaluated by a team  of
scientists  that were assembled  to  study  the  site.
Secondary  objectives  were   addressed  to   help
understand the processes  that control   the  ultimate
downgradient  migration  of  TCE in the  contaminated
aquifer, as well as to identify scale-up issues.  These
secondary objectives include:

    Determine tree growth rates  and root biomass
    Analyze tree transpiration rates to determine current
    and future water usage
    Analyze the hydrologic effects  of tree transpiration
    on the contaminated aquifer
    Analyze  contaminant   uptake  into  plant organ
    systems
    Evaluate   geochemical  indices  of  subsurface
    oxidation-reduction processes
    Evaluate   microbial   contributions  to  reductive
    dechlorination
    Collect data  to  determine implementation  and
    operation costs for the technology (see Section 3 -
    Economic Analysis)
4.2 Detailed   Description   of   the   Short
    Rotation   Woody  Crop  Groundwater
    Treatment System

In April  1996, the U.S. Air Force planted 662 eastern
cottonwood trees (Populus deltoides) to determine the
feasibility of such a planted system to attenuate a part of
the TCE-groundwater plume that is migrating beneath
the Carswell Golf  Course  north of Farmers Branch
Creek. The following sections discuss the rationale for
design decisions related to the Phytoremediation system
at the Carswell  Golf Course.  The monitoring systems
that  were employed  at  the  Carswell  site  are  also
discussed. Monitoring for this demonstration study was
more extensive than would be  necessary for an applied
remediation project because some of the data for this
demonstration were collected to help understand the
specific processes associated with a SRWCGT System.

4.2.7 Site Selection

Characterization  sampling for  site  selection and system
design was completed  in  January  of  1996.   Relative
groundwater elevations indicated that groundwater  in the
Terrace Alluvial Aquifer at the selected site generally flows
towards the southeast with an average gradient of just over
2 percent.  Depth to groundwater (at the time of sampling)
ranged from 2.5 to 4 meters (m) below  ground surface.
Aquifer   thickness  varied   between  0.5   to   1.5   m.
Horizontal-hydraulic conductivity values for the aquifer, as
determined from  eleven slug tests, range from 1 meter/day
(m/d) (1.2 x 10"3 centimeters/second (cm/s)) to 30 m/d (3.5 x
10"2 cm/s) with a  geometric mean of 6 m/d (7 x 10-3  cm/s).
Aquifer porosity, as  determined  in the laboratory,  is 25
percent.  Chemical analyses of the groundwater indicated
that TCE  concentrations ranged from 230 mg/L to 970 mg/L,
with  cis-1,2-dichloroethene  (cis-1,2-DCE) concentrations
ranging from 24 mg/L to 131 mg/L. Dissolved oxygen data (>
5mg/L) indicated that the aquifer  was well  oxygenated
(Jacobs Engineering Group Inc. 1996).  Furthermore, the
ratio of TCE to  cis-1,2-DCE from the sampling  locations
indicated   that   no  significant  reductive  dechlorination
(Chapelle,  1993) had occurred within the  selected site.
These data suggested that tree roots could reach  the water
table at the site and that the site would likely benefit from
processes that promote reductive dechlorination.

4.2.2 Site Characterization

The  eastern  cottonwood tree  (Populus  deltoides) was
selected for this study on the basis of a literature review, as
well as discussions with the  Texas  Forest  Service, the
National  Resources Conservation Service,  and  the U.S.
Forest Service  Hardwood  Laboratory.    In  summary,
cottonwoods were selected  due to their fast growth, high
transpiration rates, and  phreatophytic properties.  These
characteristics allow cottonwoods to rapidly transpire water
from a saturated zone and maximize below-ground biomass,
which is an important factor in establishing biogeochemical
reductive pathways.  Other factors that were considered
include: (1)  tolerance of cottonwoods to the contaminants of
concern,  (2) the  natural occurrence of cottonwoods  at the
selected site, (3) the perennial nature of cottonwoods, and (4)
the longevity of cottonwoods (40 -100 years).

4.2.3 Size and Configuration of the Tree  Plantations

Decisions related to the  size and placement of the  tree
plantations at the demonstration site were critical for ensuring
the success of the Phytoremediation system.  Factors that
were used  to determine  the size and configuration  of the
plantations  included the  general  direction of groundwater
flow, the extent of groundwater contamination, the volume of
groundwater that flowed through the selected site, and the
volume of groundwater stored in the aquifer beneath the site.

Two  rectangular-shaped  plantations   that  measure
                                                    33

-------
approximately 15 by 75 m were established (Figure 4-1).
The first plantation was  planted with whips, which are
sections of one-year old stems harvested from branches
during  the   dormant  season.    The  whips   were
approximately 0.5 m long at the time of planting and were
planted so  that  approximately  5  centimeters  (cm)
remained above ground. The second plantation, which
was 15 m downgradient, was planted with trees of 2.5 to
3.8 cm caliper (trunk diameter). The caliper trees were
just over 2 m tall at the time of planting. The two sizes of
trees  were selected for  inclusion in this study so that
differences in rate  of growth, contaminant reductions,
and cost based  on planting strategy could be compared.

The plantations were designed so that the long sides of
the plantations are  generally  perpendicular to the
direction of groundwater flow (Figure 4-1).  These long
sides   span  the most  concentrated  portion  of the
underlying TCE-groundwater plume. The length  of the
long sides of the plantations was constrained by logistical
factors, as well  as the experimental nature of the study.
The number of trees that were to be planted determined
the length of  short sides of the  rectangular plantations.
These short  sides are  parallel  to  the  direction  of
groundwater  flow.    The  following  information was
considered when determining the number of trees that
were to be planted:

Volume of Groundwater Flow (Volumetric Flux) Through
the Site.
The volumetric  flux of groundwater (Q) was calculated
according to Darcy's Law:
 OS-KiA
(Eqn. 4.2-1)
where K is the hydraulic conductivity of the aquifer, i is
the  hydraulic gradient  in  the  aquifer across  the
downgradient of the planted area, and A is the cross-
sectional area of the aquifer along the downgradient end
of the planted area.

Volume of Groundwater in Storage in the Aquifer at the
Site.

Volume of groundwater in storage was  calculated as
follows:

Aquifer Thickness x Study Area Size x Aquifer Porosity
    (Eqn.  4.2-2)

Data assumptions included the following:

    i = 2.25 percent
•   A = 75 m2
    Aquifer thickness is 1m
    Aquifer width is 75 meters
    The aquifer material is a medium sand with mean
    porosity of 23%.
    K (Horizontal hydraulic conductivity) =  6 m/d (7 x 10"3
    cm/s)

Using  equation  4.2-1   and  the  above  assumptions,
groundwater  flow (or flux) through the study area  was
calculated to be approximately 10,125 liters day1  (2,675
gallons day1). Using equation 4.2-2 and the site dimensions
listed in the preceding paragraph, the volume of water in
storage in the aquifer beneath the site was calculated to be
approximately 776,250  liters (205,060 gallons).    It  was
assumed that the trees would need to transpire a minimum
of 10,125 liters (2,675 gallons) of groundwater per day to
prevent contaminated  water from moving off site during the
growing season if no groundwater  were released from
storage.  A greater volume of water would need to be
transpired from the aquifer  if  water were released from
storage during the growing  season  in  response to  tree
transpiration.

According to Stomp (1993), a hybrid poplar tree occupying 4
m2 of ground can cycle  approximately 100 liters day1 (26
gallons day1) of groundwater under optimal conditions. As a
result, it was  determined that a minimum of approximately
100 trees would need to be planted at the demonstration site.
A total of 662 trees were actually planted.  Seven rows of
whips were planted approximately 1.25 meters  (4 feet) on
center in the upgradient plantation for a total of 438 trees and
seven rows of caliper trees were planted approximately 2.5
m (8 feet) on center in the downgradient plantation for a total
of 224 trees.  This is because the estimate of 100 liters day1
per tree is for optimal  conditions and field conditions at the
site may not always be optimal. It was also expected that
some trees would be  lost due to natural attrition caused by
poor planting, disease  and  insects.   In  addition,  it  was
anticipated that some transpired water would be derived from
intercepted precipitation, soil moisture or from groundwater
released from storage  rather than from groundwater flowing
into the site across the upgradient end.

4.2.4 Planting and Installation of the Irrigation System

The planting method used in this demonstration is similar to
the method used for short rotation wood culture. Whips were
obtained from the Texas Forest Service in Alto,  Texas; the
caliper trees  were obtained  from Gandy Nursery in  Ben
Wheeler, Texas.  Soil  preparation  for planting  included
trenching seven rows in each of the proposed plantations to
a depth of one meter. The whips or caliper trees were placed
within the trenched rows. Irrigation lines were also  placed
within the trenches. An agronomic assessment for macro-
                                                    34

-------
                                                                        WJEGTA523
                                                                               -1 ree
                                                                  )k
                 \>Go/f Can? Path
                  \\
                  \\
                           WJEGTA51£T\^JEGTA522 . .jHttffllilllip'" WJEGTA528


                                   ,."7>>.: .   "W>'\     .  WJEGTA525
WJEGTA512"
                                     WJEGTA510_
                                       *
                          "/- ii  rs Branch Creek
                                                                               Meters
             MONITORING WELL- Well number indcates well
             sampled throughout the entire demonstration

           •DMONITORING WELL WITH WATER LEVEL
             RECORDER
               * STREAM-STAGE GAGE
              + TENSIOMETER NEST
                                                        DDWEATHER STATION
        Figure 4-1. Short Rotation Woody Crop Groundwater Treatment System site layout.
and micro-nutrients and the presence or absence of hard
pans  was conducted.   The  need  for  fertilizer was
determined from the soil characteristics that were identified
through this  sampling and  analyses,  as well  as from
discussions with the Texas Forest Service, Tarrant County
Agricultural  Extension Service, and the Texas A&M
Horticulture Department.   A  handful  of slow  release
Osmacote  14-14-14  fertilizer was  applied around each
whip/caliper tree.  When  planting was completed, fabric
mulch and  10 cm of landscape mulch were placed along
each of the planted rows to reduce weed competition. This
was especially important for the newly planted whips.

4.2.5 Irrigation

A  drip  irrigation  system was  required  to  supplement
precipitation for the first two growing seasons. The trees
were watered liberally during this time to encourage deep
               root development.  Data from a precipitation gage at the
               site were used to help make irrigation decisions.  Because
               the roots were expected to intercept percolating irrigation
               water (Licht and  Madison,  1994), irrigation  was  not
               considered  to be  an additional  source of water to  the
               aquifer.

               4.2.6 Monitoring

               Because the processes associated with Phytoremediation
               systems require extended time  frames to develop,  the
               monitoring system had to be designed to measure small
               incremental changes in site  conditions over time.  The
               monitoring strategy for this demonstration study was more
               extensive than  would be  required for a  typical  Short
               Rotation Woody Crop Groundwater Treatment System
               project due to the  research  nature of the study.  Data
               collected from this intensive monitoring program were used
                                                     35

-------
to determine how well the system behaved overtime and
to develop models to predict future system performance.
The following monitoring  stations were employed in  the
study:

   sixty-seven  wells   installed   upgradient,   within,
   downgradient and surrounding the demonstration site,
   including the area under the mature cottonwood tree
   near the site
   continuous  water level recorders  installed in three
   monitoring wells, including one upgradient of the tree
   plantations and two within the planted area
   nine tensiometers installed upgradient  or within  the
   tree plantations
   a  weather  station  installed to  collect  site-specific
   climate data
   a stream gage installed on a creek adjacent to the site
   to record stream stage
   tree collars and  / or tree probes installed periodically
   during the  growing season to measure sapflow in
   selected trees

Figure  4-1 depicts the location of monitoring points with
respect to the tree plantations. A number of wells are not
shown  on Figure 4-1 because they are outside of the area
depicted in the  figure.  These wells were used to collect
groundwater level data surrounding the site for  use in
calibrating a groundwater-flow model of the area that could
be used to help predict out-year performance  of  the
Phytoremediation system.

4.3 Project Objectives
A SRWCGT System was studied to determine the ability of
a purposefully-planted tree system to reduce the migration
of chlorinated  ethene  contaminated  groundwater.   A
primary project objective and several secondary objectives
were established to provide cost and performance data to
determine the applicability and limitations of the technology
to similar sites with similar contaminant profiles.

4.3.7 Primary Project Objective

The primary objective of this technology demonstration was
to determine how effective the system could be in reducing
the  mass  of TCE in the  aquifer transported across  the
downgradient end of the  planted  area  (TCE mass flux).
The following goals were established:  (1) the trees  will
effect a 30 percent reduction in TCE mass flux across the
downgradient end of the study area in the second growing
season (1997), and  (2) the trees will effect a 50 percent
reduction in TCE mass flux across the downgradient end of
the study area in the third  growing season (1998).

It was  hypothesized that tree physiological processes
would  result in the  reduction of TCE  mass flux in  the
aquifer due to a combination of hydraulic control of  the
contaminant plume and in-situ reduction of the contaminant
mass  (natural  pump  and  treat).   Specifically,  it was
hypothesized  that the trees would remove contaminated
water from the  aquifer by means of their root systems,
followed by the biological alteration of TCE within the trees
or the transpiration  and  volatilization of TCE  in  the
atmosphere.  The trees would also promote microbially
mediated reductive dechlorination of dissolved TCE within
the aquifer.

To determine the mass of TCE transported in the aquifer
across the downgradient end of the planted area at a given
time,  the  volumetric  flux of  groundwater across  the
downgradient end of the site was multiplied by the average
of the TCE concentrations in a row of wells immediately
downgradient of the site (WJEGTA526 (526), WJEGTA527
(527), WJEGTA528  (528)) (Figure 4-1).  The volumetric
flux  of  groundwater  was  calculated  for  each  event
(baseline,  peak growing  season, late growing season)
according to equation 4.2-1 (presented in section 4.2.3).

The following assumptions applied:

    Horizontal-hydraulic conductivity was assumed to be
    constant  over the course of the study because
    measurements were made in the same locations. A
    value of 6 m/d was used and represents the geometric
    mean for the study area.

    The hydraulic gradient across the downgradient end of
    the  planted area  at selected times was calculated
    using groundwaterelevation data from monitoring wells
    522 and 529 (Figure 4-2). Well 522 is located between
    the tree stands near the center of the planted area.
    Well 529 is downgradient and outside the influence of
    the trees. These wells were chosen so that they did not
    reflect increases in the hydraulic gradient across the
    upgradient  end   of   the  site.  A  corresponding
    potentiometric-surface map foreach selected time was
    consulted to verify that changes in hydraulic gradient
    were due to the influence of the trees rather than to
    changes in the direction of groundwater flow.

    The thickness of the saturated zone at the selected
    times was calculated from the average thickness of the
    aquifer  in  the   monitoring  wells  immediately
    downgradient of the tree  plots (wells  526, 527, and
    528) (Figures 4-1 and 4-2). The saturated thickness in
    each of these three wells was first  normalized to wells
    in the  surrounding  area  to account for temporal
    changes  in the  saturated thickness  of the  aquifer
    unrelated  to  the   planted trees.  Specifically,  the
    water-level  data for these wells were adjusted by an
    amount equal to the difference between the water level
    at the selected time and the  water level at baseline
    (November 1996) in wells outside the influence of the
    planted trees. (November 1996 was used to represent
    baseline conditions in the aquifer because the most
    comprehensive set of water-level  and ground-water
    chemistry data for the period before  the tree roots
    reached the water table were  collected at this time.)
                                                    36

-------
       Whip Plantation

       JK
,      ,/  ,/  XV— Caliper-Tree
   x/ S|/  x/ \ Plantation
 \//^x
     \ X
        .-Mature    o    25   so
         Cottonwood
         Tree
              529
                 •  Monitoring well and number
                                                                                           DMngradient
                                                                                          endoftreatn-ert
                                                                                               system
                                                            Wrter table
                                                            during peak
                                                            transpiration
                                                         NC5TTOSGALE
                                                                             522
                                                                                   (523, 528, 527)  523
                                                                                      Wflls
 Figure 4-2.  Wells used to monitor for changes in the volumetric flux of groundwater across the downgradient end of the Short
            Rotation Woody Crop Groundwater Treatment system.
    The aquifer width that was used in the volumetric-flux
    calculations is 70 m, which is the approximate length
    of the tree plantations.
The mass flux across the downgradient end of the planted
area was subsequently calculated for the various events
(baseline, peak growing  season, late growing season)
according to the following formula:
         Mf=Q(C)
                    (Eqn. 4.3-1)
where Q is the volumetric flux of groundwater and C is the
average TCE concentration  in wells 526, 527, and 528
(immediately downgradient of the planted area) for each
event.

The  following formula  was  then used to calculate  the
percent change in the mass flux of TCE at selected times
that can be attributed to the planted trees:

                                     entOcty    (100)
 (Eqn. 4.3-2)
    Mf {ba
                               seline D
Where:
Event x is  peak (late June or beginning  of July) of the
growing season 1997, 1998,  or 1999, or late (end of
September or beginning of October) in the growing season
1997 or 1998.
4.3.2    Secondary Project Objectives
Secondary  objectives were  included in the  study to
elucidate the  biological,  hydrological, and  biochemical
processes  that  contribute to  the  effectiveness of  a
SRWCGT   system   on  shallow  TCE-contaminated
groundwater. Since a SRWCGT system can take several
years to  become  fully  effective,  much  of the  data
associated with the secondary objectives were collected to
build predictive models to determine future performance.
Measurements were primarily related to tree physiology
(tree growth, tree transpiration, contaminant translocation)
and  aquifer  characteristics  (hydraulic,  geochemical,
microbiological).   Scientists   at   Science   Applications
International Corporation (SAIC),  University of Georgia,
U.S. Forest  Service, USEPA, and USGS conducted the
work  related  to  the secondary  project  objectives in
cooperation with ASC/ENV and the USEPA SITE program.
Secondary objectives and the scope of the associated data
collection are described below:
Determine tree growth rates and root biomass: Above-
ground biomass growth was measured over the course of
the study to assess the rate-of-growth  of  the whip  and
caliper-tree  plantations.   Fifty-two  whips  and  fifty-one
caliper-trees were evaluated for the following parameters:
(1) trunk diameter, (2) tree height, and (3) canopy diameter.
The  measurements were  taken during  the  following
sampling events:  (1) December 1996, (2) May 1997, (3)
July  1997,  (4) October 1997, (5) June 1998,  and (6)
October 1998.  An additional investigation was undertaken
to quantify below ground biomass and the extent of the root
system in September of 1997. This information was used
to understand the establishment of the root system, which
is the primary means for targeting the contaminants in the
aquifer.  Differences in root characteristics between the
whip  plantings and  the  more  expensive  caliper-tree
plantings were also investigated.  Eight trees (four from
each plantation) were examined.

Analyze tree transpiration rates to determine current and
                                                     37

-------
future water usage: An important remediation mechanism
of the planted system is the interception and removal of
water  from  the  contaminated   aquifer.    Measured
transpiration rates can provide information that is critical for
evaluating  current removal of water  from the aquifer
(saturated zone) and for predicting future water usage.
Transpiration rates were quantified for the whips and the
caliper-tree plantings, as well as for several mature trees
proximal to the study area. Sapflow, leaf conductance, and
pre-dawn and mid-day leaf water potential were measured
on 14 to 16 trees from May through October in 1997 and
1998.  Climate data  were also collected at the site and
used in conjunction with the transpiration data to model
future tree transpiration.

Analyze the hydrologic effects of tree transpiration on the
contaminated aquifer: The removal of contaminated water
from the aquifer at the Carswell Golf Course site has the
potential to alter the local  groundwater flow system,
resulting in some  hydraulic control of the contaminant
plume.  Hydraulic  control may be one of the  principal
mechanisms related to reduction in TCE mass flux across
the downgradient end of the planted system.  Groundwater
level data were collected and used to assess the hydrologic
effects  of the cottonwood trees  on  the contaminated
aquifer. Specifically, data were collected in up to 62 wells
during  November and December  1996;  May, July, and
October 1997; February, June, and September 1998; and
June 1999. In addition, groundwater levels were measured
every  15  minutes in three  wells to  record seasonal
fluctuations in groundwater levels over the course of the
study.  Beginning in summer 1998, the stage in Farmers
Branch Creek was also recorded every 15 minutes so that
the  hydrologic effects of the trees could be  isolated from
other temporal changes  in the system.  Slug tests were
conducted in eleven  wells to determine the site-specific
hydraulic conductivity of the aquifer. Eleven core samples
were collected and analyzed in the laboratory to determine
site-specific aquifer porosity.  These data, along with the
transpiration data, were  used  to model future hydrologic
effects of the planted  trees on the contaminated aquifer.

Analyze contaminant uptake into plant organ systems: A
potential removal mechanism  for TCE and  other volatile
contaminants in  the aquifer is  translocation  of  the
contaminants into the plant organs. Chlorinated ethenes
may be transpired  through the stomata of the leaves or
metabolized within the plant organs to other compounds
such as simple haloacetic acids (N. Lee Wolf, U.S.EPA,
written communication 1999). To assess the presence and
magnitude of contaminant uptake and translocation at the
study area,  plant organ  samples of  roots, stems, and
leaves were acquired and analyzed for volatile organic
compounds. Samples were taken from five whip plantings,
five caliper-tree  plantings, a  mature  naturally-occurring
cottonwood, and  a naturally-occurring mesquite tree. The
trees  were  sampled during  the  following  events:  (1)
October 1996 - end of the first growing season, (2) July
1997 - peak of the second growing season, (3) October
1997- end of the second growing season, (4) June 1998 -
peak of the third growing season, and (5) October 1998 -
end of the third growing season. Tree cores were collected
from 11 species of trees surrounding the planted area and
analyzed for the presence  of TCE and  cis-1,2-DCE  in
September 1998.  In addition, leaves from seven trees
(cottonwood  whip,  cottonwood  caliper  tree,   cedar,
hackberry,  oak,  willow,  mesquite) were collected and
analyzed for dehalogenase activity to  determine whether
the leaves had the capability to break down  TCE.

Evaluate  geochemical  indices   of   subsurface
oxidation-reduction processes: Many TCE contaminated
aquifers could benefit from microbially-mediated reductive
dechlorination. Reductivedechlorination, however, cannot
take place under the aerobic conditions that are present at
many  such shallow  sites,  where  TCE  is the sole
contaminant.  Processes that promote the consumption of
oxygen  in the subsurface  can accelerate  the microbial
reductive dechlorination process.  Trees  can  promote
subsurface oxygen utilization by providing the subsurface
environment with organic matter that stimulates  aerobic
microbial activity that can result in depleted oxygen levels
and   resulting  anaerobic  conditions.     Groundwater
geochemical samples were collected at the  study area to
assess  the  development  of  an anaerobic  subsurface
environment overtime, along with any associated reductive
dechlorination  of the chlorinated ethenes. Samples were
collected from  both the groundwater and the unsaturated
soil throughout the study area.  Groundwater analyses
included  chlorinated volatile organic compounds  (VOCs,
including TCE and cis-1,2-DCE), dissolved organic carbon,
methane, sulfide, ferrous and total iron, dissolved oxygen,
and dissolved hydrogen.  Soil measurements (unsaturated
zone) included total organic carbon and pH.

Evaluate microbial contributions to reductive dechlorination:
A  microbial survey was  performed at the study  area  to
determine if the  planted trees  have driven the local
microbial  community structure  to  support  reductive
dechlorination  of TCE. Samples of soil and groundwater
were collected from thirteen locations in February and June
of 1998.  Microbial concentrations were determined using
a  five-tube  Most  Probable  Number (MPN) analysis.
Enumerations were performed to determine the populations
of the  following  types of  microorganisms: aerobes,
denitrifiers, fermenters,  iron-reducers, sulfate reducers,
total  methanogens,  acetate-utilizing  methanogens,
formate-utilizing  methanogens,  and   hydrogen-utilizing
methanogens.     Laboratory   microcosms  were  also
established to  estimate biodegradation-rate  constants for
the demonstration site.
                                                    38

-------
4.4 Performance Data
The  following sections present a  discussion  of the
technology's performance with respect to the primary and
secondary project objectives.  The purpose of the following
sections is to present and discuss the results specific to
each objective, provide an interpretive analysis from which
the conclusions are drawn, and, if relevant, offer alternative
explanations and viewpoints.

4.4.7 Summary of Results -  Primary Objective
The primary objective of the  study was to  determine the
Phytoremediation system's ability to reduce the mass flux
of TCE across the downgradient end of the  site during the
second (1997) and  third (1998) growing  season.  The
objective  called  for  a 30 percent reduction during the
second growing season and a 50 percent reduction during
the third growing season. The objective could be achieved
from a combination of the two mechanisms hypothesized
to be capable of contaminant  reduction -  hydraulic control
and in-situ reductive dechlorination.
Table 4-1 presents the results of the calculations used to
validate the primary claims described in equations 4.2-1,
4.3-1, and 4.3-2.  The SRWCGT system did not achieve
the mass flux reductions of 30 and 50 percent for the
second and third growing seasons, respectively. The TCE
mass flux was actually up 8 percent during the peak of the
second  growing  season,  as compared  to  baseline
conditions.  The planted trees reduced the outward flux of
groundwater by 5 percent during the peak of the second
season  but TCE concentrations  in the  row  of wells
immediately downgradient  of the  trees  were  higher,
resulting in the  increase in TCE mass  flux.  These data
suggest that the mass flux of TCE out of the planted area
during the peak of the second season  would have been
even greater in  the absence of the hydraulic influence of
the trees.  The TCE  mass flux during  the third growing
season was down 11 percent at the  peak of the season
and  down 8  percent  near the end of the season, as
compared to baseline conditions.  Concentrations of TCE
during the third  season in the row of downgradient wells
were  similar to  concentrations  at baseline  and  the
reduction in TCE mass flux is primarily attributed to  a
reduction in the volumetric flux of groundwater out of the
site. The flux of groundwater out of the site during the
peak of the fourth growing season was 8 percent less than
at baseline.  Groundwater was  not sampled  for TCE
concentrations  at this  time.   Variations  in  climatic
conditions are the likely explanation for the differences in
the outward flux of groundwater  between the third  and
fourth seasons.  In general, these data reveal  that the
system had begun to influence the mass of contaminants
moving   through  the  site   during  the   three-year
demonstration.

The  contributions of hydraulic  control  and  reductive
dechlorination   as   attenuation  mechanisms  can  be
evaluated from the study results. The principle mechanism
for the reductions in mass flux observed during the early
stage of the system's development was hydraulic control.
TCE concentrations from the downgradient row of wells did
not decrease during the first three growing seasons, which
indicates that reductive dechlorination processes had not
yet significantly  occurred (Table  4-1).   Although TCE
concentrations had not decreased, there was a reduction
in the mass of TCE in the plume just downgradient of the
study area because  tree transpiration had affected the
volumetric flux of contaminated water out of the site. This
is evidenced by the  decrease  in the hydraulic gradient
across the downgradient end of the planted area, as well
as the decrease in saturated thickness of the aquifer at the
downgradient end  of the site.   The  largest observed
reduction in hydraulic gradient was 10 percent (0.0159 to
0.0143) and occurred during  June 1998.  The maximum
drawdown that could be attributed to the trees during June
1998 is 10 cm and was  observed between the two tree
plots. Although a drawdown cone could be mapped at the
watertable at this stage of the system's development, there
remained a regional hydraulic gradient across the site that
resulted in most of the contaminated groundwater flowing
outward across the downgradient end of the planted area
(Figure 4-3).

A ground-water flow model of the demonstration site was
constructed using MODFLOW (McDonald and Harbaugh,
1988) to help in understanding the observed effects of tree
transpiration on  the aquifer (Eberts, et. al. In Press). The
model  illustrates that the volume of water that was
transpired from the aquifer during 1998 was greater than
the reduced outflow of groundwater that can be attributed
to the trees. This is because of an increased amount of
groundwater inflow to the demonstration site due to an
increase in hydraulic gradient  on the upgradient side of the
drawdown cone created by  the trees.  The amount of
contaminated water that was transpired from the aquifer
during the peak of the 1998 growing season (third season)
was  equal to an amount that is closer to 20 percent of the
initial volumetric flux of water through the site rather than
the observed decrease in outflow of 12 percent.

Greater hydraulic control  is anticipated  in  the  future
because the trees  did not reach  their full transpiration
potential during the time period of the demonstration study.
Predictions for out-year hydraulic control will be discussed
in greater detail  in section 4.4-2.

4.4.2 Summary of Results -  Secondary Objectives

In addition to providing the data necessary to evaluate the
primary claim, the demonstration project included several
studies designed to address secondary project objectives.
Results of these studies provide insight into the SRWCGT
System's contaminant-reduction mechanisms.  Since a
Tree  system  may  take  several  years  to  become
established, special attention was given to the derivation of
                                                    39

-------
 Table 4-1. Summary of Primary Objective Results [m, meter; d, day;ug, microgram; L, liter; g, gram] (See Appendix C)






Event
Baseline
(1996)
Peak6 2nd
Season (1 997)
Late6 2nd
Season (1 997)
Peak 3rd
Season (1 998)
Late 3rd
Season (1 998)
Peak 4th
Season (1 999)


Hydraulic
Gradient Across
Downgradient
End of Planted
Area"
0.0159

0.0154

0.0157

0.0143

0.0150

0.0153


Cross
Sectional Area
Along
Downgradient
End of Planted
Area" (m2)
84

82

83

82

83

81


Volumetric Flux
of Groundater
Across
Downgradient
End of Planted
Area0 (m3/d)
8.0

7.6

7.8

7.0

7.5

7.4

Change in
Voumetric Flux
Across
Downgradient
End of Planted
Area Attibuted
to Planted
Trees (%)


-5%

-2%

-12%

-6%

-8%

Average TCE
Concentration
in Wells Along
Downgradient
End of Planted
Area"
(ug/L)
469

535

_

483

473

_



Mass Flux of
TCE Across
Downgradient
End of Planted
Area (g/d)
3.8

4.1

_

3.4

3.5

_

Change in Mass
Flux of TCE
Across
Downgradient
End of Planted
Area Attributed
to Planted Trees
(%)


8%

_

-11%

-8%

_

    Gradient calculated between monitoring wells 522 and 529.
    An aquifer width of 70m was used for the aquifer cross-sectional area calculations; aquifer thickness was the average of the saturated thickness
    in wells 526, 527, and 528 normalized to wells from the surrounding area to account for seasonal water table fluctuations unrelated to the
    planted trees.
    A horizontal hydraulic conductivity of 6 m/day was used for the volumetric flux calculations. This is the geometric mean of the hydraulic
    conductivity values determined for the study area.
    TCE concentration is the average in wells 526, 527 and 528.
    Peak growing season is end of June or beginning of July. Late growing season is end of September or beginning of October.
°A
  Q\-
  \*'s>
  -$&/
  180 m''
                               Whrp Plantatiofi
          179.5m"
                                              - Cal>per-Tree
                                             \   DJ-a'ntatiop-
                                            *£.---
                                            .2 cm
                                                          25
                  479 mjrMature-'"'  =
                        W  GGttafiWOOd      meters

        Farmers Branch Creek *m,:
                             Tree
            2 cm       Line of equal drawdown. Interval 1.5 cm

    .	  179 m      Line of equal water table elevation. Interval 0.25 m
Figure 4-3. Drawdown at the water table that can be attributed to the trees, June 1998.
                                                       40

-------
parameters  that  could   be  used  to  model  future
performance.   In  addition,  a  mature  cottonwood  tree
located  proximal to the planted trees provided valuable
information related to the  upper bounds of contaminant
reduction.

Determine tree growth rates and root biomass

The rate of tree growth (above- and below- ground) was
important for determining the progression of the SRWCGT
system overtime. Above-ground biomass, especially leaf
area, controls transpiration rates and the ability of such a
system to influence groundwater hydrology.  The growth of
the below-ground  organs (roots)  controls a  system's
efficiency for extracting water from the aquifer (saturated
zone).

Fifty-two whips and fifty-one caliper trees were measured
for trunk diameter, tree height, and canopy diameter in
December 1996, May 1997, July 1997, October 1997, June
1998, and October 1998 by employees of SAIC. Figures 4-
4 through 4-6 graphically depict the physical changes in the
whip and caliper-tree plantations overtime. Figure 4-7 is a
photograph of the  caliper-tree plantation  at the time of
planting (April  1996).  Figure 4-8 is a photograph of the
caliper-tree plantation at the end  of the  third growing
season (October 1998).

Overall,  both  plantations grew well  and significantly
increased in all physical parameters measured over the
course of the study.  Only two of the fifty-two whips and
three of the fifty-one caliper trees did not survive to the end
of the study. (Some of the other trees in the plantations,
however, were temporarily stunted by beaver activity during
the study.) In terms of trunk diameter,  both plantations
increased overtime; 1.41 cm to 5.13 cm for the whips, and
3.83 to  8.12 cm for the  caliper trees.  Tree height also
significantly increased for both plantations. In December of
1996, tree height for the whips averaged 2.27 m and 3.77
m  for the caliper trees.  In September  of  1998, average
tree height for the whips was 5.52 m and  6.64 m for the
caliper trees. Although the caliper trees  were taller during
the first growing season, the whips were able to approach
the height of the caliper trees by  the  end of the  third
growing season. For the canopy diameter, both the whips
and calipertrees increased overtime, however, there were
minor differences between the plantations overtime.

Canopy diameter is an important parameter that controls
leaf  area  and  transpiration.   In  an   open  growth
environment, canopy diameter is dependent on the overall
growth  and maturation  of  the tree.  In  a designed
plantation,  individual  trees  are  planted  in rows at  a
specified spacing.   As the trees grow, the canopies of
individual trees can touch, which slows down furthergrowth
due to competition for light.   This limits  the  maximum
stand-level transpiration attainable for  individual trees,
however, it does not affect the maximum amount of water
that can  be transpired by the whole plantation if the tree
spacing is such that a closed canopy eventually will  be
achieved.  Trees in the whip  plantation were planted
approximately 1.25 m apart. The average canopy diameter
for the whips at the end of September 1998 (end of the
third growing season) was 2.32  m.  The whip plantation
was approaching canopy closure at this time. Trees in the
caliper-tree plantation were planted approximately 2.50 m
apart.  The average canopy diameter for the caliper trees
in  September of 1998  was 2.52 m.   The  caliper-tree
plantation was not approaching canopy closure at this time.

Root biomass and extent were examined in September of
1997 in the whip and caliper-tree plantations. Four trees
from each plantation were evaluated  for fine root biomass
and length,  coarse  root biomass, and root distribution.
Differences   in  the  fine  root  biomass  between  the
plantations were not statistically  significant: 288 g rrr2  for
whips vs. 273 g m"2 for caliper trees in the <0.5 mm range;
30 g m"2 for whips vs. 36 g m"2 for the caliper trees in the
0.5 to 1.0 mm range; and 60 g rrr2 for the whips vs.  91 g
m"2 for the caliper trees in the 1.0 to  3.0 mm range.  Fine
root length  density  in  the upper  30  cm  of  soil  was
statistically greater in the calipertrees as compared to the
whips (8942 m m"2 vs. 7109 m m"2). Coarse root mass was
significantly greater in the caliper trees in the 3.0 to 10 mm
range; 458 g tree"1 vs. 240 g tree"1.  Although the coarse
root mass in the > 10mm range was also greater in the
calipertrees than in the whips; the difference in this range
was not statistically significant.  Details of this root study
can be found in a report entitled, "Root Biomass and Extent
in Populus Plantations" (Hendrick, 1998).

At this point  in the second growing  season (September
1997), the roots of both the whips and caliper trees had
reached the water table (275 cm for the whips and 225 cm
forthe calipertrees), and the depth distribution of the roots
was quite similar (Figure 4-9).  In other words,  the more
expensive planting costs of the calipertrees did not appear
to impart any substantial benefit with  regards to root depth
and biomass. Observed differences between the whips
and the calipertrees were reported to be due as much to
inherent genotypic differences as to the different modes of
establishment.

Analyze tree transpiration rates to determine current and
future water usage

Transpiration is the evaporative loss of water from a plant.
Water transport mechanisms move  water from the soil
zone to the stomata  of the  leaf where it is lost to the
atmosphere.  Transpired water  can  be derived from the
near surface soils,  and in  the case of phreatophytic
species, from the saturated zone (aquifer). The ability of
phreatophytic species to seek  and use contaminated
groundwater is the basis of this  system technology.  The
amount of water transpired by trees throughout their  life
cycle is an important factor in
                                                    41

-------
                              May-97
                                           July-97
                                                                                    Sep-98
Figure 4-4. Trunk diameter over time.
                            May-97
                                          July-97
                                                                                    Sep-98
Figure 4-5. Tree height over time.
                                                      42

-------
                            DWhip     BCaliperTree
                                   May-97         July-97
                                                                                     Sep-98
           Figure 4-6. Canopy diameter overtime.
Figure 4-7.   Caliper-tree plantation at the time of planting, April
             1996.
Figure 4-8.   Caliper-tree plantation at the end of the
            third growing season, October 1998.
                                                            43

-------
              5O
            1OO
   Depth
    (cm)
             15O
             200
             250
                                               20
       30
Root Counts
40
5O
60
 Figure 4-9. Root counts by depth.

determining the  effectiveness  of the  technology for
containment and  remediation of a contaminant  plume.
Transpiration rates can be used in conjunction with other
site-specific characteristics (climate, soil type, hydrology)
to determine water use patterns and to help determine
process effectiveness, including future performance.

Scientists  from  the  USDA  Forest  Service,  Cowetta
Hydrologic Laboratory, conducted a transpiration study at
the  demonstration  site.     Specifically,  transpiration
measurements were taken on a statistical sampling of
whips and  caliper trees  in May, June, July, August, and
October of 1997. In addition, transpiration was measured
on six mature trees in the vicinity of the study area in May,
July, and Septemberof 1998. Transpiration measurements
on  individual  trees  were  extrapolated  to  estimate
stand-level transpiration rates.  The sapflow data were
used to (1) compare transpiration rates for the two planting
strategies  (whips  vs.  caliper  trees),   (2)  investigate
variability  over the growing  season, and (3)  determine
stand-level water usage over the entire  growing season.
Data  from the  mature trees  was used to  estimate
upper-bound levels of transpiration that may be attainable
  by the  Phytoremediation system in the  future.  The
  transpiration measurements are summarized in a report
  entitled  "Leaf Water Relations and Sapflow in Eastern
  Cottonwoods (Vose et al., 2000).

  The greatest sapflow in the planted trees occurred in June,
  while  the lowest occurred in  the month of October. In
  general, sapflow was significantly greater in individual
  caliper trees than in individual whips for all months except
  October (Figure 4-1 Oa).

  The average seasonal sapflow for the caliper trees was
  almost two times greaterthan that of the whips (0.61 kg hr1
  tree"1  vs. 0.34  kg hr1 tree"1).  Because the whips were
  considerably  smaller  than  the  caliper  trees,   the
  investigators also expressed sapflow on a per unit basal
  area basis (kg cm"2 hr1).  When expressed this way, rates
  were generally greater in the whips than in the calipertrees
  (0.033 kg cm"2 hr1 vs. 0.027 kg cm"2 hr1) (Figure 4-1 Ob).

  Mean  total daily transpiration rates were also determined.
  Mean  total daily transpiration forthe whips ranged from 9.2
  kg tree"1 day1 (2.4 gallons tree"1 day1) in June to 1.6 kg
  tree"1 day1 (0.42 gallons tree"1  day1) in October.  Mean
                                                    44

-------
 IB
 a:
 C/> 4=
 £•1.
 e -c
           1.0
           0.8
           0.6
Caliper Trees
Whips
                    May   June  July   Aug   Oct
  5
  Q- 'i-
  CD
          0.06 J
          0.04
          0.02
          0.00
                                  Cahper Trees
                                  Whips
                    May  June   July   Aug   Oct
Figure 4-10. Variation in mean hourly sapflow rate (a) expressed
           on a per tree basis and (b) expressed on a per unit
           basal area basis.  Data are sample period means
           for all months (p<0.05) differences between whips
           and calipertrees are denoted by*. Vertical lines on
           all bars represent standard errors.
total daily transpiration for the caliper trees ranged from
14.7 kg tree'1 day1 (3.89 gallons tree'1 day1) in July to 0.92
kg tree"1 day1 (0.24 gallons tree"1 day1) in October.
Preliminary estimates  of stand-level transpiration were
extrapolated from these total daily  mean  transpiration
values by assuming that the amount of sapflow measured
in  the sample  trees  represents  the population.  The
stand-level estimates  indicate that there  was very little
difference in the amount of water transpired from the whip
plantation and the caliper-tree plantation during the second
growing season. This is because the planting density of
the whips is nearly twice that  of the caliper trees.  When
sapflow values were averaged across the second growing
season, sapflow was 16,637 kg ha"1 day1 for the caliper
trees, and 15,560 kg ha"1  day1 for the whips.  Because
each plantation measures approximately 75 by 15 meters
(0.1125 hectares), the total average daily transpiration was
estimated at 1,872 liters day1 (494 gallons day1) for the
caliper-tree plantation and 1,750 liters day1 (462 gallons
day1) for the whip plantation.  These amounts correlate
with an estimated loss of water through transpiration from
the study area  of approximately 3,600 liters  day1 (950
gallons day1) during the second growing  season.  Total
estimated growing season transpiration for the second
season was estimated to be approximately 25  cm.  It was
noted that this amount of transpiration is about one-third to
one-half  of the  amount of  transpiration for mature
hardwood forests in other regions of the U.S.  (Vose and
Swank,  1992), which  indicates that substantially greater
transpiration will occur as the planted trees mature.
The  sapflow rate that was  measured  for the mature
cottonwood tree adjacent to the planted site was as high as
230 kg day1 (~60 gallons day1).  This value represents an
upper limit of potential transpiration by a single tree at the
demonstration site.  This rate, however, is non-attainable
in a  plantation  configuration.   As previously  discussed,
canopy closure in the whip and caliper-tree plantations will
eventually  limit  leaf  area and  thereby  the  maximum
potential transpiration of individual trees.  As a result, the
spacing of the trees in  the  SRWCGT  system at the
demonstration site will affect the amount of  water that
individual  trees will transpire, but should not affect the
amount of water that will be transpired  by the overall
plantations as  long  as  canopy closure is  eventually
achieved. Tree spacing will, however, affect the timing of
canopy closure.  The full  report on "Sap Flow Rates in
Large Trees  at the Carswell  Naval Air Station" can  be
found in the report entitled the same (Vose and Swank,
1998).

Because the planted trees were not expected to reach their
transpiration potential during the period of demonstration,
a  modeling  approach was necessary to predict  future
system   performance  at  the   demonstration   site.
Site-specific climate, sapflow,  soil-moisture, and tree-root
data  were  used  to parameterize  and  validate  the
physiologically-based model PROSPER  (Goldstein and
others, 1974), which was then used to predict the amount
of evapotranspiration at the site that will likely occur once
the plantations have achieved  a closed canopy (maximum
transpiration).  Predictions vary according  to assumptions
made regarding future climatic conditions, as well as soil
moisture  and  root  growth.     Predicted   stand-level
evapotranspiration forthe period when the  tree  plantations
have achieved a closed canopy (year 12 and beyond) is the
same for whips and calipertrees and ranges from 25 to 48
cm pergrowing season, depending on model assumptions.
The root biomass study (Hendrick, 1998) was conducted to
help determine the  percent of this transpired  water that
may be derived from the contaminated aquifer (saturated
zone). Predicted transpiration from the aquifer ranges from
12 to 28 cm per growing season for year 12 and beyond,
depending on model assumptions; this is 48 to 58 percent
of predicted total evapotranspiration.  The  effects of this
                                                     45

-------
amount of transpiration on the groundwaterflow system in
the study area are discussed in the next section.

Analyze the hydrologic effects of tree transpiration on the
contaminated aquifer

The ground-water flow model that was constructed to help
in understanding the observed effects of tree transpiration
on the aquifer was also used to predict the effects of future
increases in transpiration rates on the  volumetric flux of
groundwater across the downgradient end of the planted
area by incorporating the predictions of future transpiration
from the  saturated zone made by use  of the hydrologic
model PROSPER.  Hydrologists with the USGS used the
groundwater  flow  code  MODFLOW  to  construct the
groundwater flow model and to make the volumetric flux
predictions.  Site-specific data on aquifer characteristics,
groundwater levels, and stream stage,  as well as stream
discharge measurements reported in Rivers and others
(1996) were used to calibrate the groundwaterflow model
to both steady state and transient state  conditions before
the model was  used to make predictions.  (One lesson
learned during collection of continuous water-level data for
construction  of this model is that tree roots grow through
well screens and entangle downhole instrumentation, which
can  lead to loss  of data.  Sites need to be  checked
frequently and wells need to  be reamed periodically  to
remove roots.)

The  groundwater flow  model was used to predict the
magnitude and extent of the drawdown  cone that may be
expected as a result of future transpiration at the study
area. A volumetric groundwater budget  was computed for
each predictive simulation. Because the  PROSPER model
predictions simulate a range of possible climatic conditions,
as well as soil-water availability and root  growth scenarios,
there is a range of predicted drawdown and predicted
reductions in the outflow of groundwater from the planted
area. Predicted drawdown during peak growing season
afterthe trees have achieved a closed canopy (year 12 and
beyond)  ranges from 12 to 25 cm at  the center of the
drawdown cone. The diameter of the predicted drawdown
cone ranges  from  approximately 140 m to over 210  m
(Figures 4-11 and 4-12).

These  drawdown  predictions are  associated  with  a
predicted decrease in the volumetric flux of groundwater
across the  downgradient end of the planted  area that
ranges from 20 to 30 percent of the volumetric flux of water
through the  site before the trees were planted.   The
predicted volume of water transpired from the aquifer in
future  years  when  maximum transpiration  has  been
reached  ranges from  50 to  90  percent of the  initial
volumetric flux of groundwater at the site. The discrepancy
between  the  reduction in the  volumetric outflow  of
groundwater and the volume of water transpired from the
aquifer can  be attributed to the combined  increase  in
hydraulic gradient on the upgradient side of the drawdown
cone, which leads to an increase in groundwater inflow to
the site,  and the release of water from storage in the
aquifer (Figure 4-13).

These model results indicate that a regional hydraulic
gradient will remain across the planted area during future
growing seasons.  The volumetric flux of groundwater
across the downgradient end of the planted area, however,
will be notably reduced. Percent reductions in the TCE
mass flux due to tree transpiration  will be somewhat less
than reductions  in the volumetric flux of groundwater
because  membrane barriers at the root surface  prevent
TCE from being taken up at the same concentration as it
occurs in the groundwater.   The transpiration  stream
concentration factor or  fractional efficiency of uptake for
TCE has been reported to be 0.74 (Schnoor,  1997).  No
hydraulic control of the plume is predicted for the dormant
season (Novemberthrough March). Additional information
on the hydrologic effects of cottonwood trees can be found
in  the report entitled "Hydrologic effects of cottonwood
trees on a  shallow  aquifer  containing trichloroethene"
(Eberts et al., 1999).

It may be possible to achieve a greater amount of hydraulic
control if more trees are planted but increased groundwater
inflow and release of water from storage in the aquifer will
continue  to be factors that affect hydraulic control of the
contaminant plume.  It is also possible that full hydraulic
control of  the  plume  would not  be desirable  if the
demonstration project were scaled up because full control
may result in an unacceptable decrease in flow in Farmers
Branch Creek, particularly since hydraulic control is only
one  mechanism that contributes  to the  cleanup of a
groundwater plume at a phytoremediation  site. A solute
transport model of the groundwater system at the  study
area is being constructed to gain insight into the  relative
importance of various attenuation mechanisms associated
with   Tree  systems   -  hydraulic  control,   reductive
dechlorination, and sorption.

Analyze contaminant uptake into plant organ systems

During the  period of  demonstration,  employees of SAIC
collected plant tissue samples from the whips, calipertrees,
and the mature cottonwood tree five times (October  1996,
July  1997, October 1997, June 1998, and October 1998).
Specifically, leaf and stem (new growth) samples were
taken from five whips, five caliper  trees, and the mature
cottonwood  tree  during each  sampling  event.   Root
samples  were collected from the  whip and caliper-tree
plantations during the  October 1996 and June  1998
sampling events. The samples were analyzed for volatile
organic  compounds  (VOCs).   The  purpose of these
analyses was to determine (1)  if volatile compounds
(especially chlorinated VOCs) were present in the plant
                                                    46

-------
                                                                                                    Explanation
                                                                                           3  Line of equal simulated drawdown at the
                                                                                              water table. Interval 3 centimeters
Figure 4-11. Minimum predicted drawdown at the water table for closed-canopy conditions (year 12 and beyond).
                                                                                                    Explanation
                                                                                           3 Lines of equal simulated drawdown at
                                                                                             the water table. Interval 3 centimeters
Figure 4-12. Maximum predicted drawdown at the water table for closed-canopy conditions (year 12 and beyond).
                                                             47

-------
                   100%
                                              100%
                                                        B
                                                                                              20%
                                                                       110%
                                                               (10% Increase in
                                                                Inflow Due to
                                                              Increased Gradient)
                                                                                                     90%
            (15 % Increase in
             Inflow Due to
           Increased Gradient)
 (Additional Water Due
to Release from Storage)
                                                80%
                                                        D
                                                                       125%
 (25 % Increase in
  Inflow Due to
Increased Gradient)
      35%
 (Additional Water Due  -
to Release from Storage)
                                                                                                     70%
 Figure 4-13. Simulated groundwater budget (A) prior to treatment, (B) peak of the third growing season (1998), (C) peak of the
            growing season once closed canopy has been achieved (year 12 and beyond)-minimum predicted transpiration, and
            (D) peak of the growing season once closed canopy has been achieved (year 12 and beyond)-maximum predicted
            transpiration.
tissues,  (2)   whether  there  were  changes  in  the
concentration of such compounds in the plant tissues over
time, and (3) whether there were differences between the
samples collected from the plantations and those collected
from the mature tree.  The results of these analyses were
used  to determine  whether chlorinated  ethenes  are
translocated from  the subsurface  into the trees at the
demonstration site.

Table 4-2 is a summary of the plant tissue data. The table
depicts (for each sampling event) plant tissue, tree type, the
average concentration of detected volatile compounds, and
the number of tissue samples exhibiting detectable levels of
that compound. Thirty volatile compounds were scanned
as part  of the method.  However, only seven compounds
were  detected in  the tissue samples.   The  detected
compounds include trichloroethene, cis-1,2 dichloroethene,
methylene chloride, tetrachloroethene, chloroform, toluene,
and  acrolein.   Five of the  seven volatile compounds
detected  are   chlorinated.    Toluene  is  an  aromatic
compound and acrolein is an aldehyde.

The following conclusions can be drawn from this data:

1.   Chlorinated compounds were commonly encountered
    in tissue samples during all sampling events. The stem
    samples generally exhibited the greatest diversity and
    concentration of chlorinated compounds.
2.   With regards  to the  chlorinated  ethenes  in  the
                                   plantations, there was a general increase overtime in
                                   the percentage of trees that contained the compounds,
                                   as well as an increase in the average concentration.
                                   The highest concentrations of chlorinated ethenes were
                                   encountered  during the October 1998 sampling event.
                                   All five whip  and five caliper-tree samples contained
                                   detectable levels  of trichhloroethene  in  the stems.
                                   Average stem concentrations were 32.8 ug/kg for the
                                   whips and 24.6 ug/kg for the caliper trees.
                                3.  There  were no major differences between the whips
                                   and  caliper-tree  plantations  with  respect  to  the
                                   presence and concentration of VOCs.
                                4.  The concentrations of  chlorinated  ethenes in  the
                                   plantations was higherthan detected in the mature tree.
                                The presence and  increasing abundance  over  time of
                                chlorinated ethenes in the plant tissues are an indication
                                that the   plantations  progressively  translocated  more
                                contaminants from the subsurface over time.   This data
                                cannot be  used to assess the fate of these  contaminants
                                within the plant tissues orto determine if they are volatilized
                                into the atmosphere.

                                Tree  cores were collected by  USGS with an increment
                                borer from 23 mature trees surrounding the demonstration
                                site and  analyzed  for the  presence  of   TCE   and
                                cis-1,2-DCE.   Eleven  species  of trees were sampled,
                                                      48

-------
 Table 4-2.  Average concentration of detectable volatile compounds in plant tissue [concentrations are in units of ug/kg; ND, non
            detected; NS, not sampled].
Event

  i_

 5  Is-
     *3  O>
     a>
     £2
     O  1^.
     H  °>
     £  «
     a>  oo
     c  o>
     3  O>
        00
           Analyte

     Trichloroethene
           Acrolein
         Chloroform
  Methylene Chloride
cis-1,2 Dichloroethene

     Trichloroethene
           Acrolein
         Chloroform
  Methylene Chloride
           Toluene
   Tetrachloroethene

     Trichloroethene
           Acrolein
  Methylene Chloride
cis-1,2 Dichloroethene
           Toluene
   Tetrachloroethene

     Trichloroethene
           Acrolein
cis-1,2 Dichloroethene
           Toluene

     Trichloroethene
           Acrolein
cis-1,2 Dichloroethene
                                 Leaf
Whips
 Stem
Roof
Leaf
Caliper Trees
   Stem
                                                                                        Mature Cottonwood
Roof
Leaf
   Number in parentheses represents the number of trees for which analyte was detected.
   Five whips and five caliper trees were sampled (except roots).
Stem
Roof
ND
ND
ND
ND
ND
26(1)
15.2(3)
3.9(1)
15(2)
ND
ND
21 .7 (3)
ND
29(3)
ND
ND
ND
ND
ND
ND
ND
7.0 (2)
4.1 (1)
10(1)
ND
ND
9.1 (2)
ND
ND
ND
NS
NS
NS
NS
NS
ND
ND
ND
2.2
1.2
NS
NS
NS
NS
NS
ND
58.8 (5)
ND
151 (5)
0.73 (2)
ND
ND
136(3)
ND
153(3)
ND
ND
NS
NS
NS
NS
NS
NS
ND
19(1)
0.73(1)
168(5)
ND
ND
ND
46.2 (5)
ND
ND
ND
71(3)
NS
NS
NS
NS
NS
NS
ND
49
120
ND
0.7
ND
ND
35
ND
ND
ND
ND
NS
NS
NS
NS
NS
NS
1.6(2)
ND
8.3 (3)
ND
ND
ND
10.1 (3)
20(1)
6.6 (2)
1.9(3)
2.3 (3)
ND
NS
NS
NS
NS
NS
NS
10.4(3)
ND
ND
ND
4.3 (2)
ND
9.6 (3)
12.5(4)
3.6 (5)
1.6(3)
1.5(1)
5.1 (2)
NS
NS
NS
NS
NS
NS
ND
ND
6.3
ND
ND
ND
6.4
ND
2.8
10
ND
ND
NS
NS
NS
NS
NS
NS
ND
ND
ND
1.4(5)
44(1)
ND
14(1)
2.3 (2)
140
25
ND
1.1
4.5 (2)
ND
ND
1.1(2)
71 (1)
ND
15.7(3)
2.0(1)
13
ND

0.91
ND
ND
ND
ND
13
ND
ND
0.9
NS
NS
NS
NS
ND
ND
ND
32.8 (5)
14.4(3)
13.5(5)
NS
NS
NS
ND
ND
ND
24.6 (5)
ND
8.9 (4)
NS
NS
NS
ND
ND
ND
2.2
ND
2.8
NS
NS
NS
including five cottonwoods, six oaks, two live oaks, two
cedars, two willows, one hackberry,  one mesquite, one
pecan, one American elm, one unidentified elm, and one
unidentified species. Cores were collected from a height of
approximately 1.5 m above the ground surface.

Most of the trees that were sampled contained TCE and
cis-1,2-DCE.  A comparison of the results for two trees of
different species  (willow  and  cottonwood)  that  grow
immediately adjacent to each other with intertwining roots
showed  similar  TCE   concentrations   but   different
cis-1,2-DCE concentrations.  These data suggest that
concentration  differences  may  be  partly  a  result  of
tree-species differences.   As a  result, it is  practical  to
examine the data  by comparing concentrations within
individual species.  Generally, TCE concentrations found
within  individual species decreased in the directions  of
decreasing groundwater TCE concentrations.  Although
most trees contained more TCE than cis-1,2-DCE, in areas
where  the depth to  groundwater was  about one meter or
                                            less, willow, cottonwood, and American elm trees contained
                                            substantially  more cis-1,2-DCE  than  TCE.   The  data
                                            suggest the possibility that these trees promote in situ TCE
                                            dechlorination in areas where the depth to groundwater is
                                            shallow.   They also suggest that tree-core data can  be
                                            useful in  locating  areas of active dechlorination.  More
                                            cis-1,2-DCE than  TCE also was  found  in  the  only two
                                            cedars and the only pine that were tested.  These trees
                                            were in areas where the groundwater TCE concentrations
                                            were  greater  than   the  groundwater  cis-1,2-DCE
                                            concentrations, suggesting that either the  trees take  up
                                            cis-1,2-DCE more  efficiently than TCE or dechlorination of
                                            TCE occurs within  the trees. The depth to groundwater at
                                            these trees was up to 8 meters.  No TCE was found in trees
                                            that grow  in areas that contain no TCE in the groundwater.
                                            Additional information on the concentration TCE and 1,2-
                                            DCE measured in  trees within the study area is contained
                                            in   the    report   entitled   "Trichloroethene   and
                                            cis-1,2-dichloroethene concentrations in tree trunks at the
                                            Carswell Golf Course, Fort Worth, Texas (Vroblesky, 1998)
                                                      49

-------
A research team led by USEPA (Athens, GA) investigated
the kinetics  of  transformation  of TCE  for leaf samples
collected from seven trees (cedar, hackberry, oak, willow,
mesquite, cottonwood whip, cottonwood calipertree). Each
of the plant species investigated appears to have properties
that are effective in  degrading TCE. Specifically, all leaf
samples showed dehalogenase activity. Pseudo first-order
rate  constants were determined for the samples.   The
average and standard deviation for all seven rate constants
is  0.049  +-0.02 hr1 (Table4-3).  This corresponds to a
half-life of 14.1  hours. These kinetics are fast relative to
other environmental transport and transformation processes
with the exception of volatilization for TCE. As a result, it is
unlikely that degradation within the trees will  be the  rate
limiting step  in  a Phytoremediation system.  Additional
information on evidence of dehalogenase activity in  tree
tissue  samples  is  contained  in  a  report  entitled
"Dehalogenase  and  nitroreductase activity in selected tree
samples: Carswell Air Force Base" (Wolfe et al.,  1999)

Evaluate   gee-chemical   indices   of  subsurface
oxidation-reduction processes

It was hypothesized that  the  Phytoremediation system
would  promote the  biodegradation   of TCE  in  the
contaminated aquifer  by transforming conditions in the
aquifer from  aerobic to  anaerobic.  Specifically, it  was
thought that the planted system would introduce relatively
high concentrations of biologically available organic carbon
through  the  decomposition of root material  and  the
production of root exudates that would serve as the primary
substrate for microorganism  growth   and  subsequent
depletion  of  dissolved  oxygen.   Then,  the  anaerobic
microbial utilization  of this  natural  carbon source would
drive reductive decholorination of the dissolved TCE in the
aquifer (Wiedemeier and others, 1996). Thedechlorination
pathway   for  TCE    is   trichloroethene   ->
cis-1,2-dichloroethene + Cl -> vinyl chloride + 2CI -> ethene
+3CI. The efficiency of TCE degradation varies depending
on microbially mediated redox reactions (most efficient to
least efficient- methanogenesis, sulfate reduction, iron (III)
reduction, and oxidation). Thus, an accurate determination
of redox conditions in the aquifer could be used to evaluate
the potential for reductive dechlorination.

Determination  of   redox  conditions  or the   terminal
electron-accepting process (TEAP) in an  aquifer can be
accomplished  by   several  on-site  measurements of
groundwater  chemistry.  Detection  and measurement of
methane indicates that methanogenesis is occurring near
the well  sampled.   Measurement  of  the  redox  pairs
Fe2+/Fe3+ and SO42-/S2- using standard methods usually
distinguishes   between    iron   (lll)-reduction    and
sulfate-reduction processes.   If appreciable dissolved
oxygen (DO) (more  than 2 milligrams per liter (mg/L))  is
present in the groundwater, reductive dechlorination is an
unlikely process.  As these lines of evidence sometimes
conflict, the  measurement  of molecular hydrogen  (H2),
which is produced as an intermediate product of anaerobic
microbial  metabolism,  can be  an  effective method to
elucidate the predominant TEAP (Chapelle, 1993).

Data were collected to determine the concentrations and
distribution of contaminants, daughter products, and indices
of redox conditions in the aquifer.  Specifically, TCE and
cis-1,2-DCE concentrations were monitored, as were total
organic carbon  content,  methane  production,  sulfide
concentrations, ferrous and  ferric  iron ratios,  dissolved
oxygen  concentrations,  and  hydrogen gas  generation.
Samples  were  collected  from   monitoring  locations
upgradient of the plantations, within the plantations,  and
downgradient of the plantations. In addition, samples were
taken from a monitoring point immediately adjacent to the
mature cottonwood tree to provide insight into conditions in
the  aquifer  once  the  planted trees  have  matured.
Groundwatersampling locations are depicted in Figure 4-1.
(A lesson learned from this  data-collection effort is  that
metal on  groundwater-level  floats  and other downhole
instrumentation   can   interfere  with   hydrogen   gas
measurements.)

Table 4-4 summarizes  the results of the  VOC analyses
based on the  average  concentration within each of the
areas of the site (upgradient, plantations, downgradient,
mature  tree)  for  each  event.  An examination of the
summarized contaminant data indicates that there was a
general decrease in the concentration of TCE throughout
the demonstration site over the course of the study.  This
decrease, however,  does not appear to be predominantly
related to the establishment  of the  whip and caliper-tree
plantations.    This  is  because  a  decrease  in  TCE
concentration was observed in the upgradient monitoring
wells  as well as in  the wells within the plantations.  In
addition, the downgradient monitoring wells did not exhibit
a significant decrease in TCE concentration. The change
in TCE concentration within the study area over time may
be attributed to dilution from recharge to the aquifer and
volatilization of TCE from the water table.

The data also indicate that the TCE concentration in the
aquifer  beneath  the  mature  cottonwood  tree   was
significantly lowerthan elsewhere at the demonstration site.
In addition, DCE concentrations were much higher beneath
the mature tree than upgradient, within, or downgradient of
the planted trees.

Table 4-5 summarizes the ratio of TCE to cis-1,2-DCE for
each  area that  was sampled  (upgradient,  plantations,
downgradient, mature  tree).   The  ratio of  TCE to
cis-1,2-DCE can reveal subtle changes in the aquifer due to
biodegradation of TCE to its daughter product cis-1,2-DCE
that may be difficultto detect from concentration data alone.

The TCE to cis-1,2-DCE ratio in upgradient, plantation, and
downgradient wells indicate that there was a general
                                                     50

-------
    Table 4-3.   Pseudo first-order disappearance rate constants for the plant-leaf mediated transformation of
               TCE.
             Tree
             Cedar
             H ackberry
             Oak
             W illow
             M esq u ite
             Cottonwood (whip)
             Cottonwood (caliper)
TCE,  hr"
     0.052
     0.078
     0.067
     0.01 5
     0.059
     0.044
     0.027
Table 4-4. Average TCE and DCE concentration in monitoring wells.

nt
Dec-96
May-97
Jul-97
Jul-98
Sep-98


Up
Gradient a
818
771
709
480
490
TCE
ug/L


Down
Plantations'1 Gradient0
710
548
581
486
420
512
523
571
478
484


Mature
Treed
89
38
31
157
135


Up
Gradient
176
174
179
118
158
Cis-1,2-DCE
ug/L


Down
Plantations Gradient
121
114
157
109
172
101
109
143
98
145


Mature
Tree
160
230
240
150
217
Trans-1,2-DCE
ug/L


Up Down
Gradient Plantations Gradient
1 .2 2.4
3.6 1.1
3.6 3.0
1 .8 2.3
7.7 4.5
2.0
1.3
3.3
2.0
4.6


Mature
Tree
8.8
11.5
12.8
12.5
18.3
             (a) Upgradient monitoring points consist of wells 501, 502, 503, 513, and 518
             (b) Plantation monitoring points consist of wells 504, 505, 507, 508, 509, 514, 515, 524, and 525
             (c) Downgradient monitoring points consist of wells 526, 527, 528, and 529
             (d) Mature tree monitoring points consist of wells 510, 511, and 512
                                                    51

-------
Table 4-5. TCE to cis-1,2-DCE ratio.
      Event
               TCE/cis-1,2-DCE

    Up                            Down     Mature
Gradient    Plantations    Gradient    Tree
Dec-96
May-97
Jul-97
Jul-98
Sep-98
4.64
4.43
3.96
4.09
3.11
5.88
4.79
3.71
4.45
2.44
5.08
4.80
3.99
4.88
3.34
0.56
0.16
0.13
1.05
0.62
decrease  in  the   ratio   over  time  throughout  the
demonstration site. Again, the change in the ratio generally
cannot be attributed to the  planted trees because the
change was detected in the upgradient wells. An exception
to this pattern was observed in September 1998. The TCE
to cis-1,2-DCE ratio in the plantation wells at this time was
2.44, which  is  notably less than what was measured in
wells upgradient and downgradient of the planted area.
These data  may indicate  that reductive dechlorination
processes were beginning to become established beneath
the plantations by the end of the third growing season.

The data in Table 4-5 also indicate that significant reductive
dechlorination was occurring  in the vicinity of the mature
cottonwood tree during the demonstration period. The ratio
of TCE to cis-1,2-DCE was generally an order of magnitude
less than elsewhere  at the demonstration site. As will be
subsequently discussed, geochemical conditions beneath
the mature  cottonwood  tree appear  to  have been
transformed  from aerobic  to  anaerobic conditions that
support reductive dechlorination.

An investigation to determine whether the planted trees
were capable  of promoting  a shift in  the aquifer from
aerobic  to  anaerobic conditions during  the  three-year
demonstration  period was conducted by the USGS. The
results are summarized in Table 4-6. The study concluded
that the overall groundwater geochemistry  beneath the
plantations was beginning to change in response to the
planted trees by the peak of the third growing season.
Dissolved oxygen concentrations had decreased and total
iron concentrations had increased at the southern end of
the whip plantation by this time. This is in agreement with
the observed changes in the  ratio of TCE  to cis-1,2-DCE
and indicates that reducing conditions were  beginning to
support the biodegradation of TCE beneath this end of the
                                     whip  plantation.   It was also  concluded  that reducing
                                     conditions were present in the aquifer in the vicinity of the
                                     mature  cottonwood tree as  indicated by  low  dissolved
                                     oxygen  and high total iron concentrations,  as well as the
                                     detection of hydrogen and methane gases.  Additional
                                     information on this subject is contained in a report entitled
                                     "Phreatophyte influence  on reductive dechlorination  in a
                                     shallow aquifer contaminated with trichloroethene (TCE)"
                                     (Lee et al., 2000).

                                     Evaluate microbial contributions to reductive dechlorination

                                     To assess the mechanisms and rates of biodegradation in
                                     an aquifer, it is best to look at the spatial distribution of the
                                     different microbial populations on the sediment and in the
                                     pore water in addition to the concentrations and distribution
                                     of redox reactants and products in the groundwater.  As a
                                     result, a reconnaissance study of microbial activity in soil
                                     and   groundwater  beneath  the whip  plantation,  the
                                     caliper-tree plantation, and the  mature cottonwood tree
                                     near the site was conducted by the USGS in February and
                                     June of 1998. The purpose of the study was to determine
                                     the nature of the microbial community at the demonstration
                                     site and to determine  if the microbial  community  had
                                     evolved  into   one   that would support  the  reductive
                                     dechlorination of TCE and its  daughter products.  The
                                     presence of large populations of sulfate-reducing bacteria
                                     and methanogenic bacteria are indicative of environments
                                     that are favorable for reductive dechlorination.

                                     Results of the  study  are  summarized  in  Table 4-7.
                                     Specifically, Table 4-7 includes the Most Probable Number
                                     (MPN)  values for   physiologic  microbial  types  in  soil
                                     samples (S) and groundwater samples (W) throughout the
                                                   52

-------
Table 4-6.    Selected chemical data from wells used to define terminal electron accepting processes (TEAR) at the
              demonstration Site [mg/L, milligrams per liter; <, less than; nM, nanomolar per liter; pM, micromoles per liter; TEAR, terminal electron
              accepting process; E, estimated.
Area
Upgradient
Mature tree
Whip
Plantings
Caliper
Plantings
Between
Planted
Trees
Down-
gradient
d
0)
501
511
514
515
523


529

Dissolved Oxygen
(mg/L)
1997 1998
July Nov. Feb. June
3.5 3.0 3.0 4.7
1.1 0.7 0.9 0.8
2.5 1.2 0.7 1.7
3.0 2.5 1.5 2.9
3.5 3.5 3.0 4.5


3.5 4.0 3.0 2.7

Dissolved Sulfide
(mg/L)
1997 1998
July Nov. Feb. June
<0.001 <0.001 <0.001 <0.001
<0.001 0.005 0.007 <0.001
<0.001 0.120 0.056 <0.001
<0.001 <0.001 <0.001 <0.001
<0.001 <0.001 <0.001 <0.001


<0.001 <0.001 <0.001 <0.001

Total Dissolved Iron
(mg/L)
1997 1998
July Nov. Feb. June
0.1 <0.1 <0.1 <0.1
4.9 7.7 3.9 5.5
<0.1 <0.1 0.1 0.2
0.1 <0.1 0.1 <0.1
<0.1 <0.1 <0.1 <0.1


<0.1 <0.1 <0.1 <0.1

Area
Upgradient
Mature tree
Whip
Plantings
Caliper
Plantings
Between
Planted
Trees
Down-
gradient
d
1
501
511
514
515
523
529
Molecular Hydrogen
(nM)
1997 1998
July Nov. Feb. June
<0.05 <0.05 O.05 0.3
<0.05 <0.05 0.1 E 0.9E
O.05 12.2 0.7 0.5
O.05 0.8 <0.05 0.1
0.47 <0.05 <0.05 0.23
<0.05 <0.05 <0.05 0.5
Methane
(MM)
1997 1998
July Nov. Feb. June
<0.1 <0.1 <0.1 <0.1
5.1 7.5 24 15
<0.1 <0.1 <0.1 <0.1
<0.1 <0.1 <0.1 <0.1
<0.1 <0.1 <0.1 <0.1
<0.1 <0.1 <0.1 <0.1
TEAR
Reduction of dissolved oxygen
Methanogenesis
Iron (III) reduction
Reduction of dissolved oxygen
Reduction of dissolved oxygen
Reduction of dissolved oxygen
                                                            53

-------
Table 4-7. Results of microbial population survey ["S" denotes soil sample, "W" denotes water sample]
Borehole Aerobes

BUSGSTA001S 41
BUSGSTA001W 500
BUSGSTA002S 56
BUSGSTA002W 30
BUSGSTA003S 160,000
BUSGSTA003W 1 ,400
BUSGSTA004S 13,000
BUSGSTA004W <2
BUSGSTA005S ND
BUSGSTA005W ND
BUSGSTA006S 17,000
BUSGSTA006W 1,100
BUSGSTA0075 1 1 ,000
BUSGSTA007AW 5,000
BUSGSTA008S 60
BUSGSTA008W 40
BUSGSTA009S 430
BUSGSTA009W 170
BUSGSTA010S 2,200
BUSGSTA010W 500
BUSGSTA011S 370
BUSGSTA011W 140
BUSGSTA012S 1,700
BUSGSTA013AS <2
BUSGSTA013BS 1,300
BUSGSTA013BW 7,000

BUGSTA001
BUGGSTA002
BUGGSTA003
BUGGSTA004
BUGGSTA005
Denitrifiers Heterotrophic ton-reducers
Anaerobes
230 410 14
130 30 4
240 >300,000 430
80 >160,000 2,300
69 6,900 580
13 500 13
4,400 240 43,000
13 50 230
ND 4,800 3,700
ND 300 1,600
2,000,000 152,000 170
23 110 4
1,100 3,700 17
14 3,000 2
6 <2 <2
2 20 2
4 <2 <2
4 20 <2
22 54 <2
11 400 <2
50 280 <2
7 <2 2
23 370 <2
120 120 <2
<2 <2 2,100
350 800 40
Upgradient from trees in open space

Within whips, south side
Within caliper-trees, south side
Downgradient from trees in open space
Low spot west of mature cottonwood
Sulfate-
reducers
35
20
<2
4
210
20
15
60
19
70
300
26
24
2
<2
<2
37
<2
16
<2
9
<2
13
2,100
36
<2
Total
methanogens
<2
<2
37
<2
<2
<2
56
<2
48
4
650
<2
<2
30
<2
<2
<2
<2
<2
<2
<2
<2
<2
<2
<2
<2
BUGGSTA009


BUGGSTA010
BUGGSTA011
BUGGSTA012
BUGGSTA013A
Acetate-utilizing Formate-utilizing Hydrogen-utilizing
methanogens methanogens methanogens
<2
<2
<2
<2
<2
<2
24
<2
11
13
<2
<2
<2
2
<2
<2
<2
<2
<2
<2
<2
<2
<2
<2
<2
<2
Within whips, south side

Within whips, north side
Within caliper-trees, north
<2
<2
<2
<2
<2
<2
<2
<2
6
<2
<2
<2
<2
<2
<2
<2
<2
<2
<2
<2
<2
<2
<2
<2
<2
<2



side
<2
<2
<2
<2
<2
<2
<2
<2
37
2
170
<2
<2
80
<2
<2
<2
<2
<2
<2
<2
<2
<2
<2
<2
<2




In field behind house at 328 Tinker Dr.
Under mature cottonwood
in front of

house at 328 Tinker Dr., unsaturated zone
BUGGSTA006
Under mature cottonwood near site
BUGGSTA013B
Under mature cottonwood
in front of

house at 328 Tinker Dr., saturated zone
USGSTA007
BUGGSTA007A
BUGGSTA008
Under mature cottonwood near site
Under mature cottonwood near site
Within caliper-trees, south side















                                                        54

-------
study area. Microbial populations within the area of the tree
plantations (BUSGSTA002, 003, 008, 009, 010, and 011)
were similar to the background sites (BUSGSTA001 and
012) with the exception of locally  increased numbers of
anaerobic   microorganisms  and  the  presence   of
methanogenic microorganisms. These data suggest that the
microbial community appeared to be moving towards an
assemblage capable of supporting reductive dechlorination
by the third growing season.  The microbial population in
the area of the mature cottonwood tree near the site
(BUGSTA006 and 007) included a vigorous community that
supported both hydrogen-oxidizing and acetate-fermenting
methanogens. This active anaerobic population is assumed
to be responsible for the decrease in TCE concentration
and  the  generation of daughter products beneath  the
mature cottonwood tree. A sediment sample from beneath
the mature tree contained identifiable acidic compounds,
including phenol, benzole acid, and acetic acid, which  are
common intermediates observed in anaerobic ecosystems
where complex organics are undergoing biodegradation
and are consistent with the complex organic root exudates
at this location. These compounds are most likely acting as
electron donors for the reductive dechlorination of the TCE
beneath  the  mature cottonwood  tree.  The microbial
population downgradient of the plantations (BUGSTA004)
contained an anaerobic community structure  similar to
populations  present beneath the plantations.  Additional
information on the subject of microbial dechlorination in the
study area can be found in the report entitled "The role of
microbial  reductive  dechlorination  of  TCE  at   the
phytoremediation site at the Naval Air Station, Fort Worth,
Texas" (Godsy et al., 2000).

Although the microbial data suggests that the Plant system
may be capable  of modifying  the subsurface microbial
community in the aquifer beneath the planted trees to one
that can begin supporting reductive dechlorination of TCE,
TCE degradation rates cannot be determined from the data.
In order to  determine the degradation rate  of TCE in
subsurface sediments at the demonstration site, laboratory
microcosms  were established using sediment  and water
samples  collected from  locations  in and near the site.
Preliminary results indicate that TCE was converted to
cis-1,2-DCE  in a microcosm created from sediment taken
from beneath  the mature cottonwood tree  and water
collected from beneath the caliper trees. The first order
kinetic rate of TCE disappearance  in this microcosm was
0.34 day1 (Ean Warren, USGS, written commun.,  2000).
Further microcosm experiments are planned.

4.5 Discussion

The  SRWCGT system at the Carswell Golf  Course is a
low-cost, easy to implement, low-maintenance system that
is consistent with a  long-term  contaminant  reduction
strategy.  The system  produces  virtually  no process
residuals and requires minimal maintenance. Maintenance
requirements include occasional pruning and irrigation. The
system  is  an  "evolving"  process  that  increases its
effectiveness  over  time.   The  following  discussion
summarizes the predicted effectiveness of the system as
configured at the Carswell Golf Course site and presents
recommendations for implementing a  similar system at
other sites.

The  SRWCGT system  is  useful for intercepting  and
remediating a chlorinated ethene contaminant plume.  The
technology  uses two mechanisms to achieve this goal;
hydraulic  influence  and in-situ  biologically  mediated
reductive dechlorination.  Hydraulic influence involves the
interception and usage of contaminated groundwaterby the
trees.    Biologically-mediated  reductive  dechlorination
involves  the generation  of subsurface  biodegradable
organic matter by the tree root systems, which drives the
microbial communities  in the  underlying  aquifer  from
aerobic to anaerobic ones that are capable of supporting
reductive dechlorination of TCE.

With respect to hydraulic influence, the trees in both the
whip and caliper-tree plantations at the demonstration site
began to use  water from the  aquifer and reduced the
volume of contaminated groundwater leaving the site during
the three-year demonstration. The maximum reduction in
the outflow of contaminated groundwater that could be
attributed to the trees was approximately 12 percent and
was observed at the peak of the third growing season.  The
reduction in the mass flux of TCE across the downgradient
end of the treatment system at this time was closer to 11
percent because TCE concentrations were slightly higher
during the third growing season than  at baseline.  The
maximum observed drawdown of the water table occurred
near the center of the treatment system at this time and was
approximately 10 centimeters. A groundwater flow model
(MODFLOW) of the demonstration site indicates that the
volume of water that was transpired from the aquiferduring
the peak of the third growing season was probably closerto
20 percent of the initial volume of waterthat flowed through
the site  because there was an  increase in groundwater
inflow to the site due to an increase in the hydraulic gradient
on the upgradient side of the drawdown cone.

Tree-growth and  root-growth data  collected from the
demonstration site are consistent with the observations of
hydraulic influence of the trees on the contaminated aquifer.
Trees  in  the  whip plantation,  which  were planted
approximately 1.25 meters apart, were starting to approach
canopy closure by the end  of the  third growing season.
This observation indicates that the trees were transpiring a
significant amount of water  at this time.   (A plantation
approaches  its maximum transpiration potential once it
achieves a closed canopy because a closed canopy limits
leaf area.) The caliper trees were planted 2.5 meters apart
and although the plantation was not as close to achieving
a closed canopy, individual calipertrees transpired just over
twice the waterthat individual whips transpired. As a result,
                                                    55

-------
the volume of water that was  transpired by  the  two
plantations was similar because there were half as many
caliper trees as whips. Tree roots in both plantations had
reached the water table (275 cm for the whips and 225 cm
for the caliper trees) by the second growing season.

There were no data collected during the demonstration that
favored the planting of calipertrees overthe less expensive
whips. The physiologically-based model PROSPER, which
was used to predict out-year transpiration  rates at the
demonstration site, indicates that the whip and caliper-tree
plantations will  eventually transpire  a similar amount of
water- 25 to 48 centimeters per growing season depending
on  climatic  conditions, soil moisture, and  root growth.
Forty-eight  to   fifty-eight  percent   of  this   predicted
evapotranspiration is  expected to be derived  from the
contaminated aquifer (saturated zone) regardless  of the
planting strategy.  In general, the closer trees are planted,
the sooner a plantation may achieve  closed  canopy.
However, it is important to consider the increased chance
for disease when trees are closely spaced. There is a body
of literature on short rotation wood culture that can be used
to guide decisions with regard  to  tree spacing  in  a
SRWCGT system (see Appendix B, Vendor's Section 5.0).

Since the SRWCGT system had not achieved maximum
hydraulic control during the period  of demonstration,  a
modeling approach was  used to  make  predictions  with
regards to out-year hydraulic control. The groundwater flow
model  indicates that once the  tree plantations  have
achieved a closed canopy, the reduction in the volumetric
flux of contaminated groundwater across the downgradient
end of the site will likely be between 20 and 30 percent of
the initial amount of waterthat flowed through the site.  The
actual  amount of water that will be transpired from the
aquifer by the tree plantations will be closer to 50 to 90
percent of the volume of waterthat initially flowed through
the site.  The discrepancy between the reduction  in the
volumetric outflow of groundwater and the volume of water
transpired  from the aquifer  can be attributed  to the
predicted increase in groundwater inflow to the site and the
release of water from storage in the aquifer.  No hydraulic
control was observed during  the dormant season from
Novemberto March and no hydraulic control is predicted for
future dormant seasons.

In general, the  amount of  hydraulic control that can be
achieved by a Tree system is a function of site-specific
aquifer conditions.  A planted  system can be expected to
have a greater hydrologic affect on an aquifer at a site that
has  an initially  low volumetric flux of groundwater as
opposed to  a   site  where the  flux of  contaminated
groundwater is  significantly greater.  The parameters of
hydraulic  conductivity,   hydraulic  gradient,  saturated
thickness, and  aquifer width  in  the  treatment zone all
contribute to the volumetric flux of groundwater through a
site.   The  horizontal   hydraulic conductivity at  the
demonstration site in Fort Worth, Texas is approximately 6
meters/day. The natural hydraulic gradient is close to two
percent and  the saturated thickness of the aquifer  is
between 0.5 and 1.5 meters. Volume of water in storage in
an aquifer will also affect system performance. Although
the current study did not investigate the effect of aquifer
depth;   it  is  possible  that a  greater  percent  of total
evapotranspiration could be derived from an aquifer with a
shallower water table.

When designing for hydraulic control at a Phytoremediation
system, it is  important to keep the remediation  goals  in
mind. In other words, it may not be desirable to achieve full
hydraulic control at a site if full control would adversely
affect the groundwater/surfacewatersystem downgradient
of the site. At the demonstration site in Texas, the receptor
is Farmers Branch Creek, which has very low flow (less
than 1  cubic foot per second) during the summer months
(peak transpiration).  The optimal performance at such a
site may be to keep the plume from discharging into the
creek  without drying  up the  creek,  particularly since
hydraulic control is only one mechanism that contributes to
the cleanup of a groundwater plume by Phytoremediation
System. A groundwater flow model of a potential site is
ideal for addressing such design concerns.

With respect to the fate of the contaminants that were taken
up into the planted trees, TCE  and its daughter products
were commonly detected in tissue samples of roots, stems
and leaves. Generally, there was an increase overtime in
the percentage  of planted trees in which the compounds
were detected. Stem tissue generally exhibited the greatest
diversity and  concentration of chlorinated compounds.  It
was concluded  that the planted cottonwood trees have
properties that are effective in degrading TCE. Specifically,
the leaf samples showed dehalogenase  activity.  An
investigation into the kinetics of transformation of TCE for
leaf samples  concluded that it is unlikely that degradation
within  the  trees will  be the rate-limiting   step  in  a
Phytoremediation system.  As a result, it may be  better to
use species that are native to a  proposed area  rather than
to  use  genetically altered plants that  are  designed to
enhance metabolism of TCE.

With   respect   to   biologically-induced   reductive
dechlorination, there is evidence that the aquifer beneath
the planted  trees was beginning to support anaerobic
microbial communities capable of biodegradation of TCE
within three years of planting.  Specifically, microbial data
from soil  and  groundwater samples indicate that the
microbial community beneath the planted trees had  begun
to  move towards an assemblage capable  of supporting
reductive dechlorination during  the demonstration period.
In addition, dissolved oxygen concentrations had decreased
and total iron concentrations had increased at the southern
end of the whip plantation where the water table is closest
to land surface.  The ratio of TCE to cis-1,2-DCE had also
decreased at this location beneath the whip plantation,
which suggests that the shift toward anaerobic conditions in
                                                     56

-------
this part of the aquifer was beginning to support the
biodegradation of TCE. Significant contaminant reduction
by this mechanism, however, had not occurred across the
demonstration site by the end of the demonstration period.

Data from the aquifer beneath the mature cottonwood tree
near the planted site support the conclusion that reductive
dechlorination can  occur beneath cottonwood trees with
established root systems. The ratio of TCE to cis-1,2-DCE
beneath  the  mature  tree  was  typically one order of
magnitude  less than  elsewhere at  the site  during the
demonstration.  The microbial population in the area of the
mature cottonwood tree included a vigorous community that
supported both hydrogen oxidizing and acetate fermenting
methanogens. This active anaerobic population is assumed
to be  responsible for the decrease in TCE concentration
and the  generation  of daughter products beneath the
mature cottonwood tree.

The data collected during the demonstration are insufficient
to conclude when significant reductive dechlorination will
occur beneath the planted trees.  Data collected during the
fifth dormant season after the period of demonstration had
ended indicate  that the aquifer was generally anaerobic
beneath the planted trees while it was aerobic upgradient
and downgradient of the trees. It is reported in the literature
that   hybrid  poplar   plantations  sequester  significant
quantities of soil carbon due to tree root growth by the time
they are six years old.  It is likely that this increase in soil
organic carbon  would  be  enough to promote reductive
dechlorination of dissolved TCE  in the underlying aquifer,
including during the dormant season. The only conclusive
information on the future timing of significant reductive
dechlorination in the aquifer, however, can be extrapolated
from  the mature  tree. The  mature cottonwood was
approximately 20 years old during the demonstration; as a
result, the planted  site will  likely  reach this  level of
contaminant reduction within this time frame.

Even though reductive dechlorination was observed around
the mature cottonwood tree, the presence of TCE daughter
products, as well as  residual TCE,  indicate that the
reductive dechlorination process has not fully mineralized
the contaminants of concern to innocuous compounds.
There is no field  evidence from this study that suggest
complete in-situ biodegradation of TCE and its daughter
products can be achieved.

In summary, the  first three  growing  seasons  at the
Phytoremediation  system demonstration site resulted in a
reduction in the mass of contaminants moving off site. The
maximum observed  reduction in the mass flux of TCE
across the downgradient end of the demonstration site was
11 percent.  An increase in the  hydraulic influence of the
trees and the reductive dechlorination of TCE in the aquifer
is expected as  the  system matures. A  solute transport
model would be  necessary  to determine the  relative
importance of hydraulic control, reductive dechlorination,
and sorption in the out years.
                                                     57

-------
                   SECTION 5
OTHER TECHNOLOGY REQUIREMENTS
        Regulation
5.1    Environmental
       Requirements
State and local regulatory agencies may require permits
prior to implementing  a phytoremediation technology like
the Short Rotation Woody Crop Groundwater Treatment
(SRWCGT) system. Most federal permits will be issued by
the authorized  state agency.   Depending upon  the
characteristics of the  site and the nature of a particular
application, the state may  also require a Treatment,
Storage, and Disposal (TSD) Permit for on-site storage of
a hazardous waste for greater than 90 days.  An air permit
issued  by the state Air  Quality Control Region may be
required if air emissions in excess of regulatory criteria, or
of toxic concern, are anticipated. Discharge of wastewater
is highly unlikely during SRWCGT. However, wastewater
discharge permits may be required if any such wastewater
were to be discharged  to a POTW.   If remediation is
conducted at a Superfund site, federal agencies, primarily
the U.S. EPA, will provide regulatory oversight. If off-site
disposal of contaminated waste is required, the waste must
be taken to the disposal facility by a licensed transporter.

Section 2  of this report  discusses the environmental
regulations that may apply to the SRWCGT  process.

5.2    Personnel  Issues

The  number of personnel  required  to implement  the
SRWCGT technology is  largely dependent on the size of
the area to be treated. Large sites, requiring extensive site
preparation and assembly of a large  irrigation system may
require  several individuals  (inclusive  of  contractors);
especially if there  are constraints on time.   For smaller
sites, requiring minimal  site  preparation, as few  as  two
people may be needed for the actual treatment technology
related activities. After site setup, labor associated with a
tree-based phytoremediation system such  as the  one
demonstrated at the Carswell Golf Club is limited generally
to  monthly or bimonthly ground maintenance tasks  and
monitoring and sampling events. These tasks could be
accomplished by one individual over a one to three  day
period. Labor associated with  monitoring and sampling
events could be reduced somewhat through  automated
data collection using  data loggers. Data loggers would
enable real-time remote access of information pertaining to
tree  growth,  hydraulic conditions and soil  moisture.
Monitoring and sampling events will  likely involve tree
measurements (i.e., tree height, canopy width and tree
trunk  diameter), additional water level measurements,
calibration checks on automated  monitoring systems,
groundwater sampling, rhizosphere soil sampling and tree
tissue sampling

Estimated labor requirements for a hypothetical 200,000 ft2
site are discussed in detail in Section 3 of this report.

For most sites, the personnel protective equipment (PPE)
for workers will include steel-toed shoes or boots, safety
glasses,  hard hats,  and  chemical  resistant  gloves.
Depending on contaminant types, additional PPE (such as
respirators) may be required.  Noise levels would usually
not be  a  concern for  an application  of  a  SRWCGT
technology.  However some equipment used  for crop
cultivation and vegetative clearing and regrading (i.e. tillers,
mowers, chain saws, etc.) could create appreciable noise.
Thus, noise levels should be monitored forsuch equipment
to ensure that workers are not  exposed to noise levels
above the time weighted average of 85 decibels over an 8-
hourday. If this level is exceeded and cannot be reduced,
workers would be required to wear hearing protection.

5.3    Community Acceptance

Potential hazards to a surrounding community may include
exposure to  particulate matter that becomes  airborne
during regrading and tilling operations.  Air emissions of
VOCs is possible if those contaminants are also present in
the soil.  Particulate air emissions can be controlled by dust
suppression measures.

Overall,  there  are   few  environmental   disturbances
associated with SRWCGT. No appreciable noise, beyond
that  generated  by the short term use of agricultural
equipment,  is ever anticipated  for the majority of the
treatment time.  A  fence may be desirable  to  keep
unauthorized visitors from entering the site.
                         58

-------
                                            SECTION 6
                                    TECHNOLOGY STATUS
This section discusses the experience of the developers in
performing treatment using the Short Rotation Woody Crop
Groundwater  Treatment (SRWCGT)System.   It  also
examines the capability of the developers  in using the
technology at sites with contaminant mixtures.
regimes are  being conducted.  Bench  and pilot  scale
investigations testing the ability of trees to handle other
recalcitrant   compounds  like   perchloroethylene,
1,1,1,-trichloroethane, and perchlorate have also been
conducted with  promising results.
6.1    Previous Experience

In addition to the demonstration performed on chlorinated
VOCs at the Carswell Golf Club site, the Aeronautical
Systems  Center Engineering  Directorate Environmental
Safety and Health  Division  has extensive experience in
site investigations  and remediations at hundreds of site
nationwide involving a variety of metals, fuels, VOCs, and
other DoD unique compounds. Currently other field scale
site  investigations   and   remediations   employing
phreatophytic trees in a variety of climates and hydraulic
6.2    Scaling Capabilities

The planting approach employed  in this demonstration
have been used by the pulp and paper industries worldwide
at much larger scales than that of the demonstration site.
Several documents  developed  by the Department  of
Energy's Oak Ridge National Laboratory Biomass/Biofuel
Group offer recommendations with regard to the selection,
planting, care, and harvesting of various trees and grasses
amendable to short rotation energy and  fiber crops.
                                                  59

-------
                                         References Cited
Chapelle, F.H., 1993, Ground-water microbiology and geochemistry:  New York, John Wiley, 424 p.
Chapelle, F.H., 2000, Ground-water microbiology and geochemistry: New York, John Wiley, 2nd Ed.
CH2M Hill, 1984, Installation restoration program records search for Air Force Plant 4, Texas - IRP Phase I
Dietz, A.C., and J.L. Schnoor. (2001) Phytotoxicity of chlorinated aliphatics to hybrid poplar: Environmental Toxicology and
       Chemistry, 20 (2), 389-393.
Dickmann, D.I., and Stuart, K.W., 1983. The Culture of Poplars in Eastern North America: Michigan State University Press,
       East Lansing, Michigan
Eberts, S. M., and others, In Press. Phytoremediation - Transformation and Control of Contamination: Steven McCutcheon
        and J. L. Schnoor eds., Wiley & Sons.
Eberts, S. M., Schalk, C.W., Vose, J., and Harvey, G.J. 1999. Hydrologic effects of cottonwood trees on a shallow aquifer
       containing trichloroethene:  Hydrological Science and Technology, vol. 15, no. 1-4, p. 115-121.
Godsy, M.E., Warren,  E., Paganelli, V.V., 2000. The Role of Microbial Reductive Dechlorination of  TCE at the
       Phytoremediation Site at the Naval Air Station, Fort Worth, Texas - Final Report: U.S. Geological Survey, Menlo
       Park, California.
Goldstein, R.A., Mankie, J.B., and Luxmoore, R.J., 1974, Documentation of Prosper, A model of atmosphere-soil-plant
       water flow: East Deciduous Forest Biome  EDFB-IBP 73-9, 75 p.
Gore, J.A., 1985, The Restoration of Rivers and Streams - Theories and Experience: Butterworth Publishers.
Hendrick, R.L., 1998, Root Biomass and Extent  in Populus Plantations Planted for Phytoremediation Purposes - Final
       Report:  D.B.Warnell School of Forest Resources, University of Georgia.
Lee, R.W., Jones, S.A., Kuniansky, E.L., Harvey, G.J., Sherwood Lolar, B., and Slater, G.F. 2000. Phreatophyte Influence
       on Reductive Dechlorination in a Shallow Aquifer Contaminated with Trichloroethene (TCE): International Journal
       of Phytoremediation, vol. 2, no. 3, p. 193-211.
Licht,  L.A., and Madison, Mark,  1994, Proceedings of the 87th meeting of the Air  and Waste Management Association:
       Cincinnati, Ohio
McDonald, M.G. and Harbaugh, A.W., 1988, A  modular three-dimensional finite-difference ground-water flow model:
       Techniques of Water-Resources Investigations of the United States Geological  Survey, Book 6, Chapter A1,
       [variously paged].
Rivers, G.A., Baker, Jr., E.T, and Coplin, L.S., 1996, Geohydrologic units and water-level conditions in the Terrace Alluvial
       aquifer and Paluxy aquifer, May 1993 and  February 1994, near Air Force Plant 4, Fort Worth area, Texas: U.S.
       Geological Survey Water-Resources Investigations Report 96-4032, 13 p.
Schnoor, J.L., 1997, Phytoremediation: Ground-Water Remediation Technologies Analysis Center Technology Evaluation
       Report TE-98-01, 37 p.
Stomp, A.M.,1993, Genetic improvement of tree species for remediation of hazardous wastes: Vitro Cellular Development
       Biology, v.29, p.227-232.
U.S. Army Corps of Engineers, Kansas City Division, 1986, Investigation of groundwater pollution at Air Force Plant 4, Fort
       Worth, Texas
Vose, J.M., Swank, W.T., Harvey, G.J., Clinton, B.D., and Sobek, C. 2000. Leaf Water Relations and Sapflow in Eastern
       Cottonwood (Populus deltoides Bartr.) tees planted for phytoremediation of a groundwater pollutant: International
       Journal  of Phytoremediation: vol. 2, no. 1, p. 53-73.
Vose, J.M., and Swank, W.T., 1992, Water Balances, in D.W.Johnson, Lindberg, S.E., eds., Atmospheric Deposition and
       Forest Nutrient Cycling,  a Synthesis of the Intergrated Forest Study, Ecological Studies 91: Springer-Verlag, New
       York, p. 27-49.
Vose, J.M., and Swank, W.T., 1998, Sap Flow Rates in Large Trees at the Carswell  Naval Air Station - Final Report: USDA
       Forest Service,  Southern Research Station.
Vroblesky, D.A., 1998, Trichloroethene and Cis-1-2,-Dichloroethene Concentrations in Tree Trunks at the Carswell Golf
       Course, Fort Worth, Texas: U.S. Geological Survey, Columbia, South Carolina.
                                                    60

-------
Wiedemeier, T.H., Swanson, M.A., Moutoux, D.E., Gordon, E.K., Wilson, J.T., Wilson, B.H., Kampbell, D.H., Hansen, J.E.,
       Haas, Patrick, and Chapelle,  F.H., 1996, Technical protocol for evaluating natural attenuation of chlorinated
       solvents in groundwater - draft - revision 1: Air Force Center for Environmental Excellence Technology Transfer
       Division, Brooks Air Force Base, San Antonio, Texas, p. 2-19.
Wolf, N.L., Ou, T.Y., Tucker, J., Smith, L, Lewis, S., and McCutcheon, S., 1999. Dehalogenase and Nitroreductase Activity
       in Selected Tree Samples, Carswell Air Force Base: Prepared for Restoration Division Acquisition Environmental
       Management, Wright Patterson Air Force Base, Ohio.
                                                    61

-------
                Appendix A



DATA Used to Evaluate Primary Project Objective

-------
Appendix A
DATA USED TO EVALUATE PRIMARY PROJECT OBJECTIVE (SEE TABLE 4-1)
Hydraulic Gradient Across Downgradient End of Planted Area

Baseline (November 1996)
Peak 2na Season (1997)
Late 2na Season (1997)
PeakS™ Season (1998)
Late 3m Season (1998)
Peak 4tn Season (1999)


Baseline (November 1996)
Peak 2na Season (1997)
Late 2na Season (1997)
PeakS™ Season (1998)
Late 3m Season (1998)
Peak 4tn Season (1999)
Water Table Altitude - Well 522
1 79.93 m above sea level
180.13 m above sea level
180.02 m above sea level
179.76 m above sea level
179.67 m above sea level
179.83 m above sea level

Saturated Thickness - Well 526
1.59m
1.50m
1.56m
1.55m
1.56m
1.54m
Water Table Altitude - Well 529 Distance Between Wells Gradient3
1 78.96 m above sea level
179.19 m above sea level
179.06 m above sea level
178.88 m above sea level
178.75 m above sea level
1 78.9 m above sea level
Cross Sectional Area Along
61 m
61 m
61 m
61 m
61 m
61 m
Downgradient
Saturated Thickness - Well 527 Saturated Thickness
0.80m
0.80m
0.76m
0.73m
0.75m
0.71 m
1.22m
1.20m
1.24m
1.22m
1.23m
1.22m
0.0159
0.0154
0.0157
0.0143
0.0150
0.0153
End of Planted Area
- Well 528 Ave. Thick.
1.20m
1.17m
1.19m
1.17m
1.18m
1.16m







Aquifer Width
70m
70m
70m
70m
70m
70m







Cross Sectional Area
84m"
82m"
83m"
82m"
83m"
81 m"
                                                        Average of TCE Concentrations In Wells Along Downgradient End of Planted Area
Baseline (November 1996)
Peak 2na Season (1997)
Late 2na Season (1997)
PeakS™ Season (1998)
Late 3™ Season (1998)
Peak 4tn Season (1999)
TCE Concentration - Well 526
         564 ug/L
         570 ug/L
            NA
         530 ug/L
         490 ug/L
            NA
TCE Concentration - Well 527
         610 ug/L
         685 ug/L
            NA
         540 ug/L
         470 ug/L
            NA
TCE Concentration - Well 528   Ave. Cone.
         232 ug/L              469 ug/L
         350 ug/L              535 ug/L
            NA                   NA
         380 ug/L              483 ug/L
         460 ug/L              473 ug/L
            NA                   NA
[m above sea level, meters above sea level; m, meter; m  , square meter; ug/L, micrograms per liter; NA, data not available]
a Slight differences between reported measurements and calculated gradients are due to rounding errors introduced during conversion
 of units from feet to meters for presentation in this table; calculated values were derived from measurements in original units of feet

-------
                                  Appendix B

                                Vendor's Section
Note:  Information contained in this appendix was provided by the technology vendor and has not
      been independently verified by the U.S. EPA SITE Program
                                       B-l

-------
          APPENDIX B - Air Force Experience and Recommendations


This section describes steps to be taken for implementing phytoremediation and establishing a short
rotation woody crop. Knowledge of site-specific soil and climate conditions before planting can often
decrease the probability of planting failure.  This section has extensively utilized information developed
by or for the Department of Energy's Biomass/Biofuel Program,  Short Rotation Woody Crops Operations
Working Group, and the Salix Consortia of the New York State Energy Research and Development
Authority. Readers will also find additional lessons learned in the restoration of riparian zone vegetation,
points of contact, helpful web sites, references to technical reports and handbooks, and sources of hybrid
poplar, eastern cottonwoods, and willows are included in this section.

B.I     Introduction
Vascular plants have been on Earth over 400 million years. Flowering plants first emerged about 140
million years ago.  Plants survive by exploiting their surroundings as they compete for light, nutrients and
water.  Plants have evolved various strategies that allow them to exploit a given ecological niche. Some
plant groups are stress tolerators that can survive high salt and metal levels. Other plant groups compete
"best" by growing rapidly. Because plants cannot readily move themselves from sites having adverse
conditions, over time plants have developed the necessary biochemical processes to tolerate a variety of
man made and natural carcinogens, mutagens, and teratogens. Some vegetation even has the ability to
make compounds  such  as chloromethane. There are more than 3,200 chlorinated, fluorinated, and
brominated chemicals produced by living organisms and natural combustion processes (Gribble).
Chlorine is actually an essential element for plants.  In fact, natural organohalogen compounds play an
essential role in the survival of many organisms.  Trees, shrubs, grasses, flowers and vegetables can
readily handle low levels of halogenated hydrocarbons such as trihalomethane found in chlorinated
drinking water. Another indication of this tolerance is that members of Populus and Salix families are
often found in shallow ground water contaminated  by trichloroethylene and its daughter products
dichloroethylene, and vinyl chloride.  Plants can do this because they have dehalogenase and mixed
function oxidase enzymes needed to transform low  levels of halogenated hydrocarbons.

Plants form the basis for agriculture and forestry. Plants have a long history of providing us with fuel,
fiber, oils, medicines (quinine, digitalis, opiates), poisons (strynine, hemlock, etc.) and food. Perhaps the
group to first exploit plants for environmental purposes was the Incas who planted alders in the 10th
century to stabilize their planting terraces in Peru (Moore). Alders also helped maintain the fertility of the
soil by fixing nitrogen.  The Chinese have used trees since the 12th century to stabilize slopes and prevent
erosion, while the Dutch have used trees to stabilize their earthen dikes for several hundred years.  The
ability of trees to act as pumps was noted in the late 19th century when Eucalyptus trees were planted in
Italy and Algeria to dry up marshes. The incidence  of malaria in these areas subsequently decreased.

Phytoremediation is a new term, but given the diverse and long history of plant exploitation through
out world history  it can hardly be considered a new idea. Phytoremediation is currently being practiced
by some professionals with backgrounds in agronomy, biochemistry, hydrology, chemical engineering,
sedimentology and industrial hygiene  to clean up shallow groundwater and soil contaminated with
various metals and organics.  Because phytoremediation is in its commercial infancy, the people who
employ phytoremediation have often designed projects with methodologies developed from personal
experience. This knowledge is considered to be proprietary and zealously guarded even though much of
this information is already in the public domain.  About 30 years ago the United States Department of
Energy embarked on a program to  grow plants as a  source of fiber and fuel in response to the  Arab oil
embargoes of the early 1970's.  The outcome of millions of dollars and thousands of man years of effort
is  in an extensive body of public domain information on the physiology and development of short
rotation woody crops.  The information about individual species or clones that are most suitable  for a
given region, how to plant, control weeds, when and how often to fertilize, how to recognize and control
plant pathogens and other pests, and how to harvest is all in the public domain. This public domain
information gives detailed guidance on how to select and prepare potential sites. Research and

                                              B-l

-------
development is also currently being conducted in the Netherlands, Finland, Denmark, Sweden, Italy,
Australia, and the United Kingdom.

If shallow ground water contaminated with low level nitrates, phosphates, hydrocarbons, or chlorinated
hydrocarbons is encountered at a site that is suitable to growing a short rotation woody crop,
consideration should be given to employing the technology developed by the US DOE before employing
any proprietary deep planting methods.  This information is available on-line at the Biomass Information
Network or through regional biomass energy programs.

Before initiating a phytoremediation corrective action  for shallow ground water, it is imperative to
determine if natural attenuation processes (i.e.,biodegration, dispersion, sorption, or volatilzation) are
able to achieve site-specific remedial objectives within a comparatively reasonable time frame.  If site-
specific natural attenuation processes are at work and capable of reducing mass, toxicity, mobility or
volume of halogenated hydrocarbons in the soil and groundwater, the site in question MAY NOT be
considered a candidate for a phytoremediation intervention.

There are several currently available protocols and tools that have been developed by the United States
Air Force, United States Geological Survey and Environmental Protection Agency to evaluate the fate  of
chlorinated hydrocarbons in the ground.  The Technical Protocol for Evaluating Natural Attenuation of
Chlorinated Solvents in Groundwater has undergone extensive external and internal peer and
administrative review by the U.S. EPA and U.S. Air Force. The intent of the Technical Protocol for
Evaluating Natural Attenuation of Chlorinated Solvents in Groundwater is to provide guidance for data
collection and analysis to evaluate monitored natural attenuation through biological processes  It is
available from the National Technical Information Service.  Another useful resource is BIOCHLOR
Natural Attenuation Decision Support System available from the U.S. EPA Center for Subsurface
Modeling Support (CSMoS).  To obtain the BIOCHLOR program and user documentation go to the
CSMoS web site at www.epa.gov/ada.csmos.html.  Tables B.I and B.2 show the  parameters of interest
when determining if natural attenuation is likely to occur in a given aquifer.
                                             Table B.I
            Analytical Parameters and Weighting for Preliminary Screening for Anaerobic Biodegradation Processes'
Analysis
Oxygen*
Oxygen*
Nitrate*
Iron II*
Sulfate*
Sulfide*
Methane*
Oxidation Reduction
Potential* (ORP)
against Ag/AgCI
electrode
pH*
TOC
Temperature*
Concentration in
Most
Contaminated
Zone
<0.5 mg/L
>5 mg/L
<1 mg/L
>1 mg/L
<20 mg/L
>1 mg/L
<0.5 mg/L
>0.5 mg/L
<50 millivolts (mV)
<-100mV
5 pH >9
> 20 mg/L
>20°C
Interpretation
Tolerated, suppresses the reductive pathway at higher
concentrations
Not tolerated: however, VC may be oxidized aerobically
At higher concentrations may compete with reductive
pathway
Reductive pathway possible; VC may be oxidized under
Fe(lll)-reducing conditions
At higher concentrations may compete with reductive
pathway
Reductive pathway possible
VC oxidizes
Ultimate reductive daughter product, VC accumulates
Reductive pathway possible
Reductive pathway likely
Optimal range for reductive pathway
Outside optimal range for reductive pathway
Carbon and energy source; drives dechlorination; can be
natural or anthropogenic
At T >20°C biochemical process is accelerated
Value
3
-3
2
3
2
3
0
3
1
2
0
-2
2
1
                                               B-2

-------
Table B-1 continued
Carbon Dioxide
Alkalinity
Chloride*
Hydrogen
Hydrogen
Volatile Fatty Acids
BTEX*
Tetrachloroethene
Trichloroethene*
DCE*
VC*
1,1,1-Trichloroethane*
DCA
Carbon Tetrachloride
Chloroethane*
Ethene/Ethane
Chloroform
Dichloromethane
>2x background
>2x background
>2x background
>1 nM
<1 nM
> 0.1 mg/L
> 0.1 mg/L








>0.01mg/L
>0.1 mg/L


Ultimate oxidative daughter product
Results from interaction between CC>2 and aquifer minerals
Daughter product of organic chlorine
Reductive pathway possible, VC may accumulate
VC oxidized
Intermediates resulting from biodegradation of aromatic
compounds; carbon and energy source
Carbon and energy source; drives dechlorination
Material Released
Material released
Daughter product of PCE
Material released
Daughter product of TCE.
If cis is > 80% of total DCE it is likely a daughter product
1,1 -DCE can be chemical reaction product of TCA
Material released
Daughter product of DCE
Material released
Daughter product of TCA under reducing conditions
Material Released
Daughter product of DCA or VC under reducing conditions
Daughter product of VC/ethene
Material Released
Daughter Product of Carbon Tetrachloride
Material Released
Daughter Product of Chloroform
1
1
2
3
0
2
2
0
0
2a/
0
Q3/
0
f^a/
0
2
0
2
2
3
0
2
0
2
* Required analysis, a/  Points awarded only if it can be shown that the compound is a daughter product (i.e., not a constituent of the source




NAPL).
               Table B.2 Interpretation of Points Awarded During Screening Step 1
Score
Oto5
6 to 14
15 to 20
>20

Interpretation
Inadequate evidence for biodegradation* of chlorinated organics
Limited evidence for biodegradation* of chlorinated organics
Adequate evidence for biodegradation* of chlorinated organics
Strong evidence for biodegradation* of chlorinated organics
^reductive dechlorination
                                                  B-3

-------
        Review Available Site Data
         If Site Data are Adequate
      Develop Preliminary Conceptual
                 Model
                                              Gather any Additional
                                                     Data
                                             Necessary to Complete
                                            the Screening of the Site
          Screen the Site using the
                Procedure
          Presented in Fiqure B.2
                   Are
                Screening
                 Criteria
                  Met?
                                                     Are
                                                Sufficient Data
                                                  Available to
                                                   Properly
                                                 creen the Site?
                                                           Evaluate Use of
                                                              Selected
                                                             Additional
                                                          Remedial Options
                                                          Including Source
                                                             Removal or
                                                               Source
                                                            Control Along
                                                                with
                                                               Natural
                                                             Attenuation
     Does it
   Appear That
Natural Attenuation
      Alone
   Meet Regulatory
     Criteria?
        Perform Site Characterization
            to Evaluate Natural
               Attenuation
          Refine Conceptual Model
                   and
          Complete Pre-Modeling
               Calculations
                                     Hydraulic
                                   Containment
              Simulate Natural
                Attenuation
            Using Solute Fate and
              Transport Models
                                    Phytoremediation
  Enhanced
Bioremediation
         verify Model Assumptions
                   and
             Results with Site
         Use Results of Modeling and
          Site-Specific Information in
                     an
             Exposure Pathways
                  Analysis
              Will Remediatio
              Objectives Be Met
               Without Posing
                Unacceptable
                              Engineered Remediation
                                    Required,
                             Implement Other Protocols
                                                                                        Perform Site Characterization
                                                                                        to Support Remedy Decision
                                                                                      	Making	
                               Assess Potential For
                               Natural Attenuation
                                With Remediation
                                 System Installed
                                                                                        Refine Conceptual Model
                                                                                                  and
                                                                                         Complete Pre-Modeling
                                                                                              Calculations
                                                                                                            culati
                           Simulate Natural Attenuation
                             Combined with Remedial
                              Option Selected Above
                              Using Solute Transport
                          	Models	
                                                                                        Verify Model Assumptions
                                                                                                  and
                                                                                            Results with Site
                                                                                          Characterization Data
                                                                                                            [erj^a,
                                                                                                    Use Results of Modeling
                                                                                                             and
                                                                                                   Site-Specific Information in
                                                                                                    an Exposure Assessment
                                                                            NO
Fo Potential ^^^
&fK.o\s,"> ^^^
^S^ fe
YES ^
Develop Draft Plan for
Performance
Evaluation
Monitoring Wells and
Long-Term Monitoring
*

^


Determine
Measures to be
Combined with
MNA
^•v ToPo
\Recei
^ ^^
YES
                                       Does
                                Revised Remediation
                                   Strategy Meet
                                   Remediation
                                 Objectives Without
                                      Posing
                                Unacceptable Risks
                                                  Findings
                                                and Proposed
                                                 Remedy in
                                                 Feasibility
                                                   Study
Figure B. 1  Natural attenuation of chlorinated solvents flow
                                                                            B-4
                                                                            (Flowchart adapted from Technical Protocol for
                                                                            Evaluating Natural Attenuation of Groundwater)

-------
       Analyze Available Site Data
          Along Core of Plume
      to Determine if Biodegradation
              is Occurring
Collect More Screening
        Data
                                                                                                 Engineered Remediation Required,
                                                                                                    Implement Other Protocols
Locate Source(s) and Potential
Points of Exposure. Estimate
Extent of NAPL, Residual
and Free-Phase
1
r
Determine Groundwater Flow and
Solute Transport Parameters Along
Core of Plume Using
Site-Specific Data; Porosity and
Dispersivity May Be Estimated
^
r
        Estimate Biodegradation
             Rate Constant
1
r
Compare the Rate of Transport
to the Rate of Attenuation Using
Analytical Solute Transport Model
                                   NO
                Does it
          Appear That Natural
       Attenuation Alone Will Meet
          Regulatory Criteria?
       Evaluate Use of Selected
      Additional Remedial Options
     Alonq With Natural Attenuation
Proceed to
Figure B.1
      Perform Site Characterization
     to Evaluate Natural Attenuation
                  ceed to
               Figure B.1
Figure B.2  Initial screening process flow
              (Technical Protocol for Evaluating Natural Attenuation
               of Chorinated Solvents in Groundwater)
                                                              B-5

-------
               1,1 - DCE
                                            PCE
                                            TCE
                                                                       ci) Chloride Atom
                                   Carbon Atom

                                   Hydrogen Atom

                                   Single Chemical
                                   Bond
                                   Double Chemical
                                   Bond
c/s -1,2, - DCE
trans-1,2 - DCE
                                                             Complete Mineralization
                                           Ethane
                                                  (Technical Protocol for Evaluating Natural Attenuation
                                                  of Chlorinated Solvents in Groundwater)
Figure B.3  Reductive dehctlogencttion of chlorinated
                                             B-6

-------
If the presence of any significant natural attenuation processes cannot be established from tables
B.I and B.2, the next step is to determine if the site is a candidate for the establishment of a short
rotation woody crops. To determine if a site is viable for the establishment of a short rotation
woody crop, a thorough understanding of site-specific hydrology and agronomic factors is
essential. Failure to consider site- specific hydrologic factors such as pH, depth to groundwater
and pattern of seasonal precipitation, and agronomic factors such as the nutrient status and
presence of salts,  soil compaction, and clay hardpans can lead to disappointment. While trees
may grow at the site, there may be insufficient biomass to influence the geochemistry and
hydrology of the groundwater.   The establishment and management of a short rotation woody
crop usually has the following goals:

     1) Elimination of competing vegetation.
     2) Maintenance of site productivity
     3   Maximum net energy gain.
     4) Maximum biomass for minimum cost
Whether a shallow groundwater site is suitable for development of short rotation crops such as
cottonwoods, hybrid poplar, willow, eucalyptus, or other energy crops, requires consideration of
operational factors such as location of the site, depth to groundwater, soil properties and climate. The sites
should have sufficient area to plant the required biomass.  Planting a few rows of trees may have subtle
influences on groundwater flow. Keep in mind that the mere observation of diurnal variations in a water
table does not imply hydraulic control. Potential sites should be level or gently sloping in order to use
mechanical planting means whenever possible.  If a site is near an airport or flight line, determine if
Federal Aviation Administration (FAA) restrictions may limit height of trees. Small cuttings placed in
the ground can eventually become 100 foot safety impediments to the operation of aircraft.  The presence
of large stones or construction  debris may make large scale planting difficult and damage equipment.
Another site factor is wet heavy clays that can make machine access difficult or impossible.

Hardpans are compacted soil that can tend to impair the ability of plants to send deep roots. Compaction
of soil  can result from vehicular traffic and natural cementation. If hardpans are present, deep ploughing
may be necessary.  There are vendors that specialize in ripping  soil to correct this condition.
     Site soil characteristics are also important for successful establishment of biomass.  There are 16
nutrient elements that are essential for the growth and reproduction of plants. Thirteen of these essential
elements may be supplied by the soil or supplemented by fertilizers. Plants obtain carbon, hydrogen, and
oxygen from the air and water. Important soil properties are moisture and drainage, texture alteration,
depth, pH, and fertility. Information on the characteristics of soil in a given  county can be found from
the Soil Conservation Service of the Department of Agriculture. These reports provide a general idea of
the soils and climatic conditions in an area.

While soil surveys are  an excellent starting point, it is strongly recommended that additional soil testing
be conducted.  Soil testing can provide site-specific answers to concerns about pH, salts and plant nutrient
availability (i.e.,  nitrogen, phosphorus, potassium) and micronutrients such as manganese, iron, boron,
zinc, copper, molybdenum, and chlorine. The first step is to select a laboratory to conduct the required
tests.  When selecting  a soil testing laboratory, ask if they participate in a proficiency testing or quality
assurance program. Ask to see the results of the most current evaluation.  Most laboratories provide
instructions on how to  collect a representative soil sample. Laboratories offer a variety of soil analysis
options. A routine analysis consists of pH, nitrates, phosphorus, potassium, calcium, sulfur, and
conductivity. Additional testing options available  at extra cost (typically $15  to $30) are analysis for
micronutrients such as  zinc, iron, copper, and manganese, detailed salinity testing, organic matter, texture,
and boron.

 A soil sample for testing should represent a uniform area. Past land use, drainage, slope, and differences
in texture and color are important.  Areas at the proposed site in which plants appear to be doing poorly
should be tested separately.  It  is important to use  a clean rust-free tool to avoid contaminating the soil
sample with iron. Collect the sample from the soil surface to the depth desired. A clean plastic pail is a

                                              B-7

-------
good container within which to mix soil samples. Avoid using galvanized or brass containers to prevent
zinc contamination. Many soil testing facilities provide plastic bags for containing soil samples.

The pH of the soil is important because pH influences the availability of nutrients. Nitrogen is probably
the nutrient that most often limits plant growth. Soil nitrogen is present in three major forms: elemental
nitrogen, organic nitrogen, and nitrogen in fertilizers. Phosphorus (P) is an essential part of the process of
photosynthesis.
Micronutrient deficiencies are most likely to limit plant growth under the  following conditions:
     1)  Highly bleached acid sandy soil
    2)  Muck soils
    3)  Soil high in pH or lime content
    4)  Soils that have been intensively cropped and heavily fertilized with macronutrients

Some soil testing facilities provide only the results of the analysis while others also make specific
recommendations based on the tests results for the crop to be grown.  If recommendations are not
provided by the laboratory, contact your local forester, county or state cooperative extension service for
guidance.  Once site-specific soil test recommendations have been made follow them. Do not apply more
plant nutrients than recommended.  This can create a nutrient imbalance that may adversely affect the
plants being grown.
                                          TABLE  B.3

       FACTORS THAT AFFECT THE PRODUCTIVITY OF SOILS FOR HARDWOODS
SOIL PROPERTY
Physical
Moisture availability during
growing.
Nutrient availability
Aeration
BEST CONDITIONS
Deep,>4ft, soils without pans.
Loose, porous, friable soils (bulk
density< 1 .4 g/cc) . Undisturbed
site with no recent cultivation or
pasturing
Water table 3-6 ft. Level ground
or lower slopes. No flooding or
floods only early spring.
Undisturbed site or cultivated <5
years. Organic matter (A-
horizon) >3%, especially in
sandy soils. A-horizon (topsoil)
>6 in. Young, well-developed
profile. Source of basic
(calcareous) parent material in
rooting zone. pH in rooting zone
5.0-7.5.
Wet by running water only in
early spring. No mottling to 2ft.
Soil color black, brown or red.
WORST CONDITIONS
Shallow, < 1.5 ft, soils with
plowpans or natural cemented
pans. Strongly compacted, tight
soils (bulk density >1.7 g/cc)
pasturing for >20 years .
Water table <1 ft or > 10 ft.
Ridgetops, mounds, dunes.
Prone to flooding anytime.
Recent intensive cultivation for
>20 years. Organic matter (A-
horizon) <1% A-horizon
(topsoil) absent or <3 in. Old,
highly leached profile. No basic
(calcareous) parent material in
rooting zone. pH in rooting zone
<4.5 or>8.5.
Swampy, stagnant or
waterlogged condition much of
year. Mottled to surface. Soil
gray in color.
Table B.3  from The Culture of Poplars in Eastern North America by Donald Dickmann
Salt Stresses
Saline soils refer to a soil that contains sufficient soluble salts to impair its productivity. A soil is saline if
the solution extracted from a saturated soil paste has an electrical conductivity of 4 decisiemens per meter
Briggs). Saline soils  are typically found in arid and semi-arid regions. Saline soils are rare in humid
                                              B-8

-------
environments except in areas where the soil has been exposed to marine environments. In humid
environments, soluble salts often migrate downward into the groundwater. Another source of salt to
plants is from road de-icing salt spray that splashes or drifts onto the roadside. Plant damage from
roadside salt spray is linked with the amount of salt applied and the traffic volume.

High salinity often limits plant growth by inducing water stress (Neuman). Plants exhibit a wide range of
salt tolerance.  Physiological esponses to salinity tend to be species specific (Newman).  Some plants are
very tolerant of salts (i.e., halophytes) while  others are intolerant. Planting poplars or willows in areas
with high soil salinity  can be problematic (Briggs/Thomas).  Soluble salts can produce harmful effects to
plants by increasing the salt content of the soil solution and by increasing the degree of saturation of
exchangeable materials (USDA Agricultural Handbook 60). The soluble salts that occur  in soils consist
of various amounts of sodium, calcium, magnesium and the anions chloride and sulfate (USDA
Agricultural Handbook 60). The originof most salts are the primary minerals found in the soil and in the
exposed parent rock of the Earth's crust.

Individuals attempting to plant vegetation in saline soils must carefully select vegetation  that is
appropriate. It is imperative that the planting material be adapted to the site-specific conditions.  Failure to
chose plant material phenotypically adapted to site conditions can often result  in a planting failure
(Briggs). Matching salinity tolerance to site-specific soil characteristics can be difficult  (Briggs).
Willows and poplars used for riparian revegetation were noted by Briggs to start exhibiting adverse
effects when the salinity levels reach  2,000mg/l.

Flood Tolerance
Plants exhibit a wide range of tolerance to flooded or wet soil conditions. A site that is subjected to
periodic flooding or wet soil conditions can impose very difficult conditions on most vascular plants.
Some plants are much more tolerant of flooding and wet soil  conditions than others. The fundamental
difference between well drained and flooded conditions in the soil are directly and indirectly related to
depletion office oxygen (Whitlow).  The absence of oxygen creates a reducing environment. Plants that
are not adapted to wet or flooded soils exhibit reduced shoots and root growth and drop their leaves.
Trees near rivers and streams are often subjected to flooding and wet soil conditions.  Some plants can
withstand  complete inundation for months at a time, while others plants are completely flood intolerant.
Flood tolerant plants have developed the  anatomical, morphological and biochemical characteristics to
withstand  flooding and anoxic conditions. Factors that influence flood tolerance are the seasonal timings,
duration, and depth of flooding. The seasonal timing of a flood is critical to the survival  of trees and
shrubs. Flooding when plants are dormant is usually not harmful.  Flood tolerant  and even intolerant
trees like the tulip tree can withstand flooding when they are dormant.  The time during which a flood
occurs in the growing season, along with the depth and duration that an area is flooded can have a
significant impact on the survival of developing vegetation. Within a given species, greater damaged and
lower survival are associated with increased depth and duration of flooding.
Impacts of Temperature

Plants have an optimal temperature range at which they grow best. Many plants are susceptible to
damage from freezing temperatures.  The ability to withstand cold temperatures often limits the range of a
given plant or even specific clones within a given species.  Moving plant material north from southern
latitudes can often be problematic.  One  1976 study by Ying et. al. in Nebraska found that cuttings from
Mississippi, Arkansas, and Texas suffered significant dieback during the winter. Ying  et. al. concluded
that trees from southern latitudes were more prone to injury in the winter because they retained their
leaves late into the growing season. Another reason why plant material adapted to southern latitudes  fail
when moved hundreds of miles north is that they tend to leaf out earlier in the spring and are prone to
damage from late frosts. To avoid these problems people attempting to establish phytoremediation
plantations should know the origin of the plant material they purchase.
Wind
Living material grows in response to stresses that occur (Wood). The adaptive growth hypothesis states
that a tree  will grow only sufficiently strong to resist the forces that have occurred during its growth
history (Wood). Wind is a ubiquitous component of the environment  (Telewski). The mechanical failure

                                              B-9

-------
of a tree is usually the result of wind rather than gravity (Vogel). Attempts to inhibit the growth of
shallow lateral roots to enhance the growth of deep roots should be done with the knowledge that greater
damage to tree stand productivity may be incurred from wind toppling in areas subject to high velocity
winds. Wind can have profound effects on the growth and form of trees (Wind and Trees).  Damage to
short rotation woody crop plantations from high velocity winds is often  an overlooked risk factor.  Just as
there are clonal differences in susceptibility to flooding and salinity, another abiotic stress is the
mechanical stress from high velocity winds.  Research by Harrington has shown that poplar clones proved
resistant to toppling are associated with above and below ground characteristics.  Harrington found that
risk factors include trees that had less root system
                     Maryland Wind Toppled Hybrid Poplar (Photo Courtesy of Harry
                     ConiDton USEPA")

development in the wind ward quadrants.  Wind toppling was the least at the closest spacing. This
seems to be due to reducing crown sway.  Toppling was also found by Harrington to be reduced in
poly clonal plots which was believed to be the result of more rapid stand differentiation or reduction in the
"domino effect" by inclusion of more wind resistant clones in the mixture.  Hybrid poplars deep planted
in Maryland with engineering controls to inhibit shallow lateral roots had almost a 20% incidence of
toppling in the wake of Hurricane Floyd (Compton).

Biotic Stressors
Insects, fungi, viruses, bacteria, and gnawing animals can threaten the success and reduce the productivity
of poplar and willow short rotation woody crops.  Many readily available poplar trees are extremely
susceptible to certain insect pests and diseases (Ostry). Symptoms of insect infestation and disease in
poplar trees can be seen in off color foliage, missing foliage, branch die back, and cankers. Disease
susceptibility among poplar clones is usually expressed by the second growing season (Hansen).
Septoria cankers is more prevalent in the eastern United States and melansporia rust is more common in
the western states. Trees severely stressed by one disease may ultimately be predisposed to other
damaging agents such as other fungi, wood boring insects, and wind breakage. This predisposal is the
case with trees severely affected by stem cankers (Hansen). While there are hundreds of insects and plant
pathogens of poplars and willow, only a few are considered to be potentially dangerous (Ostry).  Perhaps
the most serious disease among poplar short rotation woody crops are stem canker diseases. Trees with
stem canker infection often appear with dead, swollen, or shrunken patches on their stems (Dickmann).
Sometimes the canker will stop and the wound will heal overtime, but somtimes other fungal and
bacterial infections will occur. Ready guidance about insect, disease and animal pest infestation of poplar
trees is available in the USDA Agricultural Handbook 677.  This handbook describes and illustrates with
color photos the major insect, animal pests, fungal, viral, and bacterial diseases of poplars. This
handbook enables growers of poplars to identify the causes of a problem should one develop. Being
armed  with this knowledge of the expected impact of the condition, control measures warranted, and
what control measures are available enables a grower to effectively manage his crop. A careful
                                             B-10

-------
examination of the affected trees should be made and compared to illustrative and descriptions within
USDA AgriculturalHandbook 677. If a grower cannot determine the exact cause of problem with this
handbook, it is advisable to consult a forest entomologist or forest pathologist (Ostry).  Pest management
information can also be obtained to Forest Service Offices listed at the end of this section.

Willows and cottonwood ecosystems are characterized by high diversity of both plants and animals
(Briggs). Wildlife and vegetation have co-existed for millions of years in an on going struggle for
survival by herbivores and plants.  However, unlike declining water tables which can have a severe effect
on trees wild life rarely significantly contributes to the decline of trees in a riparian ecosystem (Briggs).
Some species like deer, rabbits, moles and beavers, however, can have an impact on newly established
short rotation woody crop and riparian revegetation projects (Briggs).  Moose, white tailed deer and
beaver are all capable of eating large quantities of poplar and willow tree vegetation. Moose are only a
problem to poplar plantations in northwest Minnesota and Sweden (Nester).  Rodents such as moles, rats,
and mice can also harm young shoots by gnawing off bark and damaging above ground irrigation lines.
Rabbits and moles can be problematic in establishing poplar and willow plantations. In the Swedish
experience, establishment of willow and poplar plantations can cause the existing population of rabbits
and hares to significantly increase due to the ready abundance  of food (Christersson). The best method for
controlling rabbits and rodents has been to control weeds from the start of the plantation. When weeds
are eliminated, moles, mice, rats, gophers and rabbits are vulnerable to potential predators.

Four hundred years ago there were approximately 60 to 100 million beavers in North America.  The
demand for pelts and heavy trapping pressure so severely impacted the beaver population of North
America that by the  1800's beavers were extinct east of the Mississippi River. Today, however, beavers
are making a come back through protective legislation and a lack of predators.  Beavers are now moving
into urban environments and near urban water ways, making their presence known in such diverse areas
as Detroit, Ft Worth, and Washington B.C. to name a few.  Beavers are gregarious and can usually be
found in family groups. Young beavers leave their families at about two years. They find an area where
young poplars grow and then they build a dam. Upstream they usually build a lodge and collect poplar
branches for winter feed.  Beavers are quite strong and can readily gnaw down and remove a thirty foot
cottonwood tree almost over night. Beavers are also quite difficult to trap alive. Trapping beavers and
moving them off site can require large amounts of time and effort and is usually only temporarily
successful. Trapping beavers for their pelts is simply not as profitable as it used to be (Isebrands). Some
states also frown on releasing live trapped beaver on to public  lands. Efforts to control beavers include
erecting regular fences and employing solar or battery power electric fences. Another approach has been
to employ plastic shelter tubes 2-5  feet tall that allow the cuttings to grow. These preventative measures
sometimes are successful but more often fail.  Beavers at the Carswell Golf Course Phyto site have been
an annual concern since 1996. Numerous trees have been damaged, but over all tree mortality to date has
been very little. Willows and poplars readily sprout from cut or gnawed stumps. Virtually all poplars and
willows coppice readily after beaver damage, harvesting or damage by fire (Dickmann). Since beavers
are here to stay, beaver damage to established poplar and willow phytoremediation plantations should be
taken in stride. Beaver damaged established poplar and willow trees will usually recover. While the
above ground biomass is gone, subsurface biomass is still usually capable of establishing new above
ground biomass.   It has been our experience at Carswell that below ground short rotation woody crop
biomass can still  drive iron reducing conditions and reductive dechlorination of TCE in the absence of
significant above ground biomass.
                                             B-ll

-------
Beaver Damaged Trees      photo by Greg Harvey, USAF
                           B-12

-------
                                                                                             : .(V.. 'L'-M -"" .32 L-
 HYBRID POPLARS
SWITCHGRASS AND
REED CANARY GRASS
 BLACK LOCUST
 HYBRID POPLARS
 SILVER MAPLE
 WILLOW
HYBRID POPLARS
EUCALYPTUS
   EUCALYPTUS
   HYBRID POPLARS
   BLACK LOCUST
   SILVER MAPLE
   SORGHUM
   SWITCHGRASS
   REED CANARYGRASS
    'SYCAMORE
     SWEETGUM
     POPLARS
     BLACK LOCUST
     SWITCHGRASS
     TROPICAL GRASSES
     SORGHUM
EUCALYPTUS
     Map Courtesy of Virginia Tolbert (Oak Ridge National Laboratory)
           For trees to reach their full genetic potential, plantation managers need to be able to select disease
           resistant clones and recognize various problems as they arise (Hansen). The goal of short rotation woody
           crops is to achieve and maintain high productivity (Mitchell).  The Department of Energy has screened
           approximately 125 different plants as candidates for short rotation woody crops for fiber and fuel. The
           Department of Energy has found that certain species perform better than others in various regions of the
           United States. This finding is illustrated in the attached map of screened biomass candidates. After
           selecting the appropriate tree or trees for a given region, the next step is to select specific clones that give
           superior performance in a plantation.  An understanding of short rotation woody crop production, stress,
           and ecophysiology has allowed plantation managers to achieve optimal clone-site matches at numerous
           sites (Mitchell). Tree breeders try to find clones that are adaptable to large areas (Hansen).  Few clones
           however, are sufficiently stable for all situations in regions with varying soils and climates.  Clones with
           desirable qualities such as superior growth rate and disease resistance can be selected from nursery
           screening trials. Promising clones selected from nursery screening trials are then planted in field trials.

           Field trials are expensive and take several years to complete. Field trials have been conducted for hybrid
           poplars and cottonwoods by the United States Forest Service and for willows by the Salix Consortium of
           New York.  Because of the time and expense involved, most poplar clones have not undergone field
           testing in all locations where they are now planted.  The hybrid poplar field trials were conducted in
           eastern Ontario, the Pacific Northwest, and North Central sections of North American. A program for
           improving cottonwood was begun by the United States Forest Service in the early 1960's after it became
           apparent that hybrid poplars from the Northeastern United States and Europe did not perform well
                                                     B-13

-------
(Mohn).  The results of the extensive hybrid poplar field trials pointed to clone stability throughout the
North Central States and eastern Ontario, but site-specific stability in the Pacific Northwest (Hansen).

The greater stability of clones in the North Central eastern Ontario regions is believed to be due to a
narrower climate range (Hansen). U.S. Forest Service found  that clones DN 34, DN 17, and DN 182 in
the North Central United States had reasonable disease resistance and biomass across a range of sites.
Interestingly, Edward Hansen of the Forest Service noted that clone DN 182 performed well on sites with
harsh dry conditions and also performed well on good sites with wetter conditions. But clones DN 34 and
DN 17 that performed well on good sites were often affected more severely by disease on harsh sites.
This observation was also noted in the Pacific Northwest field trials with other clones. The reason for the
variability observed in the Pacific Northwest is believed to be that climate and soils vary greatly with
distance from the ocean, elevation and which side of the Cascades Range.

 The United States Forest Service has made several recommendations with respect to selecting clones for
a site.  First,  potential tree growers should make clone selections based on their performance of half their
projected rotation.  Growers should not assume that because a tree grew eight feet the first year and is
healthy that it is the "super tree" for a given area (Hansen). Second, poplar clones  should be selected
based on their performance in plantations.  Singular trees grown in an open field are not a good indicator
of plantation performance (Hansen). Additional information on hybrid poplar performance can be found
in the USDA Research Paper NC-320 North Central United States  in Field Performance of Populus in
Short Rotation Intensive Culture Plantations in  the North-Central U.S.  Some vendors offer cuttings in
various lengths ranging from 8 to 36 inches or more. It is often possible to get volume discounts by
ordering large quantities.  Typically the longer the cutting the more expensive it is. Prices for Spring
2000 for  8-9 inch hybrid poplar cuttings were approximately $ 0.25 each for quantities of 25 to 100 to
approximately $0.16 for orders of 5000 cuttings or more.  Spring 2000 prices  for 18 inch cuttings were
about $0.30 and 36 inch cuttings were about $0.50.  Shipping and handling charges are usually extra.
Because of the relative inexpense of cuttings in the establishment of a plantation one should order more
cuttings than  one anticipates planting. When ordering cuttings, preference should be given to male clones
which do not produce seeds.  Female poplar trees  can produce large amounts of small wind borne seeds.
These seeds can clog air conditioner heat exchangers, cover outdoor pools, and create other maintenance
roblems for people living near poplars (Baldridge). Vendors of hybrid poplars in the Pacific Northwest
and North Central United States are listed at the end of this section.

Willows are another species that have potential as a short rotation woody crop. Willows are easy to
propagate, resprout readily after cutting, and are not susceptible to Septoria canker (White ). Septoria
canker has caused serious damage to hybrid poplar planted in New York and harvested on 5-10 year
rotations (White). The field trials of various willow clones for biomass production was  initiated in 1987
in central New York State by the State University of New York College of Environmental Science and
Forestry, the University of Toronto, and the Ontario Ministry of Natural Resources.  The most promising
clone, willow clone SV1, in ultra-short rotation was found to yield 16 oven dry tons per hectare per year
during the fifth growing season (Kopp).  White's group found that fertilization significantly increased the
rate at which  clones reached their maximum biomass production. Kopp also noted large clonal variation
in biomass production and survival. For further information concerning the availability of specific clone
willow cuttings contact Timothy Volk of the State University of New York College of Environmental
Science and Forestry, One Forest Drive Syracuse, New York 13210 tavolk@mailbox.syr.edu.  There are
two commercial sources of non-proprietary eastern cottonwood cuttings for sale to the public.  One is
the Crown Vantage cottonwood clonal nursery at Fitler, Mississippi and the other  is Ripley County Farms
in Doniphan,  Missouri. Additional  information on specific eastern cottonwood clones can also be found
at the end of this section.

Storage
Careful site preparation and selection of appropriate planting material can be compromised by several
things. Perhaps the simplest is improper storage of cuttings.  Dormant cuttings improperly stored often
fail to grow.  For best results cuttings must be protected from heating and moisture loss  and should be
stored in  sealed double plastic bags in a cold room or refrigerator just above 0 degrees C or 32 degrees F

                                              B-14

-------
(Dickmann).  It is important to warm cuttings slowly before they are planted (Dickmann). This is done by
moving them to a room kept at 2 to 3 degrees C for a week or two prior to planting (Dickmann). Cuttings
used for short rotation woody crop establishment in the North Central United States are usually 20 to 30
cm in length; 50 cm cuttings are the norm in the South and Pacific Northwest (Dickmann). Optimum
diameters for cuttings range from 10-20 mm (Dickmann). On sites where moisture is limited in the upper
most soil layer, the longer the cutting the better. Of course, it is seldom necessary to plant cuttings in
excess of three feet long in the absence of hard pans. Cuttings should have numerous buds and be free of
mechanical and insect damage (Dickmann).  Cuttings that are spindly or have sprouted roots in storage
should not be planted (Dickmann).  For best results, cuttings should be warmed for 5-10 days prior to
planting (Hansen) When soaking, it is important to make sure buds point up (Hansen).
Planting
The "best" time to plant cuttings is when soil temperature reach 50 degrees F (Hansen). In the North
Central United States, planting usually occurs between mid April and early June (Hansen). In warmer
places like the Carswell Site in Ft. Worth, Texas cuttings can be planted from late February to mid-May.
Prior to planting, determine the location of above and below ground utilities, check if local ordinances
prohibit some tree species, and decide if irrigation is necessary to supplement the natural soil moisture.
Poplars and willows grow quickly and can obstruct the view of traffic if placed improperly. Special care
should be exercised along roadways and intersections. Most cities encourage the planting of long-lived
and low maintenance trees, but some local governments prohibit planting shorter-lived high maintenance
trees. For example, the city of Ft. Worth prohibits planting hackberry, sycamore, silverleaf maple,
mulberry, Arizona Ash, cottonwood, Siberan Elm and other high maintenance trees along city roadways.
If a city prohibits a particular tree, a variance can often be obtained when there is an appropriate reason
for using this type of tree.

Proper soil moisture and control of weeds are critical for a successful first year. The soil should be moist
and the cuttings kept wet and protected from the sun while planting.   Exposing cuttings to the sun for a
prolonged period can significantly damage them prior to planting. It is important to remember to plant
cuttings with their buds pointing up (Hansen).  Buds must point up because this is the direction in which
the tree will ultimately grow.  Cuttings should also be oriented as close as possible to vertical
(Dickmann).  Cuttings must also have at least one bud exposed above ground (Hansen). Any air gaps
around the cutting should be filled by pushing the soil against the cutting (Hansen). It is possible to plant
cuttings by hand or to machine plant them.  Usually  small scale sites of a few acres are planted by hand
and larger sites are planted by machine. Hand planting  rates are reported by Hansen to be 3
acres/day/person and machine planting rates are 20 acres/day/three person crew.  The trees at the Carswell
Site were spaced at 8 by 8 feet in the five gallon bucket trees and 8 by 4 feet in the whip plantation.
Spacing of the trees is often influenced by the number of years old they will be at harvest. The shorter the
cutting cycle  or rotation the closer the spacing of the trees. For poplars a cutting cycle of one to three
years can have spacing of 2 by 2 to 4 by 4 feet.  A rotation of 15 years can be spaced at 15 by 15 to 20 by
20 feet. For willows even closer spacing can be employed using the Swedish double row planting system.
Keep in mind that closely spaced, genetically identical trees are prone to insect infestations and fungal
diseases. Trees that are widely spaced apart, however, may take  longer to root to the water table. A
successful tree spacing design in phytoremediation achieves a balance where tree spacing promotes deep
rooting without fostering conditions that encourage plant pathology problems.

Harvesting several rotations of a short rotation woody crop from a site can often result in a depletion of
nutrients. Several different approaches to nutrient management for short rotation woody crops have been
advocated (Heilman). The conservative approach is not overly concerned with the depletion of nutrients
as long as production of above ground biomass is not significantly reduced (Heilman).  The cost
conservative school applies fertilizer only when soil fertility begins to impact growth. The other approach
to fertilizing short rotation woody crops seeks to maintain fertility at a high steady state (Heilman).  Here
fertilizers are applied to not only supply nutrients but also to increase soil fertility (Heilman). The main
drawback to  this approach is  the expense of maintaining high nitrogen levels and the risk of leaching
nitrogen into the groundwater. Another drawback in phytoremediation applications of short rotation
woody crops  is that maintaining optimum levels of water and nutrients through irrigation and fertilization
can decrease subsurface biomass (Dickmann). If trees are given optimum levels of nutrients and water it

                                              B-15

-------
   is unlikely that the tree will expend the resources to develop a large root system to explore the subsurface.
   Decreasing subsurface biomass may have an impact on the amount of carbon that is available for
   reductive dechlorination. Another problem with the liberal application of nutrients like nitrate is most
   studies show fertilizers are rarely 100% utilized by plants (Heilman). The liberal application of fertilizer
   in excess of what trees or other plants can use can cause leaching into the groundwater; this may impact
   the geochemistry of the groundwater making conditions unfavorable to reductive dechlorination. For
   these reasons, fertilizer applications to short rotation woody crops grown to phytoremediate shallow
   groundwater contaminated with halogenated solvents should only be done when foliar (leaf) level
   nitrogen levels fall below 3%. For further information about when to fertilize hybrid poplar platations
   obtain USDS Research Paper NC-319-A Guide to Determing When to Fertilize Hybrid Poplar
   Plantations.
                                                         1-bud  exposed
                                                  - >,    '    •.     .
          buds   point  up
Planted cutting.
Photo Courtesy of E. A. Hansen, et. aL, 1992.
                                        B-16

-------
WEED CONTROL

Weed control is imperative during the establishment phase of a short rotation woody crop. The extensive
experience of foresters throughout the world has shown that uncontrolled weeds can quickly compromise
the success of a short rotation woody crop.  Eliminating weeds reduces competition for light, water, and
nutrient and also results in less cover for rodents (Handbook of Short Rotation Woody Crops}. Omitting
post planting weed control for hardwoods results in poor survival and growth and sometimes complete
failure.
To insure a successful tree plantation, some short rotation woody crop foresters endeavor to have a 90%
weed-free plantation in year one, 80% weed-free in year two, and 70% weed-free in year three. As the
trees get bigger in the later years, they are better able to compete for light and water effectively,
controlling the weeds.

There are a number of ways to control weeds by cultivation, mulching, and herbicides.  One 1984 study
by Edward Hansen Research Note NC-317, Forest Service - U.S.P.A., titled,  Weed Control for
Establishing Intensively Cultured Hybrid Poplar Plantation compared eight weed control methods that
included cultivation, herbicides, and a legume cover by themselves or in various combinations.  The
weed control treatments were as follows:
     Glyphosate
     Linuron - Legume
     Linuron - Glyphosate
     Linuron - Cultivation
     Cultivation
     Legume
     Furrow Cultivation
     Furrow Cultivation
Hansen concluded that there was no difference in survival among poplar trees for six of the eight
treatments. The weed control treatment significantly affected first year height.  Hansen states that from
the standpoint of tree survival and growth ,the pre-emergent herbicide lenuron applied alone or combined
with other treatments gave consistently superior performance.

Glyphosate was found to be extremely difficult to apply after planting without damaging tree seedlings.
Actively growing young hybrid poplars are easily damaged by even small amounts of glyphosate spray
but are not affected through the soil (Hansen). Glyphosate damage is manifested in off color leaves and
stunted growth.

Other researchers in Canada, Sweden, Italy, and the United Kingdom seem to agree that herbicides are
consistently the most effective and cheapest means of providing the necessary degree of weed control.  In
contrast, mechanical cultivation must be done every 10-14 days to be effective. Manual weed control does
not appear to be a viable economic option for large scale poplar plantations at this time.  Manual weeding
is  labor intensive and is something to be avoided if possible even in small scale operations.

The actual choice of herbicide and application method chosen appears to depend chiefly on the nature of
the weed problem and the timing of the application. Keep in mind that dry weather may render pre-
emergent herbicides ineffective. A cautionary note is that laws regulating the use of herbicides differ
from country to country. In America, regulations require the listing of a crop species on the herbicide
label before it can be used legally on a commercial or private basis (Handbook of Short Rotation Woody
Crops). Herbicide labels are constantly changing and one should also consult specific product labels  and
information before applying any herbicide. On smaller scale for plantings near wetlands or other
sensitive areas, the use of plastic microfunnel mulches may be  another option to consider.  Ultimately, the
level of weed control required will depend on the area to be planted, the time  of year, and whether weeds
are primarily annuals or perennials.  A more in-depth review of weed management in short rotation
woody crops is provided in a 1998 paper, "WeedManagement in Short Rotation Poplar and Herbaceous
Perennial Crops Grown for Biofuel Products" by Douglas Buhler.
                                              B-17

-------
Irrigation

The decision whether to irrigate or not can often be difficult.  One must consider such factors as the depth
to ground water, the amount of annual precipitation and the timing of this precipitation. Some places like
Ft. Worth, Texas receive most of their precipitation in the spring and fall. Places with only sporadic,
scattered rain in the summer can make the establishment of cuttings difficult because  they lack an
adequate root system. An understanding of historic weather patterns is required to make an informed
decision on whether to install an irrigation system in a given area.  Fortunately, free world-wide historical
climate data can be obtained on-line from the Utah Climate Center at Utah State University at
http://climate.usu.edu/free.

Supplemental water should be applied if soil moisture falls below  75 to 80 per cent of field capacity of
below -0.05 to -0.1M Pa (0.5 to - 1.0 far) of tension (Dickman ).  Another approach is to irrigate
whenever weekly precipitation fails to reach a certain minimum amount (Dickman).  Tensiometers
installed at a depth of 18 and 60 inches are a good way to assess the amount of available soil. There are
numerous ways to apply supplemental water. Flood irrigation is the most economical but is restricted to
level terrain and soil with high water holding capacity.

Large scale short rotation woody crop plantations in the Pacific Northwest employ drip irrigation systems
that deliver millions of gallons of water per day derived from the Columbia River.  Drip irrigation allows
application of precise amounts of water to plant roots (New). This allows soil moisture in the area around
the plant to be maintained at a uniform level throughout the growing period (New). Drip irrigation is
used more often for orchard crops than for field crops (New).  Drip irrigation was employed at the
Carswell site during the first growing season. Without this irrigation system, the plantations at Carswell
would have failed because the summer of 1996 was one of the driest summers on record in  Texas.

Many planted trees are able to reach groundwater 3m below the surface when irrigated for the first two
seasons after having been planted (Briggs).  This was also our experience at the Carswell site.  A root
study conducted by the University of Georgia found that both plantations at the Carswell site had reached
the saturated zone in September of 1997, seventeen months after planting (Hendrick).  There are
numerous ways to install an irrigation system at a site. Tree roots usually only explore moist soil so when
the irrigation system is turned off roots can often be left high and dry above the water table or saturated
zone. First plantings should be irrigated the first growing season.  The length of irrigation and the amount
depend on how long it takes tree roots to reach the saturated zone.  Typically, young growing
cottonwoods require 5-8 gallons a day per tree. (19-30 liters/day/tree) Experience in the restoration of
riparian vegetation in the arid western United States has shown that the most reasonable irrigation
strategy to give trees an over abundance of water so that soil is saturated to groundwater nearly constantly
(Briggs).

 The typical components of a drip irrigation system are a main pipeline which carries water to manifolds
and lateral lines. Water flow is regulated using manual or automatic valves.  Guidance on how to plan
and operate an orchard drip irrigation system can be obtained in the booklet Planning and Operating
Orchard Drip Irrigation Systems B-l 663 from the Texas Agricultural Extension Service at  Texas A&M
University System in College Station, Texas. This booklet addressees drip irrigation system layout,
salinity management, emitter clogging control, fertilizer injection, and backflow prevention.

Salinity management is important because water from streams and aquifers usually contain  dissolved
salts.  Application of groundwater can add salt to the soil where it will accumulate unless it is moved
below the root zone by rainfall or excess irrigation water (New). When the amount of salt added exceeds
the amount removed by leaching salts, the concentration in the soil can become harmful to trees and other
plants (New). This process, called salinization, has caused the collapse of agriculture in many ancient and
modern societies (Hillel). Irrigation water is considered poor quality when it contains moderate to large
amounts of salt.  Before irrigating a phytoremediation plantation with water from a contaminated deep
aquifer it is important to know the amount of salts in this water (New). It is important not to guess about
soil and water quality.  It is advisable to have an annual salinity analysis of soil  samples from the root

                                              B-18

-------
zone to insure the long term productivity of a phytoremediation plantation irrigated with deep
contaminated water.

Emitters employed in drip irrigation frequently clog from physical, biological, and chemical processes.
Clogging reduces water emission rates and can cause stress to plants by non-uniform water distribution
(New). Physical clogging is caused by soil, sand, pipe scale, and plant material and can be prevented by
employing a filter system that is appropriate for the emitter type and size (New).  Filters with multi-stage
corrosion-resistant screens may be required when irrigation water contains large amounts of sand.
Biological clogging is usually in the lateral lines and is caused by microorganisms and algae. Biological
clogging is reduced by selecting emitters with large orifices and flushing the system with a chlorine
concentration between 10-50 ppm (New).  High concentrations and the precipitation of calcium,
magnesium, and iron in irrigation water causes chemical clogging (New).  Concentrations of calcium
and magnesium greater than 50 ppm in irrigation water often requires periodic injections of hydrochloride
solution throughout the growing season (New).

Back flow occurs when the flow of water is reversed from an irrigation system back into a potable water
supply system. If contaminants are allowed to flow back into the potable water system it is possible to
create a public health problem.  The prevention of backflow in irrigation is very important.  It is
important to have an understanding of how to prevent backflow.  Any connection between a potable water
supply and a potential source of contamination is termed a cross-connection. Backflow or the reverse flow
of liquids in a plumbing system is caused by two basic conditions backpressure or backsiphonage. The
most likely causes of backpressure; are a booster pump designed without backflow prevention devices or
interconnection with another system operated at a high pressure such as a fertigation injector system.
When a change of system pressure causes the pressure at the supply point to become lower than the
pressure at the  point of use non-potable water can be backsiphoned into the main line.   The main causes
of backsiphonage are undersized piping, line  repairs or breaks that are lower than a service point,  lower
main pressure from high water withdrawal rates and reduced supply main pressure on the suction  side of a
booster pump.  Pollutants can be controlled at the cross-connection by one of several mechanical
backflow preventers such as atmospheric or pressurized vacuum breakers, double check-valve assemblies,
and a reduced pressure principle assembly. The type of backflow preventer required is based on the risks
posed by the substance which may flow into the potable water supply system.  Local and state
regulations for codified construction  requirements need to be checked.  All backflow preventers should
be inspected after installation and checked annually to insure their proper function and operation.

MONITORING LESSONS LEARNED

The monitoring of groundwater at the Carswell Site has produced several insights. The first is that
traditional groundwater level measuring devices will likely cease to operate properly or give erroneous
readings due to roots from the planted cuttings hanging them up. The iron in the steel float can interact
with the groundwater to produce greatly elevated hydrogen levels.  This is an artifact and doesn't reflect
the influence of the plantation subsurface biomass on the geochemistry of the groundwater. The problems
with traditional floats were resolved at the Carswell Site by employing Design Analysis WATERLOG
H310 pressure  sensors.  These cost approximately $1000 a piece and work by detecting changes in flow
which correlate to changes in pressure. It is important that this pressure sensor be clamped or tied down to
fixed location where there is no velocity flow.  If the pressure is subject to open flow it is likely that the
readings will be inconsistent (Rivers). This no flow condition is achieved by suspending the sensor from
a stainless steel drop cable and using a weighted ballast or sinker (Rivers).
                                              B-19

-------
                                           Where Can I Order Hybrid Poplar Cuttings?
Lee Wholesale Nursery
Fertile, MN 56540
(218) 574-2237
Lincoln-Oaks Nurseries
Box 1601
Bismark,ND 58501
Schumacher's Nursery & Berry Farm
711 Chapman Avenue
Route 2 Box 10
Heron Lake, MN 56137
(507) 793-2288
Mike Hradel
Cold Stream Farm
2030 Free Soil Road
Free Soil, MI 49411
(616) 464-5809
Jamie DeRosier
Route 1 Box310A
Red Lake Falls, MN 56750
(218)253-2861
Insti Trees Nursery
Box 1370
Rhinelander, WI 54501
(715)365-8733
Hramor Nursery
515 9th Street
Manistee, MI 49660
(616)723-4846
Pope SWCD
24 First Avenue SE
Glenwood, MN 56334
(320) 634-5326
East Otter Tail SWCD
655 3ri Avenue Southeast
Perham, MN 56573
(218)346-2050
MN Agro-Forestry Coop
c/o WesMin RC&D Council
900 Robert Street, #104
Alexandria, MN 56308
(320) 763-4733
Mt Jefferson Farms, Inc
P.O. Box 12708
Salem, OR 97309
(503)363-0467
Segal Ranches
2342 S. Euclid Road
Grandview, WA 98930
(509)882-2146
                                                           B-20

-------
                      WHERE TO GET EASTERN COTTONWOOD CUTTINGS




                                 Eastern Cottonwood (P. deltoides)




                                  Non-Proprietary Planting Stock
             • 110804




             • 110610




             • 110412




             • 110226




             • ST75




             • ST72




             • ST70




             • ST66




             • S7C20




             • S7C15




             • S7C8




             • S7C1




NOTE:   ST clones were developed by Stoneville Lab




         S7C clones originated in Texas




         110 clones originated from various sandbars along the Mississippi River
CROWN VANTAGE




FOREST RESOURCES




5925 NORTH WASHINGTON STREET




VICKSBURG, MS 39183




OFFICE: (601) 630-9899




FAX: (601) 636-5865
                               Non-Proprietary Cottonwood Cuttings




                                        Harrison Wells




                                      Ripley County Farms




                                         P.O. Box 614




                                      Doniphan, MO 63935




                                        (573) 996-3449




                                         rcf@semo.net




                                             B-21

-------
                                           Forest Service Offices
Region 1 - Northern
Region 6 - Pacific Northwest   Northeastern Area
USDA Forest Service




State & Private Forestry




Forest Pest Managaement




Federal Building




P.O. Box 7669




Missoula, MT 59807




(406)329-3511




FTS 585-3511









Region 2 - Rocky Mountain









USDA Forest Service State & Private




Forestry




Forest Pest Management




11177W. 8th Ave.




Box 25127




Lakewood, CO 80225




(303) 236-3213




FTS 776-3213









Region 3 - Southwestern









USDA Forest Service




State & Private Forestry
USDA Forest Service




State & Private Forestry




Forest Pest Management




319 S.W. PineSt.




P.O. Box 3623




Portland, OR 97208




(503) 221-2877




FTS 423-2727









Region 8 - Southern









USDA Forest Service




State & Private Forestry




Forest Pest Management




1720 Peachtree Road N.W.




Atlanta, GA 30367




(404) 347-2989




FTS 257-2989









USDA Forest Service




State & Private Forestry




Forest Pest Management




2500 Shreveport Hwy.




Pineville, LA 71360
USDA Forest Service




State & Private Forestry




Forest Pest Management




370 Reed Road




Broomall, PA 19008




(215)461-3252




FTS 489-3252









USDA Forest Service





State & Private Forestry




Forest Pest Management




Louis C. Wyman For. Sci. Lab.




P.O. Box 640




Durham, NH 03842




(603) 868-5719




FTS 834-5765









USDA Forest Service




State & Private Forestry




Forest Pest Management






180 Canfield St.




P.O. Box 4360




Morgantown, WV 26505




(304)291-4133
                                                    B-22

-------
Forest Pest Management




Federal Building




517GoldAve. S.W.




Albuquerque, NM 87102




(505) 842-3292




FTS 476-3292




Region 4 - Intermountain









USDA Forest Service




State & Private Forestry




Forest Pest Management




Federal Building




324 25th St.




Ogden,UT 84401




(801) 625-5257




FTS 586-5257




Region 5 - Pacific Southwest
(318)473-7160




FTS 497-7160









USDA Forest Service




State & Private Forestry




Forest Pest Management




200 Weaver Blvd.




Asheville, NC 28804




(704) 672-0625




FTS 672-0625
FTS 923-4133









USDA Forest Service




State & Private Forestry




Forest Pest Management




1992 Folwell Ave.




St. Paul, MN 55108




(612)649-5261




FTS 777-5261
Region 10 -Alaska
USDA Forest Service




State & Private Forestry




Forest Pest Management




630 Sansome St.




San Francisco, CA 94111




(415) 556-6520




FTS 556-6520
USDA Forest Service




State & Private Forestry




Forest Pest Management




Federal Office Building




Box 1628




Juneau, AK 99802




(907) 261-2575




FTS 907-261-2575
                                                    B-23

-------
                        REGIONAL BIOMASS ENERGY PROGRAM









The Regional Biomass Energy Program (RBEP) carries out activities related to technology transfer,




industry support, resource assessment, and matches local resource to conversion technologies. Activities




are conducted by five regional programs (Northwest, Western, Great Lakes, Southeast and Northeast) that




promote development of biomass energy conversion technologies and feedstocks that are applicable to the




region.









                                      Michael Voorhies




                                  U.S. Department of Energy




                               Regional Biomass Energy Program




                             1000 Independence Avenue S.W. EE-31




                                 Washington, DC 20585-0001




                          (202) 586-1480 (phone), 202-586-1605 (fax)




                                 michael.voorhies@hq.doe.gov
Fred J. Kuzel




Great Lakes Regional Energy Program




35 E. Wacker Drive, #1850




Chicago, IL 60601




(312) 407-0177(phone), (312) 407-0038 (fax)




fkuzel@cglg.org




(Illinois, Indiana, Iowa, Michigan, Minnesota,




Ohio, and Wisconsin)
Jeff Graef




Dave Waltzman




P.O. Box 95085




Lincoln, NE 68509-5085




Graef: (402) 471-3218, fax (402) 471-3064




Jgraefgimail. state .ne .us




Waltzman: (303) 275-4821, fax (303) 275-4830




Dave. waltzman(g),hq.doe. gov




(Arizona, California, Colorado, Kansas, Nebraska,




Nevada, New Mexico, North Dakota, Oklahoma,




south Dakota, Texas, Utah, and Wyoming)
                                            B-24

-------
Richard Handley




Northeast Regional biomass Program




Coalition of Northeastern Governors




400 North Capital St., NW




Suite 382




Washington, D.C., 20001




(202) 624-8454 (phone), (202) 624-8463 (fax)




nrbp@sso.org




(Connecticut, Delaware, Maine, Maryland,




Massachusetts, New Hampshire, New Jersey, New




York, Pennsylvania, Rhode Island, and Vermont)
Jeff James




Northwest Regional Biomass Energy Program




800 5th Ave, Suite 3950




Seattle, WA 98104




(206) 553-2079 (phone), (206) 553-2200 (fax)




jeffrev.james@hq.doe.gov




(Alaska, Idaho, Oregon, Montana, and Washington)
Phillip Badger




Southeast Regional Biomass Energy Program




P.O. Box 26




Florence, AL 35631




(256) 740-5634 (phone), (256) 740-5530 (fax)




pcbadger(gimindspring.com




(Alabama, Arkansas, Florida, Georgia, Kentucky,




Louisiana, Mississippi, Missouri, North Carolina,




South Carolina, Tennessee, Virginia, West




Virginia, Washington, DC)
More RBEP information and reports are available




at the Biomass Resource Information




Clearinghouse.
                                             B-25

-------
                        REFERENCES for APPENDIX B
1.      Briggs, Mark K., Riparian Ecosystem Recovery in Arid Lands. The University of
       Arizona Press, Tucson 1996.

2.      Christersson, L., R Ramstedt, M., and Forsberg Pests, Diseases, and Injuries in Intensive
       Short Rotation Forestry. Chapter 7 in Ecophysiology of Short Rotation Forest Crops
       edited by C.P. Mitchell, J.B. Ford-Robertson, T. Hinckley, and L. Sennerby-Forsse,
       Elsevier Science Publishers, London, 1992 pp!85-212.

3.      Dickman, Donald and Stuart, Katherine W., The Culture of Poplars in Eastern North
       America, Michigan State University Publications, 1983.

4.      Dickman, D. I. , and Pregitzer, K.S., The Structure and Dynamics of Woody Plant
       Systems. Chapter 4 in Ecophysiology of Short Rotation Forest Crops edited by C.P.
       Mitchell,J.B. Ford-Robertson, T. Hinckely and L.Sennerby-Forsse, Elsevier Science
       Publishers, London, 1992 pp95-115.

5.      Hansen E.A., Ostry M.E., Johnson W.D., Tolsted, D.N., Netzer, D.A.,Berguson W.E.,
       and Hall, R.B. Field Performance of Poplulus in Short Rotation Intensive Culture
       Plantations in the North-Central U.S., United States Department of Agriculture Forest
       Service, North-Central Experimental Station Research Paper NC-320.

6.      Hansen E., Heilman, P., and Strobel, S., Clonal Testing and Selection for Field
       Plantations, Chapter 5 in Ecophysiology of Short Rotation Forest Crops edited by C.P.
       Mitchell, J.B. Ford-Robertson, T. Hinckley, and L. Sennerby-Forsse, Elsevier Science
       Publishers, London, 1992 pp!24-145.

7.      Hansen E.A., Netzer D..A.,  and Tolsted D.N. Guidance for Establishing Poplar
       Plantations in the North-Central U.S., United States Department of Agriculture Forest
       Service, North Central Forest Experiment Station Research Note NC-363, 1993.

8.      Hansen E. A., A Guide for Determining When to Fertilize Hybrid Poplar Plantations,
       United States Department of Agriculture Forest Service North-Central  Forest
       Experiment Station, Research Paper NC-319, 1994.

9.      Harrington, C.A., and DeBell D.S., Above and Below Ground Characteristics Associated
       with Wind Toppling in a Young Populus Plantation, Trees- Structure and Function 11
       (2): 109-118.

10.    Heilman P.E., Hinckley T.M., Roberts D.A., and Ceuleman R., Production Physiology
       Chapter 18 in Biology of Populus and its Implications for Management and Conservtion,
       edited by R.F. Stettler, H.D. Bradshaw Jr., P.E. Heilman, and T.M. Hincley Nation
       Research Council of Canada NRC Research Press Ottawa  1996 pp 459-489.

11.    Kopp,R.F., Abrahamson, L.P., White E.H., Volk,T.A., Willow Biomass Producer's
       Handbook , State University of New York College of Environmental Science and
       Forestry, Syacuse, NY.

                                          B-26

-------
12.    Neuman, D.S. Wagner, M., Braatne, J.H. and Howe, J., Stress physiology-abiotic.
      Chapter 17 Biology of Populus and its Implications for Management and Conservation.
      Edited by R.F. Stettler, H.D. Bradshaw Jr., P.E. Heilman, T.M. Hinckley. National
      Research Council of Canada, Ottawa, National Research Press, pp 423-458.

13.    New, Leon and Fipps Guy, Planning and Operating Orchard Drip Irrigation Systems B-
      1663.  Texas Agricultural Extension Service The Texas A&M University System,
      College Station, TX 1992.

14.    Ostry, M.E., Wilson, L.F. McNabb, H.S. Moore,L.M., A Guide to Insect Disease , and
      Animal Pests of Poplars, United States Department of Agriculture Forest Service ,
      Agriculture Handbook 677, 1989.

15.    Portwood, Jeff, Crown  Vantage Corporation , personal communication 22 March, 00 .

16.    Pregitzer, K.S. The Structure and Function of Populus Root Systems, Chapter 14 in
      Biology of Populus and its Implications for Management and Conservation edited by R.F.
      Stettler, H.D. Bradshaw Jr., P.E. Heilman, and T,M. Hinckley, National  Research
      Council of Canada, Ottawa National Research Press, 1996.

17.    Telewski F.W. Wind -induced physiological and developmental responses in trees,
      chapter 14 in Wind and Trees edited by M.P.Coutts and J.Grace, Cambridge University
      Press 1995, pp237-259.

18.    Tolbert, Viginia ,  Oak Ridge National Laboratory Biomass Biofuel Program , personal
      communication 3 February, 00.

19.    Vogel, S., Blowing in the Wind: Storm -Resisting Feature of the Design of Trees in
      Storms, Journal of Arboriculture 22(2) March 1996, pp92-98.

20.    Volk, Timothy, State University of New York College of Environmental Science and
      Forestry, personal communication  16 March 00.

21.    Whitlow T.H. and Harris R.W. Flood Tolerance in Plants: A State of the Art Review
      Technical Report E-79-2. U.S. Army Engineer Waterways Experiment Station  1979.

22.    Wood C.J. Understanding wind forces in trees, chapter 7 in Wind and Trees edited by
      M.P. Coutts and J.Grace Cambridge University Press, 1995 pp 133-163.

23.    Ying C.C. and Bagley, W.T., Genetic Variation of Eastern Cottowood in an Eastern
      Nebraska Provence Study
                                         B-27

-------