&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 )
^^
te for treatmel
js waste man
Hazardous wastes transported off-sil
must be accompanied by a hazardoL
.«
"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
------- |