>S-EPA
United States
Environmental Protection
Agency
Solid Waste and
Emergency Response
Washington, DC 20460
(5102G)
EPA/542/R-00/007
July 2000
www.epa.gov
www.clu-in.org
Proceedings of the Ground-Water/
Surf ace-Water Interactions
Workshop
-------
-------
CONTENTS
EXECUTIVE SUMMARY
PRESENTATION ABSTRACTS
A Federal Statutory/Regulatory/Policy Perspective on Remedial Decision-making with Respect
to Ground-Water/Surface-Water Interaction
Guy Tomassoni 13
Interaction of Ground Water and Surface Water
Thomas C. Winter 15
Hydrogeology and Biogeochemistry of the Surface Water and Ground Water Interface of a
Mountain Stream
Cliff Dahm 21
Ground-water Plume Behavior Near The Ground-Water/Surface Water Interface of a River
Brewster Conant, Jr 23
Assessment Approaches and Issues in Ecological Characterizations
G. Allen Burton, Jr. and Marc S. Greenberg .31
Delineation, Quantification, and Mitigation of Discharging Plumes
David R. Lee 35
Field Technology and Ecological Characterization of the Hyporheic Zone
D. Dudley Williams 39
DISCUSSION GROUP SUMMARIES
Hydrogeology Discussion Group Summary
Thomas C. Winter and Joseph Dlugosz 46
Chemistry Discussion Group Summary
Allen Burton and Ned Black .... 54
Biological Discussion Group Summary
Cliff Dahm and Bruce Duncan 58
-------
POSTER SESSION ABSTRACTS
Use of Multi-Parameter Sensitivity Analysis to Determine Relative Importance of Processes
Involved in Transport of Mining Contaminants
Jungyill Choi, Judson W. Harvey, and Martha H. Conklin 69
Measurements of Plant and Algal Bioaccumulation of Metals in Final and Pinto Creeks, Arizona
Justin C. Marble, Timothy L. Corley, and Martha H. Conklin 73
Tracing Groundwater Flow into Surface Waters by Application of Natural and Artificial Tracers
D. Reide Corbett, William Burnett, Jeffrey Chanton, and Kevin Dillon 77
Considerations for Calculating the Mass Loading of Metal Contaminants to a Marine
Embayment: ASARCO Superfund Site, Tacoma, WA
Gayle Garman and ASARCO Sediments/Groundwater Task Force 81
The Interaction of Ground Water and Surface Water within Fall Chinook Salmon Spawning
Areas in the Hanford Reach of the Columbia River
David R. Geist 95
Integrated Acoustic Mapping of Surface Waters: Implications for Ground-Water/Surface-Water
Linkages
Chad P. Gubala, Ullrich Krull, Joseph M. Eilers, Mike Montoya, and Jeff Condiotty 99
Delineation of VOC-Contaminated Groundwater Discharge Zone, St. Joseph River, Elkhart,
Indiana
John H. Guswa, Jonathan R. Bridge, and Michael J. Jordan 100
Measuring Enhanced Removal of Dissolved Contaminants in Hyporheic Zones and
Characterizing Causes and Consequences for Water Quality
Judson W. Harvey, Christopher C. Fuller, and Martha H. Conklin 103
Bioassessment of Hyporheic Microbial Communities Using a Specially-designed Sediment
Colonization Chamber
Susan P. Hendricks 107
Fundamentals of SPMD Sampling, Performance, and Comparability to Biomonitoring Organisms
J.N. Huckins, J.D. Petty, H.F. Prest, J.A. Lebo, C.E. Orazio, J. Eidelberg, W.L. Cranor,
R.W. Gale, andR.C. Clark .' 113
Acid Mine Drainage—The Role of Science
Briant Kimball 118
-------
Temporal and Spatial Trends in Biogeochemical Conditions at a Gxoundwater-Surfacewater
Interface
John M. Lendvay and Peter Adriaens 120
Natural Attenuation of Chlorinated Solvents in a Freshwater Tidal Wetland, Aberdeen Proving
Ground, Maryland
Michelle M. Lorah and Lisa D. Olsen 126
Discharge of Contaminated Ground Water to Surface Water: An Ecological Risk Assessment
Perspective
Mary Baker Matta and Tom Dillon 131
Defining Groundwater Outcrops in West Neck Bay, Shelter Island, New York Using Direct
Contact Resistivity Measurements and Transient Underflow Measurements
Ronald Paulsen 138
Influence of Stream Orientation on Contaminated Ground-Water Discharge
Don A. Vroblesky 143
Factors Controlling Hyporheic Exchange in a Southern Ontario Stream: Modeling Riffle-Scale
Patterns in Three Dimensions Using MODFLOW
R.G. Storey, D.D. Williams, and K.W.F. Howard 148
Solute and Solid Phase Relationships in the Surface Hyporheic Zone of a Metal Contaminated
Stream, Silver Bow Creek, MT
Johnnie N. Moore and William W. Woessner 151
APPENDICES
Appendix A: Workshop Participants List 157
Appendix B: Discussion Group Focus Issues 162
Appendix C: Case Study Summaries 165
Appendix D: MHE Push Point Sampling Tools 191
m
-------
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
EXECUTIVE SUMMARY
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
INTRODUCTION
Although ground water and surface water are usually evaluated as separate water masses, they are
connected by the ground-water/surface-water transition zone1 in a hydrologic continuum.
Understanding contaminant fate and transport in this zone is important to the U.S. Environmental
Protection Agency's (EPA's) hazardous waste site cleanup programs across the nation because about
75% of RCRA and Superfund sites are located within a half mile of a surface water body, and almost
half of all Superfund sites have impacted surface water. Investigations of ground water and surface
water need to be integrated and incorporate recent advances in investigative techniques.
Ecological risk assessments for surface water bodies have all too often focused on the water
column (where the ground-water contaminant plumes become extremely diluted), or on the sediments.
Typically there has been little or no evaluation of contaminated ground-water discharges. Impacts from
the discharge of contaminated ground water on the transition zone ecosystem have been ignored, even
though this ecosystem provides important ecological services and is the most exposed to ground-water
contaminants. Based on these considerations, the need to evaluate the transition zone is clear.
To address the technical concerns related to ecological impacts in the transition zone, the EPA's
Office of Solid Waste and Emergency Response (OSWER) sponsored a workshop in January 1999,
which was planned jointly by the Ecological Risk Assessment Forum and the Ground Water Forum.2
The workshop was organized around answering two fundamental questions:
• How important is the transition zone ecologically?
• How can we measure hydrogeological, chemical, and biological conditions and changes in this
zone?
There was a consensus among workshop participants that protecting this zone is important, and
that there is a need for studies by interdisciplinary teams to ensure that valid data are obtained from the
correct locations and at the right times so that valid conclusions are reached. Both forums plan to use
the workshop information to submit research recommendations to EPA's Office of Research and
Development, develop a list of suggested tools for investigating hydrogeological fate and transport and
ecological effects at contaminated sites, develop Agency guidance, and conduct a pilot study using this
methodology. The workshop and these proceedings provide a first step to understanding the
fundamentals of evaluating the effects of contaminated ground water discharging through the transition
zone.
WORKSHOP GOALS
The overall goal of the workshop was to provide an opportunity for individuals from various
scientific and technical backgrounds to discuss the importance of the ground-water/surface-water
transition zone and help regulators better understand environmental issues relating to the connections
1 In these proceedings, the authors may use terms other than "ground-water/surface-water transition zone" to
indicate this zone of transition. These terms may be equivalent (e.g., ground-water/surface water interface) or more
restrictive (e.g., hyporheic zone, which refers to the interface between ground water and lotic (moving) surface
waters.)
2 The Ecological Risk Assessment Forum and Ground Water Forum comprise ecological risk assessment and
ground-water specialists, respectively, from EPA's Regional Offices, Headquarters, and Office of Research and
Development. These forums help the EPA maintain consistency and develop national program guidance.
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
between ground water and surface water. Within this broad goal, the Ecological Risk Assessment
Forum and Ground Water Forum had the following additional specific goals:
Ecological Risk Assessment Forum Goals:
• Develop a conceptual model for use in ecological risk assessment at sites where contaminated
ground water discharges to surface water.
• Integrate structural, functional, and hydrogeological components and methods for evaluating
changes to the ecosystem.
Ground Water Forum Goals:
• Increase awareness of new tools used to evaluate fate and transport within the transition zone.
• Identify and understand geological, hydrological, and chemical factors that might influence
transition zone dynamics.
WORKSHOP DESIGN
A planning committee from the two forums designed the workshop to promote multidisciplinary
interaction on a set of focus issues and questions. The workshop included invited platform speakers, a
poster session, discussion groups, and an overall report-out from the groups and subsequent discussion.
This approach worked well, resulting in fairly uniform agreement on concepts and recommendations
regarding integration and use of investigatory tools.
Multidisciplinary Approach
Invited workshop participants included ecologists, geochemists, and hydrogeologists who work
with the ground-water/surface-water transition zone (Appendix A).
Conceptual Model
A draft illustration of the conceptual model representing the forums' current understanding of
ground-water/surface-water interactions for a river was presented and explained at the beginning of the
workshop. The participants were asked to review the conceptual model and improve it as greater
understanding was gained during the course of the workshop. Workshop participants also identified but
did not address the need for research into other transition zone environments, such as those for lakes,
estuaries, and wetlands.
Platform Speakers
The planning committee invited seven platform speakers to present topics representing a cross-
section of information on ground-water/surface-water interactions; the presentations helped workshop
participants address focus issues and questions in subsequent discussion groups. The abstracts of the
speakers' presentations are included in this report:
• A Federal Statutory/Regulatory/Policy Perspective on Remedial Decision-making with Respect to
Ground-Water/Surface Water Interaction (Guy Tomassoni, EPA's Office of Solid Waste)
• Interaction of Ground Water and Surface Water (Tom Winter, U.S. Geological Survey)
• Hydrogeology and Biogeochemistry of the Surface Water and Ground Water Interface of a
Mountain Stream (Cliff Dahm, University of New Mexico)
-------
Proceedings of the Ground- Water/Surface- Water Interactions Workshop
July 2000
• Ground-Water Plume Behavior Near the Ground-Water/Surface-Water Interface of a River
(Brewster Conant, University of Waterloo)
• Assessment Approaches and Issues in Ecological Characterizations (Allen Burton, Wright State
University),
• Delineation, Quantification, and Mitigation of Discharging Plumes (David Lee , AECL Chalk
River, Ontario), and
• Field Technology and Ecological Characterization of the Hyporheic Zone (Dudley Williams,
University of Toronto)
Poster Session
A poster session during the workshop allowed related papers to be presented outside of the formal
discussion agenda. Abstracts of the posters are included in this report.
Discussion Groups
The topics of the three discussion groups were hydrogeology, chemistry, and biology as they relate
to ground-water/surface-water interactions. Three of the platform speakers, Tom Winter, Allen Burton,
and Cliff Dahm, and three members of EPA, Joseph Dlugosz, Ned Black, and Bruce Duncan, served as
discussion group co-chairs to guide discussions along the focus issues listed in Appendix B. To focus
the discussions further, participants were asked to consider first the scenario of ground water
discharging to a river.
Each workshop participant was assigned to two of the three discussion groups, and each group was
organized with a balance of hydrogeologists, geochemists, ecologists, and microbiologists to encourage
dialogue among people with different academic backgrounds. When the groups rotated for the
afternoon session, the co-chairs remained to provide continuity and briefly explain what the morning
session had covered. Some of the focus group issues were not fully addressed due to lack of
information or time, however. Discussion group summaries are included in this report.
Report Out and Overall Discussion
The information from the three discussion groups was summarized by the co-chairs and presented
to all of the participants at the close of the workshop. This in turn led to a general group discussion of
topics and future needs for research.
WORKSHOP RESULTS
The workshop brought together representatives from a variety of technical disciplines to focus on
the ground-water/surface-water transition zone. Chemists, microbiologists, hydrogeologists, and
ecologists from EPA, the U.S. Geological Survey (USGS), the National Oceanic and Atmospheric
Administration (NOAA), state environmental agencies, other government agencies, academia, and
industry discussed the hydrological, chemical, and biological processes that occur in this transition
zone and how to measure and interpret changes in these processes. Discussions highlighted the need to
revise the existing conceptual model for ecological risk assessment to evaluate the important structural
and functional aspects of the transition zone. Information was provided about many tools used to
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
evaluate the hydrological, chemical, and ecological aspects of this zone and the spatial and temporal
scales at which measurements are needed.
The following is a summary of the key points drawn from the presentations by platform speakers,
discussion group dialogues, and revisions to the conceptual model.
Platform Speakers
While providing a common multidisciplinary focus on transition zones, the speakers emphasized
the following facets of transition zone hydrogeology, chemistry, and ecology:
• Physiography and climate affect the interaction of ground water and surface water across diverse
landscapes. For example, movement of water through the transition zone is influenced by the
position of surface water bodies within ground-water flow systems, small-scale geologic features
beneath surface water, climate, and hyporheic exchange (the exchange of moving surface water
with ground water). These seemingly diverse systems may be studied, analyzed, and managed
under a unifying framework based on "hydrologic landscapes." Transition zones are particularly
important ecologically because they store and retain nutrients (and potentially contaminants),
transform compounds biologically and chemically, provide refuge to benthic invertebrates, and are
a base of the aquatic food web. Virtually no research has been conducted on the effects of
contaminants on hyporheic communities. Research should evaluate indigenous microbial activity,
organic matter/nutrient cycling, invertebrate community indices, tissue residues of dominant
species, in situ toxicity, and in situ physicochemical profiles. Very site-specific research could
include novel tools such as ecological food web modeling, semi-permeable membrane devices to
evaluate bioaccumulation, toxicity identification evaluations to determine the classes of chemicals
(e.g., metals or organic compounds) responsible for observed toxicity, and identification and
evaluation of in situ stressors including physical stressors (e.g., flow or suspended solids). It also
will be critical to establish appropriate uncontaminated reference sites for comparison with
contaminated sites.
• The hydrogeology of the ground-water/surface-water transition zone strongly influences the spatial
and temporal distribution of both aerobic and anaerobic microbial processes as well as the
chemical form and concentration of nutrients, trace metals, and contaminants in surface and
ground waters. Major hydrologic events such as spring snowmelt affect biochemical components.
Studies that integrate hydrogeology, biogeochemistry, and aquatic ecology are needed to
understand fully the dynamics and importance of the transition zone.
• Determining the location and magnitude of contaminant discharges to surface waters from ground-
water plumes is a complex hydrogeological and biogeochemical problem. Although measurements
of hydraulic gradient may be sufficient to delineate large discharge areas, numerous seepage
studies have shown that areas of significant discharge can be small and easily missed. Even in
relatively homogeneous terrain, flows may be highly focused at shorelines, and solute transport
may be rapid. Geochemical conditions and contaminant concentrations may change drastically
over intervals of a few centimeters. Closely spaced measurements can be used to determine
contaminant concentrations in and flux from the streambed and to distinguish areas of high
attenuation from areas of poor attenuation (e.g., sand stringers, interconnected zones of higher
permeability, or other preferential flow paths).
• Physical and numerical model studies, like seepage studies, indicate surface-water head differences
of a few centimeters between riffles, and pools in streams can produce surface-water exchange
-------
Proceeding of the Ground- Water/Surface- Water Interactions Workshop
July 2000
flows within permeable alluvial sediments despite net discharge of ground water to the stream.
Modeling can be used to reveal interactions between surface water and ground water that are
overlooked by larger scale models but have important chemical and biological consequences for
the ground-water systems, the stream, and the biota.
Discussion Group Summaries
Each discussion group agreed on the importance of the ground-water/surfacerwater transition zone
and emphasized the need for multidisciplinary approaches to evaluating fate, transport, and effects of
contaminants in this zone. The main differences among the groups were in discussion of the tools used
by each discipline.
Hydrogeology
The hydrogeology discussion group focused on using a tiered approach to determine the movement
of ground water to surface water. The group recommended starting with a general reconnaissance of
observable indicators of ground-water discharge and evolve to very detailed and focused sampling of
hydraulics, chemistry, and biology. They recommended the following tiers:
• Use field methods that indicate ground-water discharge to surface water either indirectly (by
observations of qualitative indicators or by chemical data) or directly (by using physical data to
directly measure stage and calculate flow).
• Collect ground-water and surface-water samples over time and during different flow conditions.
• Adjust the field sampling strategy to account for different hydrologic landscapes.
The hydrogeology discussion group also suggested using a generic field design for investigating
the ground-water/surface-water transition zone that includes use of piezometer nests, wells screened
across the water table, and devices to measure or calculate the flow of water and chemicals through the
transition zone. To address the interaction of ground water and surface water, the larger-scale (relative
position of the surface water body within the ground-water flow system) hydrogeologic landscape
processes and the smaller-scale (transition zone) processes should be evaluated.
The group recommended selecting field demonstration sites for research of ground-water/surface-
water interaction in different geographic regimes that account for variation in hydrogeologic
landscapes and climate. The design and effectiveness of site-characterization methods should be tested
and evaluated, and based on the results, the conceptual model and tools for ground-water/surface-water
transition zone characterization should be improved.
Chemistry
The chemistry discussion group emphasized that chemical information is used to evaluate
contaminant chemistry and fate, biological processes, and flow paths. The group recommended the
following:
• Develop initial estimates of actual or potential risks to receptors. Collect information on site
geochemistry and contaminant flow paths—although this might be deferred until after an initial
evaluation.
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
• Develop one or more standard conceptual models to identify important questions to ask and the
data to collect at different types and scales of sites. Sampling efforts in the transition zone may be
more costly than standard sampling of surface water or shallow ground water.
• Determine chemical variations in time and space. In the transition zone, chemical and biological
processes occur over many different time scales such as daily cycles (e.g., temperature and
transpiration), short-term weather events, invertebrate and fish life cycles, seasonal changes, and
long-term climatic changes and events (such as extreme weather events). Characterizing the spatial
extent of contaminant discharge to surface water is just as important as determining the
concentration distribution in a ground-water plume. In a screening or predictive risk assessment,
contaminant concentrations are used for comparisons to toxicity benchmarks. However, the mass
flux or loading of contaminants is also important and influences both the impact of contaminants
on habitats and the physical, chemical, and biological transformations of the contaminants at the
transition zone. The flux of contaminants can change in magnitude and direction with changes in
surface water temperature and stage.
Biology
The biology discussion group concluded that the transition zone is ecologically important. Some
surface organisms have a life stage within this zone, and their productivity could be affected by
contaminants in the zone. Less is known of the unique species that permanently inhabit the transition
zone, and many have not been described. Transition zones often provide high quality habitats and are
sites of contaminant reduction and nutrient and carbon cycling. Transition zones also can provide
preferred habitat, refugia, sites of high biodiversity, habitat for the macrofaunal food base, microbial
production, and energy transfer.
The group agreed that techniques and methods are available to evaluate the structure and function
of the macrobiota and meiofauna. Methods also exist to sample organisms in the transition zone;
however, many of these methods are neither standardized nor well-developed. In particular, there is no
standard method to determine microbial community structure or activity/function. The group made the
following recommendations:
• Use standard metrics, such as community composition, density, and species richness, to compare
sample results regardless of the specific collection method. Evaluate functional feeding groups.
• Conduct bioaccumulation studies and stable isotope analyses to evaluate food chain relationships.
• Understand the basics of community structure and function at all levels before developing more
methods to conduct toxicity testing.
• Coordinate ecologically related sampling in the transition zone with hydrogeological and chemical
surveys at ground-water discharge sites. Use these surveys to help define the biological zones
likely to be affected.
Conceptual Model
To produce the conceptual model shown in Figure 1, the workshop planning committee presented a
draft model at the workshop and revised it from the comments received from participants. This model,
drawn for a river, can be adapted to other sites (lake, tidal, estuaries, marshes, etc.). It combines
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
ecological and hydrogeological concepts to focus on ecological processes in the transition zone and
tools used to investigate fate, transport, and effects of contaminants in discharging ground water.
Figure 1. Conceptual model for contaminated ground water discharging to a river illustrating the need to
look beyond surface water and benthic ecological receptors and hydrogeological fate and transport. Such a
model should consider receptor exposure in the transition zone and account for finer-scale fate, transport
and effects from the discharge of contaminated ground water within this zone.
CONCLUSIONS
General consensus was reached that protecting the transition zone is important, and there is a need
for interdisciplinary studies to understand and document the changes that occur in it. Conclusions
related to the two fundamental organizing questions are discussed below.
How Important is the Transition Zone Ecologically?
The ground-water/surface-water transition zone is an ecological community with important
ecosystem functions affecting several trophic levels from microbes to fish. As an ecotone (i.e., a
transition from the ground-water ecosystem to the surface-water ecosystem), this zone provides key
ecological services to the surface water ecosystem:
• Provides food for benthic macroinvertebrates. The microbial community serves as the food base to
the small organisms within the zone that in turn are food for the benthic macroinvertebrates.
• Provides and maintains unique habitats or refugia, particularly in upwelling zones.
• Cycles nutrients and carbon in aquatic ecosystems.
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
The microbial and biological activity within this zone also may be important for natural
attenuation, because large gradients can be created, which can result in subsurface conditions that
change from anaerobic to aerobic over short distances. Biodegradation can cause organic contaminant
concentrations to change over several orders of magnitude within this zone.
How Can We Measure Hydrogeological, Chemical, and Biological Conditions and Changes in
this Zone?
Despite many unanswered questions (see next section) there are many tools from each of the
disciplines that can be used to evaluate fate, transport, and effects in the transition zone. It was
recognized that the types, locations, Oand times of measurements required to characterize this zone can
vary depending on the questions being asked. Hydrogeologists and ecologists must work together to
obtain information that is useful to both and to efficiently and properly evaluate this zone.
KEY RECOMMENDATIONS FOR RESEARCH
The recommendations presented below were identified during the various phases of the workshop,
particularly within the discussion groups and during the report-out discussions on the final day.
Common Key Areas
The major recommendation common to all three discussion groups is that EPA should create a
series of regional study areas of contaminated transition zone sites. Hydrogeologists, chemists, and
biologists together should determine how, where, and what to sample and how to interpret the results.
These scientists are obligated to integrate their objectives into a single conceptual model to evaluate
transition zones.
Hydrogeology
EPA should encourage research in areas that increase the basic understanding of the influences of
nearby surface-water bodies on contaminant plume migration. Delineation of plumes can be improved
by more widespread application of the hydrologic landscape concepts in site characterization.
Specifically, the following are needed: (1) improved techniques for measuring hydraulic heads, in
stream and on-shore; (2) improved estimation methods of ground-water flow rates near the surface
water boundary; and (3) improved methods for delineating plume concentrations near discharge zones.
Increased use of tracers to help document and quantify the rate of ground-water discharges (or
recharges) is needed. Better gradient quantitation methods are needed, especially in zones of rapidly
fluctuating surface water stage. Also, there is a need for better assessment and evaluation of the
heterogeneity of the ground-water zones adjacent to the surface-water bodies.
Chemistry
EPA should identify a number of regionally representative sites with contaminated transition
zones—along with appropriate uncontaminated reference sites—to be studied by EPA's regional and
Office of Research and Development (ORD) laboratories and academic grantees. The sites should
reflect the scales and contaminant problems typical of each region because the transition zone
chemistry, biology, and hydrology of small mountain streams impacted by mines in Region 8, for
example, may be very different from those of a zone where chlorinated solvent plumes discharge to
one of the Great Lakes in Region 5. The study of ground-water discharge and transition zone flow in
estuaries will be further complicated by tidal fluctuations. Members of the chemistry discussion group
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
felt strongly that extrapolating data from small streams to large rivers and lakes is unacceptable. Also,
some investigations techniques work well in small streams, but not in areas of high flow. As with any
landscape approach, the chemical species and the dominant chemical and physical processes vary for
different landscapes, but some basic processes may be common to some or all of these sites.
Biology
Biological investigations rely heavily on hydrogeological and chemical investigations, particularly
for identifying discharge zones. The regional study sites recommended by the other two groups should
be used to fulfill several biological research needs. The greatest need is for basic biological research,
such as life histories, faunal surveys, and organism activity, so that the full importance of the transition
zone can be determined and changes related to contaminants can be quantified. Sampling and
evaluation tools for both contaminated and uncontaminated substrates need to be developed and
standardized to determine contaminant effects on species richness, trophic structure, and organism
growth for macrobiota, meiofauna, and microorganisms in the transition zone ecosystem. Quantitative
links are needed between site-specific chemical, hydrogeological, and ecological factors and the valued
functions of the transition zone (e.g., contaminant degradation, food base for benthic organisms, role as
a refuge, and high quality habitat).
NEXT STEPS
This workshop was the first step in creating a multidisciplinary foundation for investigating,
monitoring, and evaluating effects in the transition zone from the discharge of contaminated ground
water. Future efforts building on this foundation should take many paths. For example, the conceptual
model of the transition zone presented here is continually evolving. Conceptual models representing
discharges to water bodies other than rivers need to be considered so that approaches and tools
appropriate to wetlands, estuaries, and lakes—including those influenced by tides—can be identified
and developed. Similarly, other pathways need to be identified and addressed, such as contaminated
sediments as sources of contamination to ground water and to the transition zone where infiltration of
surface water occurs.
Based on the workshop, the Ground Water Forum and the Ecological Risk Assessment Forum
intend to:
• Submit research recommendations to ORD.
• Develop a list of suggested tools for investigating hydrogeological fate and transport and
ecological effects at contaminated sites.
• Develop Agency guidance for incorporating the transition zone into risk assessments.
• Conduct a pilot study.
10
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
ACKNOWLEDGMENTS
The Superfund and RCRA Ground Water Forum and Ecological Risk Assessment Forum would
like to thank the following people for their contributions to the Ground-Water/Surface-Water
Interactions Workshop and these proceedings:
OSWER's Technology Innovation Office
EPA's Region 8 office in Denver for providing meeting facilities
Workshop organizers and report editors:
Brace Duncan, U.S. EPA, Region 10, Seattle, Washington
Ren6 Fuentes, U.S. EPA, Region 10, Seattle, Washington
Richard Willey, U.S. EPA, Region 1, Boston, Massachusetts
Speakers and Discussion Group Co-Chairs:
Ned Black, U.S. EPA, Region 9, San Francisco, California
G. Allen Burton, Jr., Institute for Environmental Quality, Wright State University, Dayton,
Ohio
Brewster Conant, Jr., Department of Earth Sciences, University of Waterloo, Waterloo
Ontario, Canada
Cliff Dahm, Department of Biology, University of New Mexico, Albuquerque
Joe Dlugosz, U.S. EPA Environmental Effects Research Laboratory, Mid-Continent Ecology
Division, Duluth, Minnesota
Bruce Duncan, U.S. EPA, Region 10, Seattle, Washington
David R. Lee, Environmental Research Branch, AECL, Chalk River, Ontario, Canada
Guy Tomassoni, U.S. EPA, Office of Solid Waste, Washington, DC
D. Dudley Williams, Division of Life Sciences, University of Toronto at Scarborough, Canada
Thomas C. Winter, U.S. Geological Survey, Denver, Colorado
11
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
PRESENTATION ABSTRACTS
12
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
A Federal Statutory/Regulatory/Policy
Perspective on Remedial Decision-making with
Respect to Ground-Water/Surface-Water
Interaction
by Guy Tomassoni
The ground water/surface water interaction zone is important because 75% of Superfund and RCRA
sites are located within a half mile of a surface water body. Forty-seven percent of Superfund sites
have recorded impacts to surface water. Most RCRA sites are located adjacent to or near surface water
(presumably for ease of transportation and manufacturing). Within the last 25 years, the Clean Water
Act has succeeded in cleaning up point sources in the United States, and EPA now needs to consider
non-point sources.
"Risk-based decision making" (RBDM) has received a bad reputation within EPA because it has been
equated to "risk-based corrective action" (RBCA). A goal of this workshop is to provide the scientific
basis to convince policy-makers to allow RBDM. EPA supports RBDM, but places more emphasis on
site-specific evaluations based on sound science. RBDM generally requires a multidisciplinary
approach, an understanding of requirements, and flexibility in applicable statutes, regulations, and
policies.
There are many technical and policy issues regarding ground-water/surface-water interactions. Good
policy is flexible, and good policy comes from good technical information. This workshop therefore
may influence future policy. Superfund and RCRA remediation ("corrective action") programs. These
laws mandate protection of human health and environment. The Superfund National Contingency Plan
offers greater detail; RCRA relies more on program guidance.
Highlights from "Rules of Thumb for Superfund Remedy Selection" (http://www.epa.gov/
superfund/resources/rules/index.htm)
Superfund's goal is to return usable ground water to beneficial uses (current and future) where
practical. When this is not practical, Superfund strives to prevent further migration and exposure, and
to evaluate opportunities for further risk reduction. Ground water generally is considered "potable" if it
is so designated by the state, or considered so under federal drinking water guidelines. Preliminary
remedial goals are set at levels that protect resources—including surface waters—that receive
contaminated ground water, taking into account Clean Water Act requirements or state standards, if
they are more stringent. Attaining drinking water standards in contaminated ground water is not always
enough to protect sensitive ecological receptors. Final clean-up levels should be attained throughout
the plume and beyond the edge of any wastes left in place. The "point of compliance" for a surface
water body is where the release enters the surface water. Alternate concentration limits (ACLs) may be
considered where contaminated ground-water discharges to surface water, where contaminated ground
water does not lead to increased contaminants in surface water, where enforceable measures are
available to prevent exposure to ground water, or where restoring ground water is "not practicable."
There are about 23 Superfund ACLs nationwide. EPA expects to use treatment to address "principal
threats" posed by site where practical.
13
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
000
RCRA Setting, Based Upon the May 1,1996, Advance Notice of Proposed Rulemaking (http-//
www.epa.gov/correctiveaction'>
RCRA has similar requirements to Superfund with respect to: returning usable ground water to
beneficial uses; points of compliance for ground water and surface water; protection of surface water
from contaminated ground water; provisions for ACLs (but without an explicit link to "practicability");
and treatment of principal threats. If current human exposures are under control and no further
migration of contaminated ground water is expected, primary near-term goals are established using two
environmental indicators. Surface water becomes the boundary if the discharge of contaminated
ground water is within "protective" limits.
The OSWER Policy Directive on Monitored Natural Attenuation (MNA) was issued in final form, and
is pertinent to the ground water/surface water issue. It addresses dilution, dispersion, absorption, and
degradation—all of which occur in ground water/surface water interaction. The directive requires
controlling sources and monitoring; it stresses the need to look beyond obvious contaminants.
In summary, the majority of contaminated sites have serious potential to affect surface waters. The
federal framework allows for RBDM with respect to ground water/surface water interaction, but we
must still achieve the expectation of restoring ground water to beneficial use and ensure discharges of
ground water to surface water are protective. Key policy issues to ponder—and to pass to senior
managers—include:
• how to achieve short- and long-term protection;
• where, how, and how often to measure compliance;
• whether to restore ground water; even if it has no impact to surface water;
• the diversity of surface bodies;
• the relation of cleanup goals to the Clean Water Act's National Pollutant Discharge Elimination
System (NPDES) approach; and
• how to account for, track, and communicate total loads in watersheds.
AUTHOR INFORMATION
Guy Tomassoni, U.S. EPA, Office of Solid Waste
14
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Interaction of Ground Water and Surface Water
By Thomas C. Winter
INTRODUCTION
Surface water bodies are hydraulically connected to ground water in most types of landscapes; as a
result, surface-water bodies are integral parts of ground-water flow systems. Even if a surface water
body is separated from the ground-water system by an unsaturated zone, seepage from the surface
water may recharge ground water. Because of the interchange of water between these two components
of the hydrologic system, development or contamination of one commonly affects the other. The
movement of surface water and ground water is controlled to a large extent by the physiography
(land-surface form and geology) of an area. In addition, climate, through the effects of precipitation
and evapotranspiration, affects the distribution of water to—and removal from—landscapes.
Therefore, it is necessary to understand the effects of physiography and climate on surface water
runoff and ground-water flow systems in order to understand the interaction of ground water and
surface water.
The purpose of this paper is to: present an overview of how physiography and climate affect the
interaction of ground water and surface water and present the concept of hydrologic landscapes as a
unifying framework for study, analysis, and management of seemingly diverse landscapes. Specifically
discussed are the effects of the following factors on movement of water between ground water and
surface water: (a) position of surface water bodies within ground-water flow systems; (b) small-scale
geologic features in beds of surface water; (c) climate; and (d) hyporheic exchange.
GENERAL HYDROLOGICAL PROCESSES RELATED TO THE INTERACTION OF GROUND WATER
AND SURFACE WATER
Position of Surface Water Bodies With Respect to Ground-Water Flow Systems
Ground water moves along flow paths of varying lengths from areas of recharge to areas of
discharge. The source of water to the water table (ground-water recharge) is infiltration of precipitation
through the unsaturated zone. Ground-water flow systems can be of greatly different sizes and depths,
and they can overlie one another. Local flow systems are recharged at water-table highs and discharge
to adjacent lowlands or surface water. Local flow systems are the most dynamic and the shallowest
flow systems; therefore, they have the greatest interchange with surface water. Local flow systems can
be underlain by intermediate and regional flow systems. Water in these deeper flow systems have
longer flow paths, but they also eventually discharge to surface water. Surface water bodies that
receive discharge from more than one flow system receive that water through different parts of their
bed. Local flow systems discharge in the part nearest shore, and larger-magnitude flow systems
discharge to surface water further offshore. Because of the different lengths and travel times of water
within flow paths, the chemistry of water discharging into the surface water from different flow paths
can be substantially different.
In some landscapes, surface water bodies lie at intermediate altitudes between major recharge and
discharge areas. Surface water bodies in such settings commonly receive ground-water inflow on the
upgradient side and have seepage to ground water on the downgradient side. Furthermore, depending
15
-------
Proceedings of the Ground-Water/Surface- Water Interactions Workshop
July 2000
on the distribution and magnitude of recharge in the uplands, the hinge line between ground-water
inflow and outflow can move back and forth across part of the surface water bed.
The above characteristics of ground-water flow systems with respect to surface water apply in a
general regional sense to most landscapes. However, the detailed distribution of seepage to and from
surface water is controlled by: (a) the slope of the water table with respect to the slope of the surface
water surface; (b) small-scale geologic features in the beds of surface water; and (c) climate.
Effect of Local Water-Table Configuration and Geologic Conditions on Seepage Distribution in
Surface Water Beds
Upward breaks-in-slope of the water table result in upward components of ground-water flow
beneath the area of lower slope and downward breaks-in-slope of the water table result in downward
components of ground-water flow. These flow patterns apply to parts of many landscapes, but they are
particularly relevant to the interaction of ground water with surface water because water tables
generally have a steeper slope on both the inflow and outflow sides relative to the flat surface of
surface water bodies. The ground-water flux through a surface water bed associated with these
breaks-in-slope, whether the seepage is to or from the surface water, is not uniformly distributed
areally. Where ground water moves to or from a surface water body underlain by isotropic and
homogeneous porous media, the flux is greatest near the shoreline, and it decreases approximately
exponentially away from the shoreline. Anisotropy of the porous media, which is a function of the
orientation of sediment particles in the geologic materials, affects this pattern of seepage by causing
the width of areas of equal flux to increase with increasing anisotropy. Yet the decreasing seepage
away from the shoreline remains nonlinear.
Geologic heterogeneity of surface water beds also affects seepage patterns. Small-scale variations
in sediment type can cause the locations and rates of seepage to vary substantially over small distances.
For example, highly conductive sand beds within finer-grained porous media that intersect a surface
water bed results in subaqueous springs. The horizontal and vertical hydraulic conductivity of the
streambed can vary by several orders of magnitude because of the variability of streambed sediments.
The complex distribution of seepage patterns caused by the heterogeneous geology of surface water
beds has been documented by field studies in many
settings.
Effect of Climate on Seepage Distribution in Surface Water Beds
The most dynamic boundary of most ground-water flow systems is the water table. The
configuration of the water table changes continually in response to recharge to and discharge from the
ground-water system. Changing meteorological conditions strongly affect seepage patterns in surface
water beds, especially near the shoreline. The water table commonly intersects land surface at the
shoreline, resulting in no unsaturated zone at this point. Infiltrating precipitation passes rapidly through
a thin unsaturated zone adjacent to the shoreline, which causes water-table mounds to form quickly
adjacent to the surface water. This process, termed "focused recharge," can result in increased ground-
water inflow to surface water bodies, or it can cause inflow to surface water bodies that normally have
seepage to ground water. Each precipitation event has the potential to cause this highly transient flow
condition near shorelines as well as at depressions in uplands.
Transpiration by near-shore plants has the opposite effect of focused recharge. Again, because the
water table is near the land surface at edges of surface water bodies, plant roots can penetrate into the
saturated zone, allowing the plants to transpire water directly from the ground-water system.
16
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Transpiration of ground water commonly results in a drawdown of the water table much like the effect
of a pumped well. This highly variable daily and seasonal transpiration of ground water may reduce
ground-water discharge to a surface water body significantly or even cause movement of surface water
into the subsurface. In many places, it is possible to measure diurnal changes in the direction of flow
during seasons of active plant growth: that is, ground water moves into the surface water during the
night, and surface water moves into shallow ground water during the day.
These periodic changes in the direction of flow also can take place on longer time scales. Focused
recharge from precipitation predominates during wet periods, and drawdown by transpiration
predominates during dry periods. As a result, the two processes—together with the geologic controls
on seepage distribution—can cause flow conditions at the beds of surface water bodies to be extremely
variable. These processes probably affect small surface water bodies more than large surface water
bodies because the ratio of edge length to total volume is greater for small water bodies than it is for
large ones.
A type of landscape that merits special attention are those areas underlain by limestone and
dolomite. These landscapes, which are referred to as karst terrains, commonly have fractures and
solution openings that become larger with time because of dissolution of the rocks. Ground-water
recharge is very efficient in karst terrain because precipitation readily infiltrates through the rock
openings that intersect the land surface. Water moves at greatly different rates through karst aquifers; it
moves slowly through fine fractures and pores and rapidly through solution-enlarged fractures and
conduits. The paths of water movement in karst terrain are especially unpredictable because of the
many paths ground water takes through the maze of fractures and solution openings in the rock. Seeps
and springs of all sizes are characteristic features of karst terrains. In addition, the location where the
streams emerge can change, depending on the spatial distribution of ground-water recharge in relation
to individual precipitation events. Large spring inflows to streams in karst terrain contrast sharply with
the generally more-diffuse ground-water inflow characteristic of streams flowing across sand and
gravel aquifers.
Hyporheic Exchange
Streambeds and banks are unique environments because they are where ground water that drains
much of the subsurface of landscapes interacts with surface water that drains much of the surface of
landscapes. "Hyporheic exchange" is the term given to the process of water and solute exchange in
both directions across a streambed. The direction of seepage through the bed of streams commonly is
related to abrupt changes in the slope of the streambed or to meanders in the stream channel. This
process creates subsurface environments that have variable proportions of water from ground water
and surface water. Depending on the type of sediment in the streambed and banks, the variability in
slope of the streambed, and the hydraulic gradients in the adjacent ground-water system, the hyporheic
zone can be as much as several feet in depth and hundreds of feet in width. The dimensions of the
hyporheic zone generally increase with increasing width of the stream and permeability of streambed
sediments. Because of this mixing between ground water and surface water in the hyporheic zone, the
chemical and biological character of the hyporheic zone may differ markedly from adjacent surface
water and ground water.
Although most work related to hyporheic-exchange processes has been done on streams, processes
similar to hyporheic exchange also can take place in the beds of some lakes and wetlands because of
the reversals in flow caused by focused recharge and transpiration from ground water near surface
water, discussed above. Therefore, it is not enough to know only the relationship of surface water to
17
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
ground-water flow systems and to small-scale seepage patterns in surface water beds, because
hyporheic-exchange processes also can be important in some types of landscapes.
Hydrologic Landscapes as a Unifying Concept for Diverse Localities and Regions
As indicated above, many geologic and climatic factors affect the movement of water through a
basin. The many different types of landforms, geologic settings, and climate variations that make up
many regions of the Earth may make it seem that a unifying conceptual framework is impossible to
achieve. Indeed, it is not unusual for scientists and water- and land managers to emphasize the
uniqueness and complexity of a given locality rather than the similarities that it might have with other
localities. However, with respect to the movement of water and chemicals, many seemingly diverse
landscapes have some features in common, and it is these commonalities that need to be identified.
Only by evaluating landscapes from a common conceptual framework can processes common to some
or all landscapes be distinguished from processes unique to particular landscapes. A common
conceptual framework also would lead to development of field designs of data collection programs that
could be transferred to other landscapes having similar characteristics.
The concept of hydrologic landscapes is based on the idea that a single, simple physiographic
feature is the basic building block of all landscapes. This feature is termed a "fundamental landscape
unit," and is defined as an upland adjacent to a lowland separated by a steeper break in slope. Water
moves over the surface of a fundamental landscape unit depending upon the surface slope of the
upland, lowland, and intervening steeper slope, and it moves through the subsurface depending upon
the hydraulic characteristics of its internal geologic properties.
All landscapes can be conceived of as variations and multiples of fundamental landscape units.
Variations and multiples of fundamental landscape units can be used to define a number of general
landscape configurations; for example: (1) the width of the lowland, valley side or upland can range
from narrow to wide; (2) the slopes of the three surfaces can vary; (3) the height of the valley side can
range from small to large—that is, the upland can be only slightly higher than the lowland or it can be
much higher; or (4) small fundamental landscape units can be superimposed on any or all of the
surfaces of larger-scale fundamental landscape units.
General landscape configurations such as these can be used to define general landscape types that
describe major physiographic features of the Earth. For example:
(1) A landscape consisting of narrow lowlands and uplands separated by high and steep valley
sides is characteristic of mountainous terrain. This general configuration can be nested into
multiples at different scales within mountainous terrain as one moves from high mountain
basins to larger and larger valleys within a mountain range complex.
• A landscape consisting of very wide lowlands separated from much narrower uplands by steep
valley sides is characteristic of basin and range physiography and basins of interior drainage. In
this type of terrain, the uplands may range from being slightly higher to much higher than the
lowlands.
• A landscape consisting of narrow lowlands separated from very broad uplands by valley sides of
various slopes and heights is characteristic of plateaus and high plains.
• A landscape consisting of one or more small fundamental landscape units (terraces) nested within
a larger lowland is characteristic of riverine valleys and coastal terrain. A landscape consisting of
18
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
numerous small fundamental landscape units superimposed on both the uplands and lowlands of
larger fundamental landscape units is characteristic of hummocky glacial and dune terrain.
Common Hydrologic Characteristics of Generalized Hydrologic Landscapes
The movement of water over the surface and through the subsurface of generalized landscapes is
controlled by common physical principles regardless of the geographic location of the landscapes. For
example, if a landscape has low land slope and low-permeability soils, surface runoff will be slow and
recharge to ground water will be limited. In contrast, if the soils are permeable in a region of low land
slope, surface runoff may be limited but ground-water recharge will be high. In landscapes that have a
shallow water table, transpiration directly from ground water may have a substantial effect on ground-
water flow systems, and on the movement of ground water to and from surface water.
Landscapes characterized by multiples of fundamental landscape units can have complex ground-
water flow systems because small-scale local flow systems associated with each topographic break in
the landscape are superimposed on larger, more regional flow systems associated with larger
fundamental landscape units. Two seemingly diverse landscapes, such as riverine and coastal terrain,
have many of these types of physiographic characteristics in common, and presumably would have
many hydrologic characteristics in common as well. Ground-water flow conditions in hummocky
terrain are even more complex than riverine and coastal terrain because of the numerous small
fundamental landscape units superimposed somewhat randomly on larger and larger fundamental
landscape units. Indeed, in glacial and dune terrain, many multiples of fundamental-landscape-unit
scale can be present. Furthermore, generally shallow water tables characteristic of coastal, riverine, and
hummocky terrain result in the opportunity for highly transient local ground-water flow systems
caused by focused recharge and transpiration directly from ground water.
Implications for management of water and remediation of contaminated localities
Management of water, and remediation of contaminated localities, requires sound understanding of
hydrological processes. Contaminated ground water and surface water are common in all types of
landscapes. Because of the cost of studies and of remediation, it is nearly impossible to devote
adequate resources to the huge number of sites that need attention. Therefore, it is of great practical
value to seek transferability of study design, study results, and remediation techniques. It is suggested
that'the concept of hydrologic landscapes can serve as a foundation for determining the commonalities
of diverse localities, and sharpen the perspective of their differences. If this can be accomplished, the
transfer value of study designs and remediation methods should substantially reduce the cost of site
remediation.
REFERENCES
The material presented above was abstracted from:
Winter, T.C., Harvey, J.W., Franke, O.L., and W.M. Alley, 1998. Ground water and surface water—a
single resource, U.S. Geological Survey Circular 1139,79 pp.
Winter, T.C., 1999. Relation of streams, lakes, and wetlands to groundwater flow systems,
Hydrogeology Journal, v. 7, pp. 28-45.
Winter, T.C., 2000. The concept of hydrologic landscapes. Submitted to the Journal of the American
Water Resources Association, February.
19
-------
Proceedings of the Ground- Water/Surface- Water Interactions Workshop
July 2000
AUTHOR INFORMATION
Tom C. Winter, U.S. Geological Society, Denver, CO
20
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Hydrogeology and Biogeochemistry of the
Surface Water and Ground Water Interface of a
Mountain Stream
By Cliff Dahm
Our interdisciplinary research group has been studying the hydrogeology, biogeochemistry, and
ecology of the surface water and ground water interface of the Rio Calaveras in the Jemez Mountains
of northern New Mexico since 1991. Snowmelt is a prominent factor in the hydrogeology of both
surface discharge and the alluvial ground water of the site. Strong interannual variability in the strength
of the snowmelt signal affects both the biogeochemistry and ecology of the surface water and ground
water. Water table variation in drought years is small, and upwellmg and downwelling zones through
the bed of the channel show a complex spatial pattern, with distinct losing and gaining sections of
stream over a 150-meter reach throughout most of the year. Water table variation in wet years with
good snow pack ranges between 40 to >100 centimeters in the alluvial flood plain, and most of the
reach is gaining (upwelling) from March through May. Flow lines are directed towards the stream with
both ground water and saturated overland flow contributing to increased stream discharge. Drought
years are characterized by discharge increases as little as three times base flow while discharge
increases during wet years exceed two orders of magnitude above base flow.
Biogeochemical characteristics of the surface water and ground water are strongly influenced by
the hydrogeology. Snowmelt generates water that is rich in nitrate, dissolved organic carbon (DOC),
and oxygen. Much of the increase in dissolved organic matter and nutrients is derived from the region
of seasonal saturation (ROSS) that is inundated during snowmelt. Studies on the DOC leached from
the ROSS have shown that half of this DOC is labile and metabolized within one month. Alluvial
ground water shows strong vertical structure from the snowmelt inputs with peaks in oxygen, nitrate,
DOC, and low molecular weight organic acids in the upper 50 centimeters in the first few weeks
following snowmelt. As water table elevations drop, concentrations of oxygen, nitrate, sulfate, DOC,
and organic acids decrease, while byproducts of anaerobic metabolism such as ferrous iron,
manganous manganese, and methane increase. Surface water inputs of organic matter and nutrients
also reflect the changing hydrology that occurs from snowmelt to base flow conditions. For example,
nitrate and DOC levels are highest during the early stages of snowmelt and low during base flow
conditions. Algal primary production shows a nitrogen limitation during low-flow conditions but not
during times of increased stream discharge.
Interactions between surface waters and ground waters at this site also affect the biological
communities of the stream benthos. High discharge during periods of snowmelt scours benthic algae
and reduces chlorophyll concentrations and algal biomass throughout the reach. As snowmelt
discharge decreases, a diatom-dominated benthic algal bloom commonly occurs over much of the
stream bottom. As flows return to base flow conditions, a spatially heterogeneous pattern of algal
community structure and biomass emerges. Persistent upwelling zones at base flow, where ground
water discharges into surface water, are generally more productive reaches and composed of a complex
mix of diatoms, green algae, and cyanobacteria. More focused benthic invertebrate activity appears to
occur in these reaches. Persistent downwelling zones, where surface water recharges the ground water,
commonly have lower rates of algal primary production and contain a higher proportion of
21
-------
Proceedings of the Grouiid-Water/Surface- Water Interactions Workshop
July 2000
cyanobacteria in the algal community. Hydrogeology, nutrient availability, and interactions between
grazers and primary producers all play important roles in structuring the benthic algal community.
Integrative studies that combine hydrogeology, biogeochemistry, and aquatic ecology are needed to
fully understand the dynamics and importance of the ground water/surface water interface. Research at
Rio Calaveras in northern New Mexico has been designed to bring these disciplines together in a
multidisciplinary study of a well-instramented 150-meter reach of mountain stream. This research has
shown the importance of major hydrologic events such as spring snowmelt in the overall hydrology,
biogeochemistry, and ecology of this ecosystem. In addition, the distribution of aerobic and anaerobic
microbial processes in the alluvial ground water system and the chemical form and concentration of
nutrients and trace metals in the surface waters and ground waters are strongly affected by the
hydrogeology of the ground water/surface water interface.
AUTHOR INFORMATION
Cliff Dahm, Department of Biology, University of New Mexico, Albuquerque, NM 87131,
cdahm@sevilleta.unm.edu.
22
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Ground-water Plume Behavior Near The Ground-
Water/Surface Water Interface of a Rsver
By Brewster Conant, Jr.
INTRODUCTION
What happens to ground-water contaminant plumes as they discharge through river beds and the
ground water/surface water interface (GWSI) is not well understood. Relatively few published studies
address this issue, even though an estimated 51 percent of National Priority List sites are thought to
impact surface water (U.S. EPA, 1991) and the most common route for the contaminants to migrate
into the surface water was via ground-water transport (U.S. EPA, 1989). Understanding processes
occurring beneath and near rivers becomes particularly relevant when making remediation decisions
that are risk-based or involve natural attenuation. Such decisions could benefit greatly by identifying
important plume transport and fate processes and by conducting detailed hydrogeological studies of
plumes to characterize the spatial and temporal variations of contaminant discharges to rivers.
GROUND-WATER PLUME DEVELOPMENT
Many factors influence the transport and fate of contaminants in the subsurface prior to a ground-
water plume discharging to the surface water of a river. To understand the significance of these factors,
it is useful to consider the fundamentals of how dissolved-phase contaminant ground-water plumes are
created. Several factors play important roles in plume development:
• Physical and chemical characteristics of the contaminants
• Geometry and temporal variations in the contaminant source zone
• Transport mechanisms (advection and dispersion)
• Reactions (destructive and non-destructive)
Many of these factors are just as applicable to contaminant behavior near and beneath rivers as they are
away from the river. Knowing the behavior and concentration distribution of plumes, before they enter
the complex conditions near and beneath a river, allows better assessment of what modifying effects
near river processes have on the plume.
Contaminant Characteristics
A contaminant's physical and chemical characteristics play an important role in how the
contaminant is transported and redistributed in the subsurface and the hazard it poses to aquatic life.
Many types of contaminants are found in the subsurface including; synthetic organics, hydrocarbons,
metals, other inorganics (e.g., nitrate), radionuclides, and pathogens (e.g., viruses and bacteria).
Contaminants can be present as solids, liquids (e.g., non-aqueous phase liquids [NAPL]), dissolved in
water, or present as gasses. Each contaminant has a different propensity to solubilize, sorb,
bioconcentrate, volatilize, or react, and these characteristics affect both their mobility and toxicity.
With respect to organic compounds, the strongly hydrophobic organic compounds (e.g., PCBs,
pesticides, and PAHs) have higher bioconcentration factors and tend to be more toxic to aquatic life
than less hydrophobic organics such as chlorinated volatile organic compounds (CVOCs). The strongly
hydrophobic compounds generally have low aqueous solubilities and, when dissolved in water, move
23
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
much more slowly than ground water (i.e., adsorb and are retarded), whereas, the CVOCs have higher
solubilities and are less retarded. Consequently, many of the longer and higher concentration
dissolved-phase organic plumes in ground water are dominated by the more mobile CVOCs, which are
generally thought to be "less toxic" to aquatic life. However, aquatic biota located in the streambed and
at the GWSI (i.e., not in the surface water) may still be adversely affected by CVOCs because they
may be exposed to high concentrations in the discharging ground water prior to any dilution by surface
water. If ground-water concentrations are higher than freshwater aquatic life standards or guidelines,
the locations of these discharge zones may represent a hazard to both the benthic and hyporheic aquatic
life in the streambed, regardless of how these contaminants might later attenuate in the surface waters
of the open river channel.
Contaminant Source Zone
At many Superfund and RCRA sites, considerable effort is spent trying to delineate the source of
contaminants impacting the ground water. These sites, particularly those involving CVOCs, typically
involve so-called "point sources" of ground-water contamination resulting from spills or releases
limited over relatively discrete release areas. This paper does not address "non-point" sources of
contamination, such as nitrate and pesticide contamination from large-scale agricultural applications,
even though such "source areas" cover more of the watershed area contributing water to the stream.
Each individual contaminant source zone has a particular distribution in the subsurface. The
location, mass, and type of contaminants in the subsurface, along with characteristics of the subsurface
geology and ground-water flow, will influence whether the source produces a ground-water plume with
a continuous, variable, or a "slug" input. A source below the water table consisting of dense non-
aqueous phase liquid (DNAPL) results in continuous dissolved-phase plumes that can persist for tens
to hundreds of years if left to naturally dissolve (Feenstra, et al., 1996). Variable source plumes may be
caused by variations in waste stream inputs, or by preferential dissolution and depletion of multi-
component contaminant sources over time (Feenstra and Guiguer, 1996). Slug inputs are
"instantaneous" or short duration releases that do not persist at the initial release location and move
through the flow system as a localized mass. Of particular concern for impacts on surface water are the
continuous and variable sources which represent long term sources of contaminants to a river.
Continuous-source plumes may result in areas of the streambed being constantly exposed to high
concentrations of contaminated ground water. Because contaminants enter streambed from the ground
water below, the sediments become contaminated at ground water discharge locations. Even if those
sediments are eroded away and transported down stream, the clean materials redeposited in their place
will be subsequently contaminated by further ground water discharge.
TRANSPORT
Ground-Water Flow
The primary mechanism by which contaminants are transported away from source zones and
toward ultimate points of discharge, such as rivers, is advection (i.e., dissolved phase contaminants
moving with the ground water). Therefore, the ground-water flow system plays a fundamental role in
determining where a dissolved phase plume from a contaminant source zone will go and whether a
given surface water body may be affected. Many factors affect ground water flow including; climate
(particularly precipitation recharge), watershed characteristics, geology, hydraulic conditions (water
table slope and ground-water potential), and hydrogeologic boundary conditions (such as discharge or
"no-flow" locations). Characterizing the ground-water flow system at a site can be more large scale or
regional when initially conceptualizing potential contaminant plume flow paths. However, when
24
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
investigating point-source plumes that reach rivers, the focus needs to be on smaller scale flow
characteristics in order to accurately determine specific locations of contaminated ground water
discharge to the river.
Several types of significant vertical ground-water flow behavior can occur both on a regional scale
and on a more local scale in the vicinity of streams. Depending on the depth and location of a source
zone, the plume may be transported through what Toth (1963) termed local, intermediate, regional,
ground water flow systems. If the ground water plume develops in a shallow local flow system, it may
discharge to the nearest surface water body. If the plume develops from a deeper source zone (e.g.,
DNAPL) or is located within a regional or intermediate flow system, it may travel beneath several
lakes or streams before ultimately discharging to one of them. Winter (1999) shows some examples of
vertical cross-sectional views of ground water interactions for streams, lakes, and wetlands. Different
types of ground water/river interactions are also shown in Bear (1979, p. 52).
The lateral component ground water flow (i.e., in plan-view) near rivers exhibits a variety of
behaviors. In a study of rivers in large alluvial aquifers by Larkin and Sharp (1992) showed that ground
water flow could be base flow, under flow, or mixed flow, depending on the slope, sinuosity, and
depth of penetration of the river in the aquifer. Base flow occurs when ground water flows essentially
perpendicular to the river and discharges to it. Under flow occurs when ground-water flow near the
river is parallel to the river and does not discharge to the river channel (at least not for some great
distance). Mixed flow is a combination of base flow and under flow where ground water near the river
flows at an angle to the river and discharges to it some distance downstream. Woessner (1998) presents
some other variations in this behavior. One consequence of these possible behaviors is that plumes
entering alluvial valleys may not necessarily travel straight across the flood plain toward the river but
instead may travel down valley large distances before discharging through the stream bottom and into
the stream. In such regimes, simply trying to determine the path of a plume near a river becomes a
challenge and finding the exact areas of discharge may be very difficult.
Defining and Locating The Ground-Water/Surface-Water Interface
As ground water travels through the subsurface, it eventually reaches the GWSI near the stream or
river. At the GWSI, a transition occurs between the hydraulic, biochemical, thermal, and ecological
conditions of the surface water and those associated with the ground water. Because changes in these
parameters may be gradational, defining the location of the GWSI is not simple. The location of the
GWSI is not static and may change as a result of daily or seasonal fluctuations in river stage and
ground water flow. The GWSI can be defined as the location where water having some portion of
surface water is in contact with 100 percent ground water. This contact may occur right at the
streambed-water column interface, or it may exist at some depth within the streambed or stream banks.
The contact between the contrasting waters may be reasonably sharp or transitional. The primary
reason that the GWSI may exist within the streambed materials, as opposed to the upper surface of
them, is due to topographic variations in the streambed and changes in the slope of the river (i.e.,
hydraulic potential). Surface water may enter the sediments at downwelling zones and reenter the river
at upwelling zones (Vaux, 1968, and Boulton,1993). Downwelling generally occurs at the head of
riffles and upwelling (along with ground water discharge) occurs at the upstream edge and base of
pools. Figure 1 is a'schematic depicting downwelling and upwelling zones (in vertical cross section)
and the effect on the location of the GWSI and a discharging ground water plume. The surface water
can also leave the channel laterally and travel several meters or more into the streambanks and
eventually reenter the channel down stream (Harvey and Bencala, 1993). Where surface water leaves
the stream channel, ground water can not directly enter the channel; therefore, the GWSI and ground
25
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
water plume will be pushed away from those locations and the plume may ultimately discharge
elsewhere (Figure 1).
The GWSI is not synonymous with the term "hyporheic zone." The hyporheic zone is an
ecological term that generally refers to an ecotone where both ground water and surface water are
present in a streambed along with a specific set of biota (i.e., the hyporheos). Hyporheic zones occur as
a result of flowing waters (e.g., streams) and so the term is not applicable to quiet waters (e.g., lakes)
even though they have GWSIs too. A broader definition of the hyporheic zone has been proposed by
White (1993) that includes any area impacted by channel (i.e., surface) water, but one set of specific
criteria defining this zone has not yet been agreed upon. Delineating both the GWSI and the hyporheic
zone is important when considering ecotoxicological impacts because a unique set of benthic and
hyporheic aquatic life have adapted to the stream environment. The hyporheic zone may represent an
ecological resource needing protection. Other work suggests the GWSI may also be an important
natural attenuation zone for contaminated ground water discharge.
Dispersion
Dispersion of contaminants in ground water refers to a process by which dissolved phase
concentrations are reduced by the spreading out of the plume and hydrodynamic mixing of the water
with cleaner surrounding ground water. Reductions in plume concentrations by dispersion in ground
water flowing in aquifer sands and gravels is a very, very weak process compared to the turbulent
mixing processes that occur in the open channel flow of rivers. Because of low lateral dispersion,
plumes emanating from discrete source zones (e.g., DNAPL) are generally long thin "snake" like
plumes (Rivett, et al., 1994) rather than wide "fan" shaped plumes. One important implication of low
dispersion is that high concentration "cores" of ground water plumes (Cherry 1996), measured a short
distance downgradient of the source, may not diminish much before reaching the river. Therefore, it is
possible for very high concentration portions of the plume to reach discharge areas unless other
reactions (e.g., biodegradation) occur along those flow paths to reduce the concentrations.
In locations where surface water enters the streambed, a hyporheic zone "mixing" of surface water
with ground water may occur. This mixing process will result in what may appear to be quite
substantial reductions in plume concentrations. The mechanisms causing this type of mixing are not
well understood and result in "apparent" dispersion. Some of the uncertainty may stem from the fact
that the hyporheic zone represents primarily a "surface water" flow path as opposed to a "ground
water" flow path. For instance, the "mixing" that supposedly occurs in the hyporheic zone may
actually be the result of ground water mixing with surface water at the base of the water column which
then reenters the subsurface at a nearby downwelling zone (see the downstream downwelling zone in
Figure 1).
Reactions
Two types of reactions can
occur in the subsurface,
destructive and non-
destructive. Destructive
reactions destroy or
irreversibly transform the
contaminant into other
compounds. These reactions
include biodegradation, abiotic
FLOW
STREAM SURFACE
Downwelling
^ STREAMBED
Groundwater
Surface Water
Interface
Figure 1. Groundwater/Surface-Water Interface
(GWSI) of a Stream and Influence on Plume Discharge
26
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
reactions, and radioactive decay. Non-destructive reactions are reversible processes that may result in
changes in contaminant concentrations in the ground water but do not destroy or transform the
compound. These reactions include such things as adsorption, precipitation and dissolution, and ion
exchange. A good discussion of these types of reactions as they apply to natural attenuation of
chlorinated solvents can be found in U.S. EPA (1998). The relative importance of these'reactions may
be different in the immediate vicinity of the river than in the rest of the aquifer. In the streambed, high
organic carbon content deposits contribute to higher adsorption than is typical for the surrounding
aquifer. Adsorption of contaminants slows down (retards the movement of the contaminants relative to
ground water flow) and sequesters them for later release. Adsorption results in contaminant loading of
the sediments and in delayed breakthrough of contaminants flowing into the stream channel.
Moreover, the high organic carbon and nutrient cycling also sustains a microbiological community that
contributes to a greater potential for biodegradation. Biodegradation may greatly reduce contaminant
concentrations. In some cases these reactions may be beneficial but in others the transformation
products may be more toxic than the parent compound. In some instances (particularly petroleum
product plumes), reactions that transform organic contaminants may also consume all the dissolved
phase oxygen in the ground water and cause the ground water plume to become anaerobic. The adverse
effect of this anaerobic water on the hyporheic and benthic aquatic life (that require oxygen to live)
may be even greater than the toxic effects of the contaminants. ,
A TETRACHLOROETHYLENE (PCE) GROUND-WATER PLUME DISCHARGING TO A RIVER
To illustrate the importance of some of the above factors, results of investigations are presented for
a site located in Angus Ontario, where a dissolved-phase PCE ground water plume from a dry cleaning
facility discharges into the nearby Pine River. Previous subsurface investigations at this site using the
Waterloo Profiler (Pitkin, 1994; Writt 1996) and recent work (Conant, unpublished data) have
delineated a dissolved phase ground water plume that emanates from a PCE DNAPL source area. The
plume travels 205 m laterally through a shallow but locally confined aquifer before discharging upward
through a silt and peat semi-confining unit and then the sandy streambed deposits underlying the Pine
River. The plume is approximately 50 m wide and has a vertical thickness of 4 to 6 m. Water quality
data collected with the Waterloo Profiler show that the peak PCE concentrations in the plume at the
bank of the river (<5 m from the river) are about 8000 ug/i Virtually no PCE degradation products
were detected in the aquifer beneath the stream bank. Drivepoint piezometers screened in the aquifer at
the river's edge show that there is a strong upward hydraulic gradient at the river. These piezometers
have water levels approximately 1 m higher than the river stage. Water quality testing beneath the
opposite bank of the river shows that the plume does not pass beyond the opposite bank.
Periodic sampling of the river water where the ground water plume discharges has detected no
contamination, or very low PCE concentrations, generally less than 2 ug/i No PCE degradation
products have been detected in the surface water. The river is about 14 m wide and during most of the
year is generally less than 0.75 m deep and flows at approximately 1.5 to 2.9 cubic meters per second.
The estimated total flux of dissolved PCE contamination traveling within the aquifer ground water
toward the river each year (expressed as equivalent pure phase PCE) is approximately 15 to 40 liters
(Writt 1996). In the River channel massive dilution of the discharging PCE ground water plume by the
surface water occurs and the plume does not appear to significantly impact the surface water quality.
However, high concentrations of contaminants within the streambed itself represent locations where
adverse ecological impacts may be occurring. At some locations, concentrations in water samples
collected from within the streambed were much higher than EPA's Freshwater Aquatic Life Chronic
Toxicity Standard for PCE of 840 ug/0 and the Canadian Water Quality Guideline of 110 ug/« for the
protection of aquatic life.
27
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
My 2000
Plume water traveling through the streambed deposits is subject to a wide range of hydrological
and geochemical (redox) conditions which are spatially variable on a scale of centimeters to meters.
Streambed temperature surveys have identified areas of the streambed dominated by ground water
discharge. Hundreds of water samples have been collected to characterize discharge zones and locate
the plume. The Waterloo Profiler, the newly developed "Mini-Waterloo Profiler," and "driveable
multilevel samplers" have been used to collect interstitial water samples from the streambed and
underlying shallow aquifer. These samples have been analyzed for both inorganic and organic
parameters. Soil coring, ground penetrating radar surveys, and slug testing of streambed mini-
piezometers have also been used to help develop a conceptual model of the subsurface system.
(?) NONE
^\ Substream Flow
jS \Downwelling-
m \
)SHORT
CIRCUIT
Springs
Seeps
(T) MODERATE
^-f Semi-Permeable
Deposits
(4)HIGH
Geologic
Window
HIGH PERMEABILITY
AQUIFER
Four different types of flow
conditions have been observed
beneath the river at the site and are
been associated with varying
geochemical conditions. The four
types of ground water flow in the
streambed include: no flow, short
circuit, high flow, and low to
moderate flow (see Figure 2). In no
flow locations, no ground water is
discharging to the stream as a result
of geological barriers or hydraulic
barriers like downwelling.
Consequently, at those locations the Figure 2. Types of Groundwater Discharge
interstitial water in the streambed is geochemically quite similar to surface water and is not
contaminated. The "short circuit" condition refers to discharge at springs and seeps where PCE
contaminated ground water flows rapidly up through very localized gaps in the semi-confining unit and
undergoes little or no attenuation or modification. In high flow areas, more permeable deposits result
in areas of higher ground water flux. These areas are reflected in strong temperature anomalies at the
streambed surface. More rapid flow and shorter residence times in the streambed deposits results in the
discharge of contaminated ground water that has been only briefly exposed to reducing conditions (i.e.,
anaerobic and nitrate reduction). Consequently, PCE contaminated ground water has undergone very
little degradation and attenuation. In the low to moderate ground-water discharge zones, contaminated
ground water flows up through moderately permeable geological deposits where sulfate reducing and
methanogenic conditions occur and substantial reductive dehalogenation of PCE is indicated by the
presence of relatively high concentrations of degradation products (i.e., 100s to 1,000s of [ig/{ of
trichloroethylene, cw-l,2-dichloroethylene, vinyl chloride, ethene, and ethane). PCE concentrations at
one location dropped from about 3700 ug/0 to less than 5 ug/0 within a vertical distance of 15 cm and
there was a corresponding increase in the concentrations of degradation products which was primarily
cw-l,2-dichloroethylene (see Figure 3). In low to moderate ground-water discharge areas PCE
concentrations in the streambed are.reduced to below the EPA's Freshwater Aquatic Life Chronic
Toxicity Standard. At some of those locations, however, 100s up to a 1800 p.g/£ of vinyl chloride (a
human carcinogen) has been created. The potential hazard posed by vinyl chloride is unknown because
it does not have an aquatic life water quality standard or guideline. In the short circuit and high ground
water discharge zones the concentrations in the streambed were observed to be higher than the EPA
standard for PCE. At this site the potential impact of the plume is clearly quite spatially variable.
In terms of the overall plume behavior, it is important to note that the only place where substantial
degradation and transformation of PCE is observed is in the last 3 m of the plume's flow path from the
source area. Some portions of the plume that have traveled 200 m laterally through the aquifer and
28
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
arrive at the streambed as PCE, may
end up transforming completely and
discharging to the surface water
column as vinyl chloride or cis-1,2-
dichloroethylene instead, At this site,
water quality monitoring in the aquifer
upgradient and immediately adjacent to
the river does not fully characterize the o
type or concentration of contaminants £
that ultimately enter the surface water.
VOCs
CONCENTRATION ug/L
0 2000 4000 6000 8000 10000
REDOX PARAMETERS
CONCENTRATION mg/L
0 10 20 30 40
a.
UJ
a
SUMMARY
Determining the location and
magnitude of contaminant discharges to
rivers from ground-water plumes-is a Groundwater Concentrations at a
complex hydrogeological and 9 Moderate Discharge Zone
biogeochemical problem. Determining
specific ground-water flow paths near a stream and its GWSI is not an easy task. Moreover, the effect
of transport and fate processes on the plume near the GWSI and within streambed deposits may be
quite different from those observed in the aquifer further away from the stream. Large changes in
geochemical conditions and plume concentrations may occur in the streambed over intervals of only
centimeters, both vertically and horizontally. Measurements of ground water plume concentrations
made adjacent to the stream or in the aquifer underlying the stream banks may not accurately reflect
either the concentrations of contaminants in the streambed or the contaminant flux that ultimately
reaches the surface water. The Angus study shows that a range of different plume discharge behaviors
can occur at a single site and that closely spaced vertical and horizontal water quality sampling is
necessary to detect these behaviors. In some places, reactions in the streambed transformed
contaminants to daughter products and reduced the overall concentration of contaminants discharging
to the river. In other places no attenuation of contaminants occurred and aquatic life in the streambed
at these discharge zones had the greatest exposure to the parent compound. Aquatic life in the surface
water column is typically less at risk from ground water contamination than benthic organisms because
of dilution with clean surface water. The current challenge for hydrogeologists is to assist ecologists in
identifying potential problem discharge zones so the toxicological impacts on benthic and hyporheic
aquatic life can be assessed.
REFERENCES
Bear, J., 1979. Hydraulics of Ground Water. McGraw-Hill Book Company, New York, p. 569.
Boulton, J. A., 1993. Stream ecology and surface-hyporheic hydrologic exchange: implications,
techniques and limitations, Australian Journal of Marine and Freshwater Research, v. 44, pp. 553-
564.
Cherry, J. A., 1996. "Conceptual models for chlorinated solvent plumes and their relevance to intrinsic
remediation," Symposium on Natural Attenuation of Chlorinated Organics in Ground Water,
Dallas Texas, Sept. 11-13. EPA/540/R-96/509, pp. 29-30.
Feenstra, S. and N. Guiguer, 1996. Dissolution of dense non-aqueous phase liquids (DNAPLs) in the
subsurface," Chapter 7 in Dense Chlorinated Solvents and other DNAPLS in Ground Water. J.F.
Pankow and J.A. Cherry Eds., Waterloo Press, Portland Oregon, pp. 203-232.
29
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Feenstra, S., J.A. Cherry, and B.L. Parker, 1996. "Conceptual models for the behavior of dense non-
aqueous phase Liquids (DNAPLS) in the subsurface," Chapter 2 in Dense Chlorinated Solvents and
other DNAPLS in Ground Water. J.F. Pankow and J.A. Cherry Eds., Waterloo Press, Portland
Oregon, pp. 53-88.
Harvey, J. W. and K. E. Bencala, 1993. The effect of streambed topography on surface-subsurface
water exchange in mountain catchments, Water Resources Research, v. 29, no. 1, pp. 89-98.
Larkin, R. G. and J. M. Sharp Jr., 1992. On the relationship between river-basin geomorphology,
aquifer hydraulics, and ground-water flow direction in alluvial aquifers, Geological Society of
America Bulletin, v. 104, pp. 1608-1620.
Pitkin, S. E., 1994. A point sample profiling approach to the investigation of ground water
contamination, M.Sc. Thesis, University of Waterloo, Waterloo, Ontario Canada, p. 167.
Rivett, R.O., S. Feenstra, and J. A. Cherry, 1994. Transport of a dissolved-phase plume from a residual
source in a sand aquifer, Journal of Hydrology, v. 159, pp. 27-41.
Toth, J. 1963. A theoretical analysis of ground water flow in a small drainage basins, J. Geophys. Res
v. 68, pp. 4795-4812.
U.S. EPA, 1989. The Nature and Extent of Ecological Risks at Superfund Sites and RCRA Facilities,
EPA-230-03-89-043, U.S. EPA Office of Policy Analysis, Office of Policy, Planning, and
Evaluation, U.S. EPA, Washington, DC, p. 212.
U.S. EPA, 1991. National Results of NPL Characterization Project. EPA/540/8-91/069. U.S. EPA
Office of Solid Waste and Emergency Response, Washington DC, p. 108.
U.S. EPA, 1998. Technical Protocol For Evaluating Natural Attenuation of Chlorinated Solvents in
Ground Water, EPA/600/R-98/128, U.S. EPA Office of Research and Development, Washington
DC, p.78.
Vaux, W.G. 1968. Intragravel flow and interchange of water in a streambed, Fishery Bulletin, 66, pp.
479-489.
White, D. S., 1993. Perspectives on defining and delineating hyporheic zones, Journal of the North
American Benthological Society, v. 12, no. 1, pp. 61-69.
Winter, T. C., 1999. Relation of streams, lakes, and wetlands to ground water flow systems,
Hydrogeology Journal, v. 7, no. 1, pp.28-45.
Woessner, W. W., 1998. Changing views of stream-ground water interaction, Eds. J. Van Brahana, Y.
Eckstein, L.K. Ongley, R Schneider and J.E. Moore. Proceedings of the Joint Meeting of the
XXVm Congress of the International Association of Hydrogeologists and the Annual Meeting of
the American Institute of Hydrogeology: Gambling with Ground water—Physical, Chemical and
Biological Aspects of Aquifer-Stream Relations. Las Vegas, Nevada, Sept. 28-Oct. 2, 1998, pp.l-
Writt, R.J., 1996. The Angus PCE Plume—Aquifer Sedimentology and Plume Anatomy, MSc. Thesis
Project, University of Waterloo, Waterloo, Ontario Canada, p. 200.
AUTHOR INFORMATION -
Brewster Conant Jr., Department of Earth Sciences, University of Waterloo, Waterloo Ontario,
Canada
30
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Assessment Approaches and Issues in
Ecological Characterizations
By G. Allen Burton, Jr. and Marc S. Greenberg
Ecosystems are extremely complex; consisting of a multitude of species that vary widely in the
sensitivity to contaminants and who are dependent on each other to varying degrees. Ecosystems are
routinely impacted by natural disturbances (e.g., high or low flows, habitat alteration, food
availability), some of which can be quite severe and cover over large areas (e.g., hurricanes, flooding,
drought, anoxia, temperature shock, invasive species, disease). These natural disturbance events must
be considered when trying to ascertain the role of human (anthropogenic) disturbances. Ecosystems are
also dynamic and vary through space (spatially) and time (temporally). These variations can be
important at the millimeter scale where microenvironments determine nutrient and contaminant
availability. However, distances of kilometers may be more significant for biogeographical issues such
as forest fragmentation, foraging, and migration. Practical time scale issues vary in importance from
minutes to decades. So, when we try and discern whether or not ecosystems are impacted by
anthropogenic disturbances, we must do so in the context of these ecosystem complexity issues. The
importance of an anthropogenic disturbance, such as exposure to chemicals, follows these natural
spatial and temporal processes to a large extent. In other words, the significance of chemical exposure
to an organism, population, or community may vary in importance over distances of mm to km and
time periods of minutes to years, depending on the organism's behavior and the chemical's fate.
However, these somber realities of complexity are not insurmountable. The following discussion will
show effective ways of determining whether ecosystems are significantly impacted and which stressors
are causing the primary problems.
Traditional water quality assessments typically focus on water quality standards, which assume if a
single chemical criteria is exceeded then impairment to the receiving water or its beneficial use
designation may exist. A limited number of states, such as Ohio and North Carolina, have also ,
developed biocriteria, which rank indigenous fish and benthic macroinvertebrate communities into
classifications ranging from poor to excellent. Toxicity testing of surrogate species, such as the fathead
minnow (Pimephales promelas) and water flea (Ceriodaphnia dubia), have been incorporated into the
National Pollutant Discharge Elimination System (NPDES) permit program for wastewater effluents.
Toxicity testing requirements are occasionally incorporated into a permit and require testing of
upstream water, effluent, and near- and far-field receiving water samples. More recently, sediment
toxicity test methods have been developed by the U.S. Environmental Protection Agency (U.S. EPA)
(U.S. EPA 1994); however, these have not been incorporated into "NPDES permits and have been used
to only a limited extent in assessments of aquatic ecosystem contamination. Each of these approaches
has associated strengths and weaknesses, describing one aspect of contaminant effects under a certain
set of exposure assumptions, which may or may not be realistic. These approaches can be used with
confidence in situations where gross contamination exists. However, most of our current
environmental concerns are more complex and often of a chronic toxicity nature. Often in remediation
projects one must decide to what point or level clean-up should extend. In complex watersheds, there
often is a need to decipher to what degree each potential source of pollution is contributing to
impairment. It is now well accepted by those in the field of ecotoxicology that an integrated approach
that combines several traditional assessment approaches, plus other non-standardized methods is
necessary to reduce the uncertainty of whether significant ecosystem contamination exists (e.g., Burton
31
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
1999, Chapman, et al., 1992). This integrated approach is described below in the context of its
application to ground water and surface water transition zones.
All ecosystems and their resident species are stressed at one time or another. We tend to focus on
that subset of ecosystems where anthropogenic stressors are at issue. Since natural and anthropogenic
stressors can be physical, chemical, or biological, the assessment process must consider all of them.
Ecosystems, their interacting components, and the stressors which affect them are dynamic and not in
equilibrium. So the assessment process must also consider organism exposures to stressors from a
magnitude, frequency, and duration perspective. These realities dictate that an integrated assessment
contain the components listed in Table 1.
Table 1. Elements of an Integrated Assessment of Aquatic Ecosystems
Component
Habitat
Chemistry
Biota
Toxicity
Sampling Media
Drainage area, riparian zone, waterway
Frequency
Seasonal
Drainage area soil, water, sediment, pollutant sources, and Low and high
tissues of key receptors flow
Benthos, fish, and fish-eating wildlife Seasonal
Surface water, pore water and sediment (laboratory and in Low and high
situ) flow
Within the four general components of habitat, chemistry, indigenous biota, and toxicity the
primary stressors and receptors can be identified with the proper sampling and test design. This
approach can follow the ecological risk assessment paradigm whereby there is a problem formulation
step, followed by field and laboratory assessments of exposure and effects and finally a risk
characterization via a weight-of-evidence approach.
For assessing potential contamination in ground water/surface water transition zones it is critical to
team hydrologists, hydrogeologists, aquatic biologists/toxicologists, and environmental chemists in the
assessment process. A tiered assessment approach is the most cost effective way to conduct an
integrated assessment, eliminating the collection of data which may not be necessary (Table 2). The
specific measurement methods that are used in these approaches should be optimized for each study,
depending on the problem and questions being asked. For example, in freshwater systems this means
optimizing the indicator species used for toxicity testing and response endpoints (e.g., sublethal
biomarkers, growth, reproduction, tissue residues, mortality), selecting the appropriate exposure in situ
(e.g., surficial vs. deep sediments, small mesh to reduce suspended solids, UV blockers to prevent
photo-induced toxicity from polycyclic aromatic hydrocarbons), or selecting the appropriate data
analysis methods for the benthic invertebrates (e.g., metrics like Invertebrate Community Index,
orthogonal comparisons).
Assessing the ecological significance of ground water/surface water transition zones will present
some new challenges. Virtually no contaminant effects research has been conducted on biological
communities which inhabit the hyporheic zones. It will be critical to establish good reference sites as a
point of comparison. These transition zones are particularly important in the storage and retention of
nutrients (and possibly contaminants), biological and chemical transformations, as a refugia for
invertebrates, and a base of the aquatic food web. Therefore, the measurement endpoints should be
focused on determining effects on these traits. Appropriate measurement endpoints could include:
indigenous microbial activity, organic matter/nutrient cycling (for more advanced studies), invertebrate
32
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Table 2. Tiered Assessment Approach for Characterizing Ground water/Surface Water Transition
Zone Contamination*
Tier la:Hydrological characterization of transition zone locations, upwelling vs. downwelling,
rates, surface water dynamics.
Tier IbtCharacterization of benthic invertebrates (sediment surface and hyporheous, grabs,
colonization, transplants) and habitat quality.
Tier 2: Toxicity testing of indicator species (sediment (laboratory); surface water (high and low
flow), surficial sediment and pore water (in situ)). Tissue residue analysis of Lumbriculus
variegatas (in situ exposure) and dominant indigenous species.
Tier 3: Site-specific studies to separate physical and chemical stressors with associated
chemical analyses, if needed.
* Assumes initial problem formulation process has identified contamination of ground water or surface
water with potential transfer to the other.
community indices (meiofaunal and macrofaunal-grab and colonization), tissue residues of dominant
species, in situ toxicity, and in situ physicochemical profiles (e.g., via peepers, datasondes).
If Tiers land 2 indicate that the surface or ground waters are toxic and/or are impacting the
indigenous community then Tier 3 may be necessary to tease out which stressors dominate at the site.
These are very site-specific based designs, but can include novel, yet proven, tools such as ecological
food web modeling, semi-permeable membrane devices (SPMDs) to look at bioaccumulation potential,
toxicity identification evaluations (TIEs) which fractionate chemical classes for toxicity testing, and
stressor identification evaluations (SEEs) which are in situ based TIEs but incorporate other physical
stressor determinations (Burton et al 1996 and 1998; Greenberg et al 1998), and more detailed
characterizations of community effects and exposure dynamics.
REFERENCES
Burton, G.A., Jr., 1999. Realistic assessments of ecotoxicity using traditional and novel approaches,
Aquatic Ecosystem Health and Management, v. 2, pp.1-8.
Burton, G.A., Jr., Hickey, C.W., DeWitt, T.H., Roper, D.S., Morrisey, D.J., Nipper, M.G. 1996. In situ
toxicity testing: teasing out the environmental stressors, SETAC News, v. 16, no.5, pp. 20-22.
Burton, G.A., Jr., Rowland, C., Greenberg, M., Lavoie, D., and J. Brooker, 1998. Determining the
effect of ammonia at complex sites: laboratory and in situ approaches, Abstr. Annu. Meeting Soc.
Environ. Toxicol. Chem. Charlotte, NC.
Chapman, P.M., Powers, E., and G.A. Burton, Jr., 1992. Integrative assessments in aquatic
ecosystems, in Sediment Toxicity Assessment, G.A. Burton, Jr. (ed.), Lewis Publishers, Boca
Raton, FL. p. 167.
Greenberg, M., Rowland, C., Burton, G.A., Hickey, C., Stubblefield, W., Clements, W., and P.
Landrum,1998. Isolating individual stressor effects at sites with contaminated sediments and
waters, Abstr. Annu. Meeting Soc. Environ. Toxicol. Chem. Charlotte, NC.
U.S. Environmental Protection Agency, 1994. Methods for Measuring the Toxicity and
Bioaccumulation of Sediment-associated Contaminants with Freshwater Invertebrates, Office of
Research and Development, Washington, DC, EPA/600/R-94/024.
33
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
AUTHOR INFORMATION
G. Allen Burton, Jr. and Marc S. Greenberg, Institute for Environmental Quality, Wright State
University, Dayton, Ohio 45435
34
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
DeSineaiion, Quantification, and Mitigation of
Discharging P9umes
By David R. Lee
INTRODUCTION
Methods have been developed for locating and sampling ground water and solute discharge areas
on the beds of surface waters. In many settings, these can aid in the assessment of natural attenuation
or in estimating the direct flux of ground water contaminants to surface waters. Where plumes are not
sufficiently attenuated by natural mechanisms before reaching surface waters, passive subsurface
treatment methods, as exemplified by the Chalk River wall and curtain, are now demonstrated at full
scale. The purpose of this presentation was to highlight the author's approach to these problems.
The concept of monitored, natural attenuation sounds good, but putting it into practice will take
careful work. If it has been difficult to monitor natural attenuation in relatively simple, well-
characterized hydrogeological settings, then it will be even more difficult to perform such monitoring
near the beds of surface water bodies. Transients in flow and changes in water levels are only a part of
the difficulty.
Another unappreciated difficulty, is the profound influence of geologic heterogeneity on
contaminant migration. Heterogeneity can result in orders of magnitude variations in flow within a
relatively small volume of earth. Many people believe that hydraulic conductivities at a site vary by
factors of 1.2 to 1.5. However, in actual fact at most sites, hydraulic conductivities vary by factors of
10 to 300! Since one of the controls on attenuation is ground-water residence time, attenuation may
vary widely across most sites. Therefore, the technical information on which to base an evaluation of
attenuation at real sites depends upon the determination of spatial distributions in flow, particularly on
finding the faster flow areas at each site.
Measurements of hydraulic gradient can indicate large discharge areas. However, the results of
numerous seepage studies have shown that areas of rapid discharge can be small and easily missed. If
not located, zones of contaminant entry will not be assessed. In other words, if flow is focused, as it
often is, the impacts of the discharge and the processes or evidence for attenuation may have to be
monitored within the relatively small, fast-flow areas, which have the greatest potential for poor
attenuation and transport of contaminants to surface. If flow rates exceed the required reaction times,
the potential for subsurface attenuation may not be realized. High flow areas occur where there are
preferential flow paths, such as sand stringers or interconnected zones of higher permeability. The
areas where these flow paths intersect surface waters may be overlooked without thorough field work.
Even in relatively homogeneous terrain, flow may be highly focused at the shorelines and transport
may be rapid.
IS ATTENUATION WISHFUL THINKING OR REALITY?
While it is reasonable to expect some attenuation for many contaminants at most sites, those who
seek to monitor attenuation or to measure impact face many pitfalls. Sampling must include the faster
flow areas in order for measurements of flow and contaminant concentrations to be representative. If
the act of sampling dilutes the ground-water concentrations, and this is easy to do near the
35
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
sediment/water interface, the sample and the resulting chemical analyses may be inappropriate for
contaminant flux calculations. Thus conclusions may be biased and non-conservative as a result of
incomplete or improper sampling. It may be easier to find evidence for attenuation than to establish
sufficient attenuation.
DEVELOPMENT OF METHODS TO LOCATE AREAS OF SIGNIFICANT DISCHARGE
There is a growing awareness that the application of existing technologies is key to valid
monitoring of natural attenuation. One promising method is the sediment probe, a specifically designed
for the detection of ground water upwelling (Lee ,1985; Lee and Beattie, 1991). Towed behind a
moving boat, the sediment probe is in contact with sediments, and it measures sediment properties.
Once areas of ground water discharge have been found and delineated, they may be assessed using
traditional, quantitative methods (Lee and Dal Bianco, 1994; Harvey, et al., 1997; Lee, et al., 1999).
Traditional methods such as piezometers (e.g., Lee and Harvey 1996; Geist, et al., 1998) may be used
for pore water collection and measurement of hydraulic head and conductivity. Under some conditions,
seepage meters (e.g., Lee and Cherry, 1978; Lee. 1977; Lee and Hynes 1978) may be appropriate for
measuring the flux of ground water across the sediment/water interface.
The sediment probe has been used to find and confirm discharge areas on the cobble sediments and
in the 2m/s currents of the Columbia River (Lee, et al. 1999). In that work, quantitative samplers
showed that, without exception, probe "hot spots" were areas of ground water inflow and some of
these inflows bore contaminants.
The sediment probe has also been used to locate ground water discharge into the shallow ocean
(Vanek and Lee, 1991). Other methods have been developed to aid in demonstrating attenuation near
the interface (e.g., Lee, 1988; Winters and Lee 1987).
Having been proven in a variety of settings, the sediment-probe method is now ready for use in
identifying areas where it may be necessary to monitor attenuation, or the lack thereof. This is
essentially a reconnaissance method, a targeting tool. It requires a slight contrast in dissolved solids
concentrations between the ground water of interest and the overlying surface water. Where the plume
itself is different in dissolved solids, it can tell us, "No, the contaminant is not here," or "Yes, it is,
and, the signal keeps getting larger as we move in this direction." By applying such methods, it is
possible to design a monitoring system for contaminant attenuation and to provide a basis for deciding
whether to rely on the process of natural attenuation. Clearly, in order to show that attenuation is
sufficient, it must be known where discharge occurs, particularly where it is most rapid, and evidence
of acceptable flux of solutes must be obtained.
There is potential for incorporating additional sensors on the sediment probe to make it sensitive to
conditions other than electrical conductance.
DEVELOPMENT OF METHODS FOR PLUME MITIGATION
In settings where attenuation is found to be insufficient, subsurface treatment systems, like those
first described by McMurty and Elton (1985), can be constructed to enhance natural attenuation
mechanisms and therefore minimize impacts on surface waters. An example of such a treatment
system is the wall and curtain at the Atomic Energy of Canada, Ltd.'s (AECL) Chalk River
Laboratories.
36
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
The wall-and-curtain treatment system was installed in 1998 to mitigate the discharge of a
strontium-90 plume. In this system, contaminated ground water is directed through a subsurface,
permeable, granular curtain of a natural, ion exchange mineral, called clinoptilolite. Based on the
results of in situ testing, clinoptilolite was highly absorbent for strontium. A bed of clinoptilolite 2 m
thick was installed underground. It is predicted to retain the strontium-90 for at least 60 years, during
which time its concentration will decay to one-fourth (or less) of the input concentrations. It is
expected that this subsurface facility will operate passively at low cost with no maintenance except for
the required effluent monitoring. Unlike other methods for subsurface treatment, the wall-and-curtain
provides an adjustable capture zone and a single point of flow for checking regulatory compliance
(Lee,etal., 1998).
FINAL COMMENT
In the process of exercising these methods at major contaminant sites, I have concluded that two
factors have combined to create a vicious circle. The factors are 1.general lack of understanding of
ground water-contaminant seepage to surface water and 2. self interest among plume owners The
vicious circle is as follows: if there is little proof of a problem and little public understanding, there is
little regulatory demand for better information and little funding for developing and applying methods.
Many contaminant plumes have been mapped to the margin of a river, lake, wetland or estuary.
But, there is little advantage for a plume owner to map it further unless this is required. Piped effluents
must meet or exceed drinking water standards, but there is little enforcement of the same water-quality
standards where it is a ground water contaminant plume, not a pipe-flow, that is entering surface
waters. Without measurements, there is little understanding and no violations. Or, if measurements at 2
or 3 points looked OK, then the discharge was deemed OK. We humans tend not to seek what we fear
we might find. When things are out of sight, they are out of mind.
CONCLUSION
Methods have been developed, applied successfully and have shown the movement of ground
water contaminants to surface waters. It is hoped that this workshop will result in broader application
of the methods highlighted here and other, equally appropriate, methods that have not been mentioned
(my apologies). Hopefully with the issuance of these workshop proceedings, the EPA will begin the
task of requiring site-specific evidence where natural attenuation is claimed to be a remedy, but is not
monitored, and will require mitigation where attenuation is not sufficient.
ACKNOWLEDGMENT
The support and collaboration of colleagues with Atomic Energy of Canada, the University of
Waterloo and the University of North Dakota and the help of several graduate students, was essential
to my work in this field for 30 years.
REFERENCES
U.S. Environmental Protection Agency, 1990. An Annotated Bibliography of the Literature Addressing
Non Point Source Contaminated Ground-water Discharge to Surface Water (Draft), Office of
Ground-Water Protection, Washington, DC, 302 pp.
Geist, D.R., M.C. Joy, D.R. Lee ,and T. Gonser, 1998. A Method for Installing Piezometers in Large
Cobble Bed Rivers, Ground Water Monitoring & Remediation, v. 18, no. 1, pp. 78-82.
37
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Harvey, F.E., D.R. Lee, D.L. Rudolph, and S.K. Frape, 1997. Locating Ground water Discharge in
Large Lakes Using Bottom Sediment Electrical Conductivity Mapping, Water Resources Research,
v. 33, no. 11, pp. 2609-2615.
Lee, D.R., 1977. A Device for Measuring Seepage Flux in Lakes and Estuaries, Limnology &
Oceanography, v. 22, no. 1, pp. 140-147.
Lee, D.R. and J.A. Cherry, 1978. A Field Exercise on Ground water Flow Using Seepage Meters and
Mini-Piezometers, J. of Geological Education, v. 27, pp. 6-10.
Lee, D.R. and H.B.N. Hynes, 1978. Identification of Ground water Discharge Zones in a Reach of
Hillman Creek in Southern Ontario Water, Poll. Res. Canada, v. 13, pp. 121-133.
Lee, D.R. 1985. Method for Locating Sediment Anomalies in Lakebeds that can be Caused by Ground
water Flow, J. of Hydrology, v. 79, pp. 187-193.
Lee, D.R., 1988. Six In Situ Methods for Study of Ground water Discharge, in Proc. of the Int.
Symposium on Interaction between Ground water and Surface Water, May 30-June 3, Ystad,
Sweden, pp. 556-566.
Lee, D.R. and W.J. Beattie, 1991. Gamma Survey Probe for Use on Ocean, Lake, Estuary and River
Sediments, U.S. Patent #5,050,525.
Lee, D.R. and R. Dal Bianco, 1994. Methodology for Locating and Quantifying Acid Mine Drainage in
Ground Waters Entering Surface Waters, in Int. Land Reclamation and Mine Drainage Conference
and Third Int. Conference on the Abatement of Acidic Drainage, Vol. 1, Proceedings of a
Conference, April 24-29, Pittsburgh, PA, pp. 327-335.
Lee, D.R. and F.E. Harvey, 1996. Installing Piezometers in Deepwater Sediments, Water Resources
Research, v. 32, no. 4, pp. 1113-1117.
Lee, D.R., D.J.A. Smyth, S.G. Shikaze, R. Jowett, D.S. Hartwig, and C. Milloy, 1998. Wall-and-
Curtain for Passive Collection/Treatment of Contaminant Plumes, in Designing and Applying
Treatment Technologies (Eds. G.B. Wickramanayake and R.E. Hinchee), 1st Int Conf on
Remediation of Chlorinated and Recalcitrant Compounds, May 18-21, Monterey, CA, pp. 77-84 .
Lee, D.R., D.R. Geist, K. Saldi, D. Hartwig, and T. Cooper, 1999. Locating Ground-Water Discharge
in the Hanford Reach of the Columbia River, In preparation.
McMurty, D.C. and R.O. Elton HI, 1985. New Approach to In-Situ Treatment of Contaminated
Ground waters, Environmental Progress, v. 4, no. 3, pp. 168-170.
Vanek, V. and D.R. Lee. 1991. Mapping Submarine Ground water Discharge Areas—An Example
From Laholm Bay, Southwest Sweden, Limnology & Oceanography, v. 36, no. 6, pp. 1250-1262.
Winters, S.L. and D.R. Lee, 1987. In Situ Retardation of Trace Organics in Ground water Discharge to
a Sandy Stream Bed, Environmental Science & Technology, v. 21, no. 12., pp. 1182-1186.
AUTHOR INFORMATION
David R. Lee, Environmental Research Branch, AECL, Chalk River, Ontario, Canada
38
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Field Technology and Ecological
Characterization of the Hyporheic Zone
By D. Dudley Williams
The hyporheic zone is a 3-dimensional aquatic interstitial ecotone formed within the mixed
substrate particles that comprise the bed of a natural, running water channel (Figure 1). It is a middle
zone bordered by the surface water of the stream or river above, and by the true ground water below.
Although it receives water from both of these sources, the relative strengths of input depend on the
configuration of the bed materials and interstitial flow paths, and on the prevailing hydraulic heads.
These heads vary spatially and seasonally to alter hyporheic habitat volume arid to produce ragged-
edged boundaries to the zone (Williams, 1993). Water that flows across these boundaries is subject to
changes brought about by distinctive, local chemical and physical properties, microbial processes, and
metazoan community dynamics.
Hyporheic research has been progressing at varying'rates over the past 30 years, although, recently,
progress has been more sustained and intense. Undoubtedly, one of the major factors that limited
progress in the 1970s and 1980s was the perception that it is very difficult to sample the hyporheic
Water
table
-? Water
:',.'• table
Figure 1. Diagrammatic section through a stream channel showing the approximate position of the
hyporheic zone during winter, low flow conditions
39
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
zone in any meaningfully quantitative manner. True, extracting largely soft-bodied invertebrates from
the interstices among highly heterogeneous and hard lotic bed materials is difficult. However, a
sufficient number of techniques now exists (some of them old, but with recent modifications) that
makes ecological characterization of this zone possible.
Many running water invertebrates can be collected from hyporheic sediments. Typically, maximum
densities may occur around 10 to 40 cm below the streambed surface, but densities of-700
invertebrates per 1 liter of sediment at 100 cm depths are not uncommon (Williams and Hynes, 1974).
The hyporheic fauna itself has two main components (Table 1). Differences in the spatial and temporal
residence profiles of these two components suggest different functional roles for the two groups within
the zone. Although not conclusively proven, there is evidence to suggest that the hyporheic zone may
act as a refuge from extreme conditions on the streambed surface (Williams and Hynes, 1977). For
example, spates are known to wash benthic organisms downstream as surface substrates are scoured,
and droughts and toxic pollutant plumes kill surface-dwelling animals (Hynes, et al. 1974; Williams,'
1987). The rapidity with which certain taxa recolonize these denuded substrates has been shown to be
due, at least in part, to vertical migration from the hyporheic zone (Dole-Olivier, et al., 1997). Again,
the discovery of diapausing nymphs of the cool water-adapted winter stonefly Allocapnia vivipara in
the hyporheic zone during the summer warm-water phase of temperate streams is further evidence of a
refugium (Harper and Hynes, 1970).
Table 1. The two primary components of the hyporheos (after Williams and Hynes 1974).
(1) Species derived from hypogean environments such as ground water, subterranean water bodies, and
waterlogged soil. These have been dubbed "permanent" members of the hyporheos as they complete
their entire life cycles in the interstices. The permanent hyporheos includes rotifers, nematode worms,
oligochaetes, mites, copepods, ostracods, cladocerans, tardigrades, and syncarid and peracarid
crustaceans.
(2) Species derived from the streambed benthos—particularly the early-instar larvae of aquatic insects.
These spend only part of their life cycles in the hyporheic zone, having to return to the stream surface
in order to metamorphose into a terrestrial, adult stage. These have been dubbed "occasional" members
of the hyporheos, although "transient" members may be a better term.
While the hyporheic zone is a fascinating system for the furthering of purely academic enquiry, it
also is emerging as an important site for the transformation and storage of nutrients (Triska, et al.,
1994). For example, nitrification, a major chemolithotrophic process, occurs in the hyporheic zone,
converting the predominant form of inorganic nitrogen in incoming waters from ammonium to nitrate.
Although the amount and rate of production of biomass contributed to the lotic food web by nitrifying
bacteria are typically lower than those generated by heterotrophs, in streams receiving high levels of
nitrogen from riparian agriculture production through nitrification could be quite significant. Similarly,
bacterial alkaline phosphatase activity is known to occur in the hyporheic zone, and release of
phosphorus from organic P may supply this important nutrient to surface (benthic) and hyporheic biota.
In addition, there is some evidence that lithological and geochemical processes in the hyporheic
zone may mediate the availability of N and P (Storey, et al., 1999). For example, substrate particles
that have a high cation exchange capacity, as a consequence of their chemical composition and size,
will tend to sorb inorganic P and ammonium. In the latter case, hyporheic sediments have the capacity
to function as a transient storage pool for dissolved inorganic nitrogen. In these respects, the hyporheic
40
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
zone should be of interest to water managers and conservationists, as custodians of national water
resources.
Hyporheic sampling techniques roughly fall into four categories (Table 2). Unfortunately, virtually
all of these samplers have limitations. For example, well digging cannot be used in mid-stream and is
not very quantitative; freeze cores may drive organisms away as they form; mechanical corers may
have depth or substrate particle size limitations; and artificial substrates may fail to re-estabh'sh natural
sediment profiles and/or detrital components. Further, many of these samplers have neither been
evaluated in more than one location, nor evaluated against each other.
Table 2. The four main categories of hyporheic samplers.
(1) digging of small wells in the exposed (above water) areas of Karaman-Chappuis technique,
gravel bars and stream margins to reach the water table, and then see Schwocrbel (1970)
straining the interstitial water so exposed through a fine-mesh Sassuchin (1930)
net;
(2) freeze cores that use chemicals such as liquid nitrogen,
liquid carbon dioxide, or a mixture of "dry ice" (crushed solid
carbon dioxide) and acetone or alcohol to freeze the substrate
around a standpipe driven into the bed;
Efford (1960)
Stacker and Williams (1972)
DanielopOl, et al. (1980)
Bretschko and Klemens (1986)
(3) mechanical corers that, when driven into the bed, either Bou and Rouch (1967)
isolate a sample of the surrounding substratum and its fauna for Husmann (1971)
subsequent removal, or suck up interstitial water and organisms Mundie (1971)
from a desired depth; Williams and Hynes (1974)
(4) artificial substrate samplers that involve placing a sterilized Moon (1935)
portion of natural stream bed into perforated containers that are Coleman and Hynes (1970)
sunk into the bed and then removed after a desired period of Hynes (1974)
colonization. Panek(1991)
Fraser. etal. (19961
Recently, we compared the field performance of four hyporheic samplers at a single riffle on the
Speed River, Ontario (Fraser and Williams, 1997). These samplers were: the standpipe corer, the
freeze corer, a pump sampler, and the colonization corer. Each sampler was assessed, at different
sediment depths, for accuracy and precision in terms of total invertebrate density, taxon richness, and
invertebrate size distribution.
Since previous studies have concluded that the standpipe corer and the freeze corer, following in
situ electropositioning, provide good estimates of hyporheic density (Williams,
1981; Bretschko and Klemens, 1986), the a priori assumption was made to accept their data
as the accuracy standard. Sampler precision was calculated as the coefficient of variation (CV), which
is the standard deviation expressed as the percentage of the mean.
In terms of faunal density (Figure 2), the colonization corer estimates were significantly less than
those obtained by the other three corers all of which produced very similar results.
In terms of overall taxon richness, there were no detectable differences among the samplers.
Further, all of the samplers captured individuals representing greater than 90% of the available taxon
41
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
pool. For example, all of the samplers captured nematodes, molluscs, ostracods, copepods, mites,
mayflies, stoneflies, caddisflies, beetles, and dipterans. However, tardigrades were captured only by
the freeze and pump samplers; cladocerans were not captured by the freeze corer; and amphipods were
not captured by the pump sampler. In terms of the percent insect larvae captured (another measure of
taxon bias), the pump sampler collected the fewest, although this was significantly so only at a depth
of20 cm.
In terms of invertebrate size, as measured by chironomid larval length, there was a decrease with
increasing depth for all of the samplers. The only difference detected among the samplers was that, at
20 cm, the pump sampler captured slightly smaller larvae than the other three.
For all four samplers tested and all of the measures compared (density, richness, and size), the
level of precision was generally between 20 and 40%, but increased with depth. No sampler yielded a
consistently higher level of precision than any other.
The conclusions that may be drawn from this comparative study are:
(1) All four samplers would suffice for collecting purely qualitative data.
(2) In terms of removing an exact, representative portion of habitat (to obtain absolute measures),
only the freeze corer qualified. However, and in support of the a priori assumption, no
statistical differences were detected between this sampler and the standpipe corer for any of
the measured variables, at any depth.
(3) The colonization corer consistently underestimated total invertebrate density.
(4) The pump sampler was capture selective both in terms of invertebrate type and size - the bias
towards non-insects and smaller insects probably reflecting a filtering effect of the interstices.
As to recommendations for possible standardization of hyporheic sampling are concerned,
pragmatically the goals should determine the means. Some examples are given in Table 3. Regrettably,
the holy grail of a perfect hyporheic sampler still seems to evade us and, indeed, may never be
attainable. Nevertheless, samplers do exist that allow acceptable levels of sediment description, water
sampling, and faunal characterization to be made—although perhaps not through one apparatus alone.
Such techniques have the potential, either singly or in combination, to help researchers answer some of
the sophisticated questions that 30 years of hyporheic study is now demanding.
Table 3. Examples of hyporheic samplers suited to specific information goals
(1) If survey information is required, relatively quickly, on invertebrate densities and types at a
variety of depths, then the standpipe corer would be suitable. This corer has been shown to produce a
mean error density estimate of around 19%, and captures virtually.all of the common taxa found in the
hyporheic zone (Williams, 1981). Both Cummins (1975) and Elliott (1977) have suggested that this
level of accuracy is acceptable in estimating benthic densities, and so perhaps the same should be
applied to the hyporheos.
(2) If a larger sample volume, together with a description of invertebrates and the undisturbed
sediments in which they live, is required, then the freeze corer (preceded by electropositioning) would
be the choice.
(3) If periodic assessment (with moderate precision) of the hyporheos is required from a particular
site, with minimal, long-term habitat disturbance, then the colonization corer would be appropriate -
especially if routine hydrogeological and chemical data are needed also.
42
-------
Proceedings of the Ground-Water/Surface- Water Interactions Workshop
July 2000
(4) The colonization corer also would be the most suited to manipulative studies of hyporheic
dynamics—as it allows different combinations of hyporheic sediments (e.g., particle size and/or
organic content) to be presented for colonization.
ACKNOWLEDGMENTS
I am grateful to Brian Fraser and Richard Storey for allowing me to draw on some of their ideas
and data from their respective University of Toronto theses. Their work was supported by grants from
the Natural Sciences and Engineering Research Council of Canada Research. I would also like to thank
Brace Duncan, U.S. EPA Region 10, for inviting me to attend the workshop.
REFERENCES
Bretschko, G. and W.E. Klemens, 1986. Quantitative methods and aspects in the study of the
interstitial fauna of running waters, Stygologia , v. 2, pp. 297-316.
Bou, C. and R. Rouch,1967. Un nouveau champ de recherches sur la faune aquatique souterraine.
Compt. rend. hebd. s&inces de 1'Acad. Sci., v. 265, pp. 369-370.
Coleman, MJ. and H.B.N. Hynes, 1970. The vertical distribution of the invertebrate fauna in the bed
of a stream, Limnol. Oceanogr., v. 15, pp. 31-40.
Danielopol, D.L., Ginner, R. and H. Waidbacher, 1980. Some comments on the freezing core method
of Stacker and Williams (1972), Stygo News, v. 3, pp. 4-5.
Dole-Olivier, M.-J., Marmonier, P., and J.-L. Beffy, 1997. Response of invertebrates to lotic
disturbance: is the hyporheic zone a patchy refugium?, Freshwat. Biol., v. 37, pp. 257-276.
Efford, I.E., 1960. A method of studying the vertical distribution of the bottom fauna in shallow
waters, Hydrobiologia, v. 16, pp. 288-292.
Fraser, E.G. and D.D. Williams, 1997. Accuracy and precision in sampling hyporheic fauna, Can. J.
Fish. Aquat. Sci., v. 54, pp. 1135-1141.
Fraser, B.G., Williams, D.D. and Howard, K.W.F., 1996. Monitoring biotic and abiotic processes
across the hyporheic/groundwater interface, Hydrogeol. J., v. 4, pp. 36-50.
Harper, P.P. and H.B.N. Hynes, 1970. Diapause in the nymphs of Canadian winter stoneflies, Ecology,
v. 51, pp. 925-927.
Husmann, S., 1971. Eine neue Methode zur Entnahme von Interstitialwasser aus subaquatischen
Lockergesteinen. Arch. Hydrobiol., v. 68, pp. 519-527.
Hynes, H.B.N., 1974. Further studies on the distribution of stream animals within the substratum,
Limnol. Oceanogr., v. 19, pp. 92-99.
Hynes, H.B.N., Kaushik, N.K., Lock, M.A., Lush, D.L., Stacker, Z.S.J., Wallace, R.R., and D.D.
Williams, 1974. Benthos and allochthonous organic matter in streams, J. Fish. Res. Board Can., v.
31, pp. 545-553.
Moon, H.P., 1935. Methods and apparatus suitable for an investigation of the littoral region of
oligotrophic lakes, Int. Revue ges., Hydrobiol., v. 32, pp. 319-333.
Mundie, J.H., 1971. Sampling benthos and substrate materials, down to 50 microns in size, in shallow
streams, /. Fish. Res. Board Can., v. 28, pp. 849-860.
Panek, K.L.J., 1991. Migrations of the macrozoobenthos within the bedsediments of a gravel stream
(Ritrodat-Lunz study area, Austria), Verh. Int. Ver. Theor. Angew. Limnol., v. 24, pp. 1944-1947.
Sassuchin, D.N. 1930. Materialien zur Frage uber die Organismen des Flugsandes in der
Kargisensteppe, Russ. Hydrobiol. Z. saratow v. 9, pp. 121-130.
Schwoerbel, J. 1970. Methods of Hvdrobiology. Pergamon Press, Oxford, 200 pp.
Stacker, Z.S.J. and D.D. Williams, 1972. A freezing core method for describing the vertical
distribution of sediments in a stream bed, Limnol. Oceanogr., v. 17, pp. 136-138.
43
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Storey, R.G., Fulthorpe, R.R. and D.D. Williams, 1999. Perspectives and predictions on the microbial
ecology of the hyporheic zone. Freshwat. BioL, v. 40.
Triska, F.J., Jackman, A.P., Duff, J.H. and Avanzino, RJ. 1994. Ammonium sorption to channel and
riparian sediments. A transient storage pool for dissolved inorganic nitrogen. Biogeochem v 26
pp. 67-83.
Williams, D.D., 1981. Evaluation of a standpipe corer for sampling aquatic interstitial bio topes.
Hydrobiologia, v. 83, pp. 257-260.
Williams, D.D., 1987. The Ecology of Temporary Waters. Timber Press, Portland, Oregon, 205 pp.
Williams, D.D., 1993. Nutrient and flow vector dynamics at the hyporheic/groundwater interface and
their effects on the interstitial fauna. Hydrobiologia, v. 251, pp. 185-198.
Williams, D.D. and H.B.N Hynes, 1974. The occurrence of benthos deep in the substratum of a
stream. Freshwat. BioL, v. 4, pp. 233-256.
Williams, D.D. and H.B.N. Hynes, 1977. The ecology of temporary streams II. General remarks on
temporary streams. Int. Revue ges, Hydrobiol, v. 62, pp. 53-61.
AUTHOR INFORMATION
D. Dudley Williams, Surface and Groundwater Ecology Research Group, Division of Life Sciences,
University of Toronto at Scarborough, 1265 Military Trail, Scarborough, Ontario, Canada MIC
1A4
44
-------
DISCUSSION GROUP SUMMARIES
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Hydrogeology Discussion Group Summary
By Thomas C. Winter and Joseph Dlugosz
INTRODUCTION
^ great variety of sediment types in the beds of most surface water bodies results in substantial
variability in the location and rates of seepage across the bed. The exchange of water between ground
water and surface water ranges from slow, diffuse seepage to rapid, concentrated flow at specific
localities. Determining the location, rate, volume, and chemistry of water moving between these two
components of the hydrologic system is difficult, expensive, and highly uncertain. Nevertheless, the
need for understanding the hydrologic processes and measuring the interaction of water and dissolved .
chemicals between ground water and surface water is fundamental to environmental management. To
address these challenges and needs, the hydrogeology discussion group focused on the hydrogeologic
aspects of understanding and measuring the interaction of water and dissolved chemicals between
ground water and surface water at sites where ground water has been contaminated.
To focus the discussion on the interface between ground water and surface water, the group made
several presumptions: (1) the hydrogeologic framework of a site has been defined; (2) the source area
of the contaminant is known; (3) the flow pathways and plume configuration are reasonably well
defined; (4) the chemical characteristics and decomposition products of the contaminants are known;
and (5) the contaminant is a potential threat to the environment. Given this information, it was
suggested that the actual determination of the movement of ground water to the surface water body
could be accomplished through a tiered approach: A sequence of actions could be followed that begins
with a general reconnaissance of observable indicators of ground-water discharge and evolves to very
detailed and focused sampling of hydraulic head, chemistry, and biology.
This summary of the discussion group presents:
(1) Field methods that can be used for (a) reconnaissance of observable qualitative indicators of
ground-water discharge to surface water, (b) direct measurement and calculated flow of water
between ground water and surface water using physical data, and (c) indicators of flow
between ground water and surface water using chemical data;
(2) Considerations for temporal sampling of water flow and chemistry; and
(3) Variations of field sampling strategies that may be needed' in different hydrologic landscapes.
The material presented here is considered to be a supplement to another EPA report (U.S. EPA,
1991) that presented a review of methods for assessing non-point source contributions of contaminants
to surface water. Some of the information presented briefly in this summary is discussed in much more
detail in the EPA report.
46
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
FIELD METHODS FOR DETERMINING THE INTERACTION OF GROUND WATER AND SURFACE
WATER
Observable Qualitative Indicators of Ground-Water Discharge to Surface Water
Many indicators of ground-water discharge to surface water can be used to determine specific
localities where a contaminant plume may be entering a surface-water body. The most common
indicators are seeps and springs; infrared mapping; aquatic plants; phreatophytes; unique sediment
zones such as mineral precipitates; water color; odor from contaminants; and mapping of lineaments in
fractured-rock settings. It was suggested that a field reconnaissance of these easily observable
characteristics would identify specific localities where detailed measurements and sampling could be
focused. If the skills of biologists are available, benthic organisms also can be useful indicators of
ground-water discharge.
Observation of seeps and springs is relatively straightforward if the flow rates are high. In
fractured-rock landscapes, mapping of lineaments can be useful if the fractures are open. Ground-
water flow concentrated in the fractures enter surface-water bodies as springs. In settings where
seepage rates are low, it is easier to observe seeps during colder times of year when ground water and
air temperatures are considerably different, because the water vapor above seeps is visible. Further-
more, in climates where surface water freezes or snow is on the ground, areas of appreciable ground-
water inflow remain open. The difference in temperature between ground water and surface water also
makes infrared mapping a useful reconnaissance tool, especially in mid-summer when the difference in
temperatures of ground water and surface water are at a maximum.
Some chemical constituents dissolved in anoxic ground water precipitate upon contacting
oxygenated surface water. For example, iron and manganese oxides are common indicators of seep
areas. Contaminated ground water commonly has color and odor. Water color and odor from
contaminants can be used as an indicator of ground-water inflow, especially if the inflow consists of
the contaminated water.
Aquatic plants can be indicators of ground-water discharge. The following are a few examples: (1)
Swanson, et al. (1984) indicated that cattails are indicators of fresh ground-water input to saline prairie
lakes in North Dakota, (2) Rosenberry, et al. (in review) indicated that Marsh Marigold was an
indicator of springs in Minnesota, (3) Lodge, et al. (1989) indicated that submerged aquatic plant
biomass was greater where ground-water inflow velocity was greater, and (4) Klijn and Witte (1999)
discussed the relationship of plants to ground-water flow systems. In addition to aquatic plants, upland
phreatophytic plants near a surface-water body are indicators of the presence of ground water at
shallow depths.
Benthic organisms can be indicators of ground-water discharge to surface water. Numerous
examples of the relationship of organisms to water flow and chemistry are provided by studies of the
hyporheic zone beneath streams. With respect to lakes and wetlands as well as streams, ostracods are
especially useful because they have specific tolerances to water temperature and chemistry. An
additional benefit to using ostracods is that some of the chemical constituents and isotopes that are
present in the water while the organisms are alive are incorporated into their shells. Therefore, study of
ostracod shells in sediments can provide a valuable record of past ground water and surface water
relationships.
47
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Direct Measurement and Calculated Flow of Water Between Ground Water and Surface Water
Using Physical Data
The reconnaissance methods discussed above may be useful for identifying locations of ground-
water inflow to surface water, but they do not indicate the quantities of water that move across the
interface. Measurement of water quantity can be done by (1) using instruments that directly measure
the water flux, or a physical or chemical property from which flux can be calculated, at the specific
locality of the instrument (herein referred to as direct measurements); or (2) calculating the flux over a
broader area of surface-water bed using streamflow data or ground-water flow nets. A drawback of
direct measurements is that they sample a point in space, and, because of the great variation in
sediment types in most surface water beds, measurements need to be taken at many places in the bed.
Furthermore, most measurements are taken at a point in time because the devices generally are not
equipped with recorders. For these reasons, it also is desirable to calculate the flux through broader
areas of surface-water beds to obtain independent estimates of flux. This approach averages out the
spatial variability of flux and it provides a check on values determined by direct measurements.
Direct measurements: Methods for directly measuring the flux of water between ground water and
surface water include the use of seepage meters, mini-piezometers, temperature profiles in the
sediments, heat-flow meters, hydraulic properties of sediments determined from cores, and direct-
contact resistivity probes. Although these were considered by the discussion group to be methods for
direct measurements, only seepage meters can be used for direct measurements of water flux. The
other methods use devices that make direct measurements of hydraulic head, hydraulic conductivity,
temperature, or electrical conductance, and the water flux then needs to be calculated from these data.
Seepage meters are chambers (commonly, cut-off 55-gallon drums) that are set on the bed of a
surface water body (Lee, 1977). After the chamber is pushed into and allowed to settle into the
sediment, a tube is inserted into an opening in the top or side of the chamber. The tube has a small bag
attached at the end and a valve positioned between the chamber and the bag. The bag can be attached
empty if ground water is known to be seeping in, or filled with a known volume of water if the
direction of seepage is unknown or if it is known that surface water is seeping out. To measure the
flux, the valve is opened and the change in water volume in the bag over a given period of time is a
measure of flux per that period of time. Seepage meters are perhaps the most commonly used devices
for measuring water flux between ground water and surface water, and different sizes and types of
chambers other than 55-gallon drums have been used. A number of studies have evaluated the
uncertainties in using the seepage-meter method for determining flux through surface-water beds
(Shaw and Prepas, 1990; Belanger and Montgomery, 1992). Seepage meters have been used largely to
make discrete measurements at a point in time, but a recording seepage meter was developed recently
by Paulsen, et al. (unpublished manuscript) using ultrasonic flow technology.
Mini-piezometers are used to determine the hydraulic gradient between a surface-water body and
the ground water beneath it. A small diameter well is inserted into the surface-water bed, and, in the
most common design, a flexible tubing is attached from the well to a manometer board. Another piece
of tubing is attached to the other side of the manometer and the other end is placed in the surface
water. Both ground water and surface water are drawn into the manometer using a hand pump. After
air is bled back into the manometer and the water levels in each tube stabilized, the difference in head
can be measured directly (Lee and Cherry, 1978; Winter, et al., 1988). The difference in head between
ground water and surface water can also be determined simply by measuring the level of ground water
in the well and the level of surface water outside the well. Mini-piezometers provide data only on
hydraulic gradient. To determine water flux, hydraulic conductivity of the sediments need to be
determined as well as the cross-sectional area of the flux.
48
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
The transport of heat by flowing water has been used to determine the interaction of ground water
and surface water. By measuring the temperature of surface water and the temperature at shallow
depths in sediments, Silliman and Booth (1993) mapped gaining and losing reaches of a stream in
Indiana. Sediment temperatures had little diurnal variability in areas of ground-water inflow because of
the stability of ground-water temperatures. Sediment temperatures had much more variability in areas
of surface water flow to ground water because they reflected the large diurnal variability of the surface
water. This approach is useful for determining flow direction. Lapham (1989) used sediment-
temperature data to determine flow rates and hydraulic conductivity of the sediments based on
fundamental properties of heat transport. Heat-flow meters, consisting of a heating element and a ring
of temperature sensors, placed at a distance from the heater, have been used to measure the rate and
direction of water movement through sediments. A pulse of heat is applied to a heating device and the
rate and direction of water movement is determined by measuring the time it takes for the heat pulse to
be sensed by the thermistors in the direction of flow.
Hydraulic properties of sediments can be determined by laboratory studies of sediment cores.
These data can then be used to calculate ground-water flux if the hydraulic gradient and area of
surface-water bed through which the water flux is taking place is known. Probes that measure
electrical resistivity have been used to locate contaminant plumes entering surface water. These probes
are most effective if the conductance of the contaminant is substantially different than the conductance
of the ambient ground water.
Calculated from stream/low data and flow nets: The quantity of water moving between ground
water and surface water over scales larger than can be determined by direct measurement using
individual sensors generally is determined by stream discharge data or by ground-water flow nets. The
most direct method for determining ground-water inflow or stream losses to ground water is to make
stream discharge measurements at different locations along a stream. The difference in discharge
between two localities is the quantity of gain or loss of water for the reach of stream between the
measurement sites. The accuracy of the values is related almost entirely to the accuracy of the
discharge measurements.
The flow-net approach is probably the most common method used for determining the interaction
of ground water and surface water. The term flow net is used broadly herein as any calculation of
ground-water flux, including simulation models, that makes use of a network of wells for determining
hydraulic gradients, estimates of hydraulic conductivity of the geologic units and sediments, and cross-
sectional area of the interface of ground water and surface water. The accuracy of the values is related
to the quantity and quality of the hydrogeologic data, and the grid spacing that is justified by these
data.
Indicators of Flow Between Ground Water and Surface Water Using Chemical Data
The chemistries of ground water and surface water commonly are different enough—especially at
contaminated sites— that some chemical constituents or isotopic properties of water can be used to
determine the interaction of ground water and surface water. Devices for collection of water samples
for determination of the chemical characteristics of water passing through sediments consist of two
basic types: (1) collection at the sediment-water interface; and (2) collection at various depths in the
sediment by inserting a device into the sediments.
Constituents: Nearly all chemical constituents have the potential to be useful in determining the
contribution of ground water to surface water. By calculating mass balances of the constituents, the
flux of water can be quantified. Isotopes of some elements, such as nitrogen and radon, are particularly
49
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
useful because in some cases a specific contaminant source can be identified. Isotopes of water are
among the most useful because they are part of the water molecule itself and are not subject to
modification by chemical reactions. The age of ground water can be determined by analyzing for
tritium and chlorofluorocarbons, which are useful for identifying ground-water flow paths.
Sampling at the sediment-water interface: Devices that have been developed for sampling water at
the sediment-water interface include drag probes, seepage meters, diffusion bags, bubble collectors,
and biosensors. Of these devices, seepage meters are the only ones that actually collect a water sample
large enough to be analyzed in the laboratory for many constituents. Furthermore, by knowing the
water flux, the flux of a constituent or isotope can be calculated. Drag probes, such as used for
measurement of temperature, specific conductance, and radioactivity, are used primarily to locate areas
of inflow. Vapor diffusion samplers are placed in the sediments and can collect certain contaminants
that diffuse into the bag, and they also can measure microbiological activity through the production of
hydrogen. Devices that collect gas bubbles are used to determine the chemical constituents in the
bubbles, which are an indication of the gases being produced in the sediments.
Sampling at depth in sediments: Devices that have been developed for sampling or measuring
water chemistry at depth in sediments consist of (1) multi-level samplers that are driven into the
sediments; and (2) probes through which individual samples can be drawn from any depth—or a
constituent measured—but can then be driven deeper to collect samples at other specific depths.
Examples of the first are pore-water peepers, gel samplers, and multi-level samplers. Pore-water
peepers are blocks of plastic that have chambers machined into them at specified intervals (Hesslein,
1976). A porous membrane is placed over the chambers and held in place by another cover of plastic
that has holes machined at the same intervals. The chambers are filled with deionized water, and the
device is driven into the sediments. The device is left in place for a period of time for the chemicals to
diffuse across the membrane and equilibrate with the ambient pore water (usually weeks). The device
is then removed and the water in the chamber is extracted and analyzed. Gel samplers are similar, but
the collection device is a thin film of polyacrylamide gel that is placed on a flat Perspex probe, covered
with a'porous membrane, and held in place by a thin plate that has a window cut the full length of the
probe (Krom, et al., 1994). The device also is driven into the sediments and left to equilibrate (usually
only minutes to a day). After equilibration, the device is removed and the gel sectioned at any desired
interval to obtain the samples.
Multi-level samplers are rigid tubes that have ports machined into them at specified intervals.
Flexible tubing is attached to each port and brought to the surface. Water samples can then be drawn
from individual ports using a pump at the surface. Squeezing or centrifuging pore water from segments
of sediment cores can also be considered multi-level sampling. Of these methods, only rigid-tube
multi-level samplers can be used for repeated sampling of precisely the same location and depth
because the device can be left in place. ' "
^ Examples of probes through which water samples can be withdrawn, or a measurement made, from
a single depth and then pushed deeper to collect other individual samples include mini-piezometers and
Geoprobes.
CONSIDERATIONS FOR TEMPORAL SAMPLING OF WATER FLOW AND CHEMISTRY
The time interval for sampling water flow and chemistry depends on the phase of the program,
physical characteristics or chemical constituents of interest, climatic setting, and hydrogeologic setting.
In general, more sampling is needed in the initial phases of a program when the extent of a problem is
being determined, and less sampling is needed for long-term monitoring. For example, it commonly is
50
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
desirable to continuously monitor water flow and hydraulic head in the initial phases of a study to
characterize the variability on daily, seasonal, and annual scales. At the same time, sampling for
chemical constituents also needs to be done more frequently at this time to relate the concentrations
and mass transport of constituents to flow regime and to climate. Once the relationship of mass
transport to flow and climate is reasonably well understood, the frequency of sampling can be reduced.
Hydrogeologic setting comes into play in sampling frequency because some settings are inherently
more simple, thus easier to characterize and monitor flow and chemistry, than others. Similarly, the
climate that drives the hydrologic system is much less variable, thus easier to characterize and monitor,
in some regions than in others. If a sampling program includes biological factors, sampling frequency
may need to include considerations related to the life cycles of the organisms.
An important climate consideration in both initial site characterization and long-term monitoring is
the effect of extreme climatic events. Extreme climatic events, such as droughts and deluges of
precipitation, can have a greater effect on a site than many years of more normal conditions. These
effects include rearrangement of bed sediments, changes in water flow paths, mass-transport of
chemicals, and biological conditions of a surface water bed. One catastrophic event can greatly alter
the perception of how well a hydrologic system is understood, and how it should be managed or
mitigated. Although difficult to anticipate, a plan for sampling during catastrophic events should be in
place.
VARIATIONS IN FIELD MONITORING AND SAMPLING STRATEGIES FOR DIFFERENT
HYDROLOGIC LANDSCAPES
A generic field design for determining the interaction of surface water with ground water includes
the use of piezometer nests, water-table wells, and devices to measure or calculate the flow of water
and chemicals across the surface-water bed. The conceptual model in Figure 1 of the Executive
Summary shows ground-water seepage inflow on one side of the surface-water body and surface-water
seepage out on the other. Actual conditions could be as indicated, have ground-water inflow on both
sides, or have surface-water seepage out on both sides. The important point of the diagram is to stress
that the interaction of ground water and surface water can be reasonably well understood only by
addressing the larger-scale processes related to the position of the surface-water body within ground-
water flow systems as well as the smaller-scale processes related to geology of the surface-water bed
and climate.
The advantage of having permanent installations, such as wells and piezometers, in the upland is
that they can be easily equipped to obtain continuous records. The disadvantage of having these
installations is that they do not indicate the precise location or chemistry of seepage across the
sediment-water interface. The advantage of the devices used within the surface-water body is that they
can be used to pinpoint the location, rates, and chemistry of seepage water. The disadvantage of using
these devices is that few can be used to obtain continuous records. Furthermore, few devices used
within the surface-water body can be left in place for long periods of time because of floods, currents,
ice, and water safety.
Although the generic field design may be applicable to many actual field settings, it is conceivable
that the design would need to be altered somewhat for different hydrologic landscapes. For example,
some landscapes, such as riverine and coastal, have wetlands at the base of terraces in the uplands. If a
source of ground-water contamination was located on the terrace, the contaminant plume could
conceivably discharge to the wetlands at the base of the terrace. In this case it would be desirable to
place an additional piezometer nest in the wetland. Other modifications to the field design might be
51
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
related to the geologic complexity of the site. For example, if the geologic framework has a series of
aquifers and aquitards or lateral geologic discontinuities, it might be necessary to place piezometers in
the different geologic units in order to better understand the ground-water flow paths.
Frequency of sampling for chemical constituents also would depend on hydrologic setting. For
example, in coastal areas affected by tides, the water flow and chemical transport paths could be
greatly affected by the tidal exchange and storm surges. In northern and mountainous areas, runoff and
ground-water recharge from snow melt can have a substantial effect on ground-water flow paths and
chemical transport.
Because of the variety of hydrologic landscapes and the variability of climate, a need exists for
development of type localities that would become benchmarks for the various landscape types. At
these type localities, design of field installations, effectiveness of various sensors and devices,
sampling frequency, and study and site characterization approaches could be tested and evaluated.
Such knowledge could lead to efficient and cost effective approaches to dealing with contaminated
sites in the hydrologic landscapes represented by a given type locality.
REFERENCES
Belanger T.V. and M.T. Montgomery, 1992, Seepage meter errors: Limnology and Oceanography v
37, pp. 1787-1795.
Hesslein, R.H., 1976, An in situ sampler for close interval pore water studies: Limnology and
Oceanography, v. 21, pp. 912-914.
Klijn, F. and J-P. Witte, 1999. Eco-hydrology: groundwater flow and site factors in plant ecology,
Hydrogeology Journal, v. 7, pp. 65-77.
Krom, M.D., Davison, P., Zhang, H., and W. Davison, 1994, High-resolution pore-water sampling
with a gel sampler, Limnology and Oceanography, v. 39, no. 8, pp. 1967-1972.
Lapham, W.W., 1989. Use of temperature profiles beneath streams to determine rates of vertical
ground-water flow and vertical hydraulic conductivity, U.S. Geological Survey Water-Supply
Paper 2337, 35 pp.
Lee, D.R., 1977. A device for measuring seepage flux in lakes and estuaries, Limnology and
Oceanography, v. 22, p. 155-163.
Lee, D.R. and J.A. Cherry, 1978, A field exercise on groundwater flow using seepage meters and mini-
piezometers, Journal of Geological Education, v. 27, p. 6-10.
Lodge, D.M., Krabbenhoft, D.P., and R.G. Striegl, 1989, A positive relationship between groundwater
velocity and submersed macrophyte biomass in Sparkling Lake, Wisconsin, Limnology and
Oceanography, v. 34, pp. 235-239.
Paulsen, R.J., Smith, C.F., and T-f. Wong. Development and evaluation of an ultrasonic groundwater
seepage meter, unpublished manuscript. •
Rosenberry, D.O., Striegl, R.G., andD.C. Hudson. Plants as indicators of rapid ground-water
discharge to a northern Minnesota lake, unpublished manuscript.
Shaw, R.D. and Prepas, E.E., 1990. Groundwater-lake interactions: I. Accuracy of seepage meter
estimates of lake seepage, Journal of Hydrology, v. 119, pp. 105-120.
Silliman, S.E., and D.F. Booth, 1993. Analysis of time series measurements of sediment temperature
for identification of gaining versus losing portions of Juday Creek, Indiana, Journal of Hydrology,
v. 146, pp. 131-148.
Swanson, G.A., Adomaitis, V.A., Lee, F.B., Serie, J.R., and J.A. Shoesmith, 1984. Limnological
conditions influencing duckling use of saline lakes in south-central North Dakota, Journal of
wildlife Management, v. 48, pp. 340-349.
52
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
U.S. EPA, 1991. A review of methods for assessing nonpoint source contaminated ground-water
discharge to surface water; U.S. Environmental Protection Agency Report EPA 570/9-91-010, 99
pp.
Winter, T.C., LaBaugh, J.W., and D.O. Rosenberry, 1988. The design and use of a hydraulic
potentiomanometer for direct measurement of differences in hydraulic head between ground water
and surface water, Limnology and Oceanography, v. 33, pp. 1209-1214.
53
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Chemistry Discussion Group Summary
By Allen Burton and Ned Black
INTRODUCTION
The chemistry discussion group agreed to adopt the broad term ground-water/surface-water
transition zone," unless it was specifically addressing the classical stream hyporheic zone. In this
summary, individual topics that were discussed frequently over the course of the day are summarized
under single headings. The group's discussions sometimes veered into issues belonging to the biology
discussion group, such as the importance of establishing clear reasons for adding the transition zone
habitat to the risk assessments performed at contaminated sites. Some group members expressed
concern that project managers should establish the justification for sampling a transition zone site (e.g.,
complete pathways to receptors) prior to extensive use of the sampling and analytic techniques we
discussed.
An obvious—but important—point to remember is that the contaminants in question are the same
ones (e.g., dissolved or NAPL chlorinated solvents and petroleum hydrocarbons, pesticides, dissolved
or particle-bound metals) that we encounter in contaminated ground waters and surface waters. Thus,
we need to collect information on the same parameters we use to predict the geochemical fate of these
contaminants in both ground water and surface-water bodies. We also need to collect the chemical and
physical information commonly used in ecological risk assessments and natural attenuation assess-
ments to determine the dominant biological processes and the potential confounding factors in
bioassays. Finally, we need to collect chemical information which helps locate zones where a ground-
water plume or hyporheic flow is entering a surface-water body. There is overlap among these
parameters, but we should remember the three different uses of chemical information:
1. Contaminant chemistry and fate
2. Biological processes
3. Identification of flow paths
The transport of dissolved contaminants from surface water into the subsurface through hyporheic
flow or ground-water recharge from a losing stream was included in our discussion of the transition
zone. With regard to flow paths and sources of contamination, the deposition of contaminated
sediments was excluded from our discussion. Other groups within EPA are addressing the issue of
contaminated sediments.
It is possible to list many chemical and physical parameters (see below) to measure in order to
satisfy the three information needs listed above. As for any ecological risk assessment, a screening
process will determine what level of site chemistry characterization should be performed. In other
words, it is not necessary to collect the same information at all sites. In order to justify extensive work
on a site, a screen must demonstrate the presence of contaminants at levels sufficient to present risk to
actual or potential receptors. For the chemistry discussion group, screening information also included
parameters for determining site geochemistry and contaminant flow paths, although collection of this
information might be deferred until after a screen.
One or more standard conceptual model should be developed to identify the important questions to
ask and data to collect at different types and scales of sites. Sampling efforts in the transition zone may
54
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
be more costly than standard sampling of surface water or shallow ground water. At the very least,
project managers and responsible parties familiar with only surface waters or only ground water will
have to be taught to use different tools.
LIST OF PARAMETERS AND TOOLS
Screening Tools
• Semi-permeable membrane devices (SPMDs)—
Widely accepted as a presence/absence screening tool. Requires extensive calibration (e.g., of
equilibration times) and sensitivity analysis to determine exact concentrations. EPA researchers, in
cooperation with other government or academic scientists, should perform sensitivity experiments
to determine if there are situations where SPMDs can be easily used to measure concentrations.
• Drag probes for temperature, conductivity, and gamma anomalies—
Useful in lakes, estuaries, and large rivers to determine zones of ground-water discharge.
• Piezometers and mini piezometers—
Multiple piezometers with low-flow sampling can provide adequate samples of transition zone
interstitial water and, of course, ground water. In order to sample just the transition zone, extreme
care is required in depth placement of the screens. Piezometers can be placed both on land and in
stream or lake beds.
• Freeze sampling techniques—
Typically used to obtain biological samples, but could also be used to sample water and substratum
for chemical analysis.
• Colonization corers—
Also a biological sampler, but can incorporate nested piezometers.
• Bead pipes (ceramic beads).
• Dye tracers of ground-water and stream flow.
• Walk river bed with a hand auger.
• During low flow, note odor and visual observations.
• Photoionization detector (PID).
• Passive diffusion samplers.
• Analyze bubbles of gas (marsh or lake setting).
• Multi-level samplers.
• Seepage meters.
• Cores (solids analysis and visual).
55
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
• Laser-induced fluorescence (LIF), qualitatively determine VOC presence, BTEX, SVOCs, dense
non-aqueous phase liquids (DNAPLs).
• Cores of trees (For instance, in a mangrove swamp. However, the contaminant may actually be
metabolized in roots so false negatives are possible.).
• Field chemistry with a HACK spectrometer (nitrate, ammonia).
• Chemetrics for sulfides.
• Differential global positioning system (GPS).
• Velocity meter.
• Tidal stage.
Post-Screening Tools
• Multi-level wells.
• Everything on screening tools list.
TIME SCALES
Hyporheic and transition zone chemical and biological processes follow several different time
scales. At a minimum, these can be described as daily cycles (e.g., temperature and river stage),
normal weather changes, invertebrate and fish life cycles, seasonal changes and long-term climatic
changes and events (such as extreme weather events). The difficulties of meshing the natural time
scales of the environment with our schedules for sampling contaminated sites are shared with risk
assessments and cleanups at all outdoor sites. Clearly, an environment such as the transition zone with
strong diurnal and seasonal controls on biology and chemistry requires multiple sampling events if we
desire great confidence that all pertinent processes are understood. And just as clearly, constraints on
sampling budgets and the desire of regulators to respond to contaminated sites with an appropriate
level of effort make limited sampling schedules the overwhelming norm. The most protective option
may be to plan our sampling to coincide with the expected worst-case time of day and season. For the
transition zone in a variety of habitats, the worst case sampling time may not be known. Thus, one of
the mandates of the Regional study areas recommended below will be to determine the worst (i.e., the
best) times to sample. For some transition zone habitats, recognized international experts will be able
to offer suggestions for sampling schedules.
SPATIAL CONSIDERATIONS
As with a ground-water plume, the spatial extent of contaminants is important information. For
sites with a contaminant plume flowing from the subsurface into a water body, the effect in the
transition zone may be limited to a discrete discharge zone. Also, the discharge zone for a contaminant
plume may occur some distance from shore. An effective way to locate a discharge zone is to sample
along a series of transects in the ground water. For a stream, it is also important to sample the bank
opposing the discharge area. It must be remembered that a ground-water plume can flow entirely under
a stream without any discharge. For classic hyporheic transport parallel to the flow of a stream,
discharge can occur anywhere in the bed. For a site with a hard substratum, the impact of the
56
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
contaminants will be in the open water column. Although contaminants so discharged are an
environmental problem, the impact on the transition zone, or the exact nature of the transition zone
itself, may be hard to define. In lakes, zones of discharge from and recharge to ground water can occur
in complex patterns.
CONCENTRATION AND FLUX
In a screening or predictive risk assessment, contaminant concentrations are used for comparisons
to toxicity benchmarks. However, the flux, or loading, of contaminants is also important information
that bears on both the impact of the contaminants on the habitat and on the physical, chemical, and
biological transformations of the contaminants at the transition zone. The flux of contaminants can
change in magnitude and direction with changes in surface water temperature and flow stage.
DETECTION LIMITS
The issue of detection limits for transition zone sampling is the same as for all other sites subject to
risk assessments. Before a sampling and analysis plan is developed, the exact values of the toxicity
benchmarks to be used for screening purposes must be determined. Otherwise, the sampling budget
may be used to collect information of no use to the risk assessors.
RECOMMENDATIONS
EPA should create a series of Regional study areas of contaminated transition zone sites, with
appropriate uncontaminated reference sites. These would be studied by EPA Regional and ORD
laboratories and academic grantees. The sites should be scaled appropriately to the typical sites for the
Region. For instance, the hyporheic chemistry, biology, and hydrology of small mountain streams
impacted by mines could be very different than a zone of chlorinated solvent-contaminated ground-
water discharge in one of the Great Lakes. Ground-water discharge and hyporheic flow in estuaries
will have the further complicating factor of tides. Sites of all sizes will be encountered by the Agency.
Members of the chemistry discussion group felt strongly that extrapolating from small streams to large
rivers and lakes is unacceptable. Also, some methods work in small streams, but not in areas of high
flow. As with any landscape approach, the species and the dominant chemical and physical processes
of the environment change with different landscapes.
57
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Biological Discussion Group Summary
By Cliff Dahm and Bruce Duncan
This session opened with the following question: "Is the hyporheic zone considered an ecological
habitat to be protected or a 'treatment opportunity' zone for restoration of contaminated ground-water
discharges to surface water?"
The group agreed early in the discussion to define the zone of interest (the ground-water/surface-
water transition zone) as the "transition zone" rather than use the term "hyporheic zone," which has a
more restricted meaning where surface waters and ground waters are actively mixing. Mixing in this
zone is very important, and in a stream, surface water moving into this zone can return back to surface
water within a short distance and be "processed" through the transition zone multiple times.
An early question raised by the participants was how the zone can be defined biologically in order
to focus on and demonstrate exposure of organisms. This requires more than a hydrological definition.
There also is a need to link the transition zone to valued resources, such as fish. If there is an impact on
the meiofaunal community, does that affect trout? This characterization of food web links, which is
needed to demonstrate risk and answer the question "who cares?," led to two important points: (1)
What are the important services that this zone performs? and (2) if these services are impaired, how
can we make that determination? Superfund managers now accept the importance of benthic
macroinvertebrates to stream ecosystems; there is not the same recognition for organisms such as
meiofauna or microbes in the transition zone.
Scale was another concern. There is a need to look at the spatial extent of impact to assess whether
the contaminant discharge results in a risk to critical habitat such that action is warranted. Some
hydrogeologists expressed frustration that they already know there is contamination in upwelling areas,
but biologists countered that: (1) we do not know what the "pristine" state should be; and (2) even if
the contamination is not cleaned up, there are other communities in other parts of the stream. So would
analysis of the transition zone really matter? One attitude was: If someone is discharging without a
permit, then they are in violation. "Who cares" is not an issue. Often, "no action" is what happens
because an adverse impact cannot be demonstrated over a realistic scale.
A concern was raised about the reluctance of managers to invest in studies of transition zones.
Given that we are not successful in getting biological measurements in ground water or surface water,
how can we convince managers to do biological measurements at the interface? How do you convince
someone that the transition zone is important when there are competing resources requiring protection?
The solution is to demonstrate the functions that occur in the transition zone and what happens when
those functions are lost.
The Guidelines for Ecological Risk Assessmentl should be used to evaluate the transition zone:
• Who is present or affected? What do stakeholders care about in the system? What are the
management goals (some are predefined such as no net loss of wetlands, or meeting Ambient
Water Quality Criteria)?
1 Risk Assessment Forum. EPA/630-R-95/002F, April 1998.171 p. http://www.epa.gov/ncea/ecorisk.html.
58
-------
Proceedings of the Ground-Water/Swface- Water Interactions Workshop
July 2000
• Identify the assessment endpoints (i.e., some biological entity or function that you care about), the
exposures, the measurements to be made, and then the effects. The ecoological risk paradigm
should cover everything and help maintain a big picture perspective.
The link between contaminated sediments and contaminated ground water in the transition zone
was another issue. How is the issue of contaminated ground water different from the issue of
contaminated sediment? The biological definition of the transition zone does not cover this; change in
chemical conditions and rates are needed as well.
There are examples where removal of contaminants occurs within the transition zone with no
removal in ground water. Ground-water wells cannot reveal the full story. More thought should be put
into field sampling of mobile contaminants. You cannot just sample sediment. For example, you might
have sand that appears very clean, but has contaminated ground water moving through it. Sediment and
water are part of a system and need to be dealt with together, not separately nor sequentially. Also,
there is a need to consider the contribution from contaminated sediments (top down) into the ground
water. Sources need to be distinguished because of the polluter's perspective.
During the presentation session on the first day of the workshop, the following questions
predominated:
• Why should we be interested in biology?
• Why should the public care or be interested?
• What are the services and processes that the transition zone provides?
• Why is the transition zone important ecologically?
• What biogeochemical measures would be ideal?
Participants were interested in contaminant migration and fate; others were interested in the effects
on biological resources (macrobiota, communities, microbial processes) in the transition zone. When
considering applicable biological measures, the biological discussion group had difficulty identifying
microbial measurements with broad applicability. There is good success with macroinvertebrate
indicators, less so with microbiota and meiofauna. A multidisciplinary approach is needed to provide
synergy.
The discussion followed three aspects of the transition zone: (1) Why is the zone important
ecologically? (2) What are the methods that can be used to assess ecological structure and function?
and (3) What research is needed to better determine the ecological importance of this transition zone
and to develop needed tools for sampling this zone?
WHY IS THE TRANSITION ZONE IMPORTANT ECOLOGICALLY? WHAT ECOLOGICAL SERVICES
ARE PROVIDED?
These issues led to additional questions: Do all transition zones need to be protected, especially if
you see no impact to the surface water? Is there intrinsic value to the transition zone itself, apart from
the surface water? Historically, people study "ecological entities." The recent trend is to look at
transition zones or ecotones. We do not know much about ecotones as an ecosystem entities. The
hyporheic zone is one important ecotone. Some surface organisms have a phase in the hyporheic zone,
which implies that productivity could be affected. The hyporheic zone also serves as a "nursery" for
secondary producers. Less is known of the permanent hyporheic zone species—they often can be
distinct, undescribed species.
59
-------
Proceedings of the Ground- Water/Surface- Water Interactions Workshop
July 2000
The group discussed the importance of transition-zone function and compared it to wetlands.
Regulations require restoration of wetlands if they are destroyed. This concept could be applied to the
transition zone; the goal could be to restore function rather than restore appearance (no net loss). It was
pointed out that we need both function and structure (species).
Another question was "why should the public care about the important function of microbiology?"
or "what would the environmental effect be from the loss of that function?" Several structural and
functional elements are extremely important in this system. Transition zones often provide high quality
habitat and are sites of contaminant reduction and nutrient and carbon cycling. A good example was
made for fish. Three major biological services are tied to fish: refugia, food sources, and reproductive
zones. The links from microbes to macrobiota to fish are essential to the aquatic food web. Trout are
known to seek out transition zones. When a river is contaminated, refugia can sustain the fish. The
table below summarizes functional values identified for microbiota and macrobiota/fish.
Transition Zone Functional Values
1. Food source
2. Preferred habitat for some species (upwelling area)
3. Refugia for macro (predator avoidance)(biodiversity)
4. Microbially active zone
5. Habitat for food base
6. Cleaning zone (filters), vegetation, aquatic and riparian
7. Energy transfer
8. Discharge areas may have high biodiversity
Microbiota
V
V
Macrobiota/Fish
V
V
V
(1) High quality habitats/refugia
Discharge zones can provide thermal refugia for anadromous fish both for resting and for
spawning. Upwelling areas may be important by providing chemical/olfactory signals to anadromous
and migratory fish. The zone provides a microbial food supply to the fish and the upwelling areas can
act as incubators. Salmon need high quality water including cool water refugia in otherwise warm
stream reaches. Conversely, ground-water discharge environments may be the only areas where it is
warm enough to survive in very cold areas. Snow dimples have been used for years as surface
manifestations of ground-water discharges. Also, small areas in a lake could provide a large percentage
of the trout population with support. These can be unique habitats and important energy sources.
Certain fish seek out upwelling areas and shellfish may also live in these zones. Macrophytes (e.g.,
shallow eelgrass beds) may also benefit. Macrophytes may establish preferentially in beds related to
discharging ground water. Sometimes ground-water discharges into marine areas are the only areas
where emergent vegetation can grow. Another question is'whether some macroinvertebrates and fish
avoid contaminated ground-water upwelh'ng areas. Trout have good olfactory sense and will avoid
metals at concentrations well below toxic levels.
These zones also may limit benthic invertebrate exposure to low oxygen and contaminants by
creating oxygenated, clean zones. These zones may also be areas of plant and animal biodiversity.
They can be areas of high water quality in alluvial aquifers. Some European countries are interested in
identifying high quality ground-water discharge zones (good quality refugia) in the midst of
contaminated rivers to preserve as critical habitat.
60
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
(X) Contaminant attenuation/removal.
The transition zone is important for chemical and biochemical reactions that influence the quality
of the ground water discharged into the surface water. Metals, halogenated organic solvents, polycyclic
aromatic hydrocarbons, volatile organic compounds (VOCs), and nutrients can be degraded or
removed from ground water within the transition zone.
Volatile organic compounds (VOCs) were discussed in particular. The issue was whether there are
concentration thresholds of VOCs above which they poison biological communities. Where there is a
large VOC plume, there could also be bioaccumulating contaminants. If the VOCs were then degraded,
but the bioaccumulative contaminants (e.g., PCBs, creosote) were not, then bioaccumulation of
toxicants still would occur. This has implications for remedial decisions, especially if contaminants are
brought in through ground water. Some participants expressed the opinion that VOCs are ignored
generally because their toxicity thresholds are much greater than those for heavier contaminants, and
therefore they seem to show no risk in the water column. However, risk thresholds based on
continuous exposure to a hazard such as VOCs are different than those used in water quality criteria.
(3) Cycling of nutrients and carbon
Nutrients and carbon cycle very actively in this zone. Strong redox gradients enhance biogeo-
chemical activity and micrpbial processes. Both aerobic and anaerobic processes often occur within
close proximity of each other. Microbial biomass can serve as the base of a detrital food chain that can
be important to overall ecoystem productivity.
(4) Food base for benthic organisms
Microbes and fungi can provide food for other transition zone organisms that are more intimately
involved in the benthic food web of the surface water body. Many macroinvertebrates use the
transition zone extensively, and they are food for other organisms. If the zone is contaminated, the
result for invertebrates could be mortality, biomagnification and/or bioaccumulation.
WHAT METHODS CAN BE USED TO ASSESS THIS TRANSITION ZONE ECOSYSTEM?
Current methods for studying transition zones generally are not standardized and sometimes not
well developed. For example, scales may be mismatched (wells are too big to sample over decimeter or
centimeter gradients). Regardless of these difficulties, it is very important that ecologically related
sampling in the transition zone be coordinated with hydrogeological and chemical surveys at ground
water discharge sites. It was useful in the discussion to distinguish two groups of organisms,
microbiota and meiofauna/macrobiota. It was noted that it is difficult (but important) to show
contaminant effects on these groups.
CD Microbiota . . ,
a. Community structure. There is no standard method to determine microbial community structure.
Some methods in use include culturing, metabolic profiling, fatty acid fingerprinting, molecular
probes, or nucleic acid characterization. These methods are either limited or time-consuming.
Other methods involve 1) collection using ceramic beads or other artificial substrates that collect a
sample population in the transition zone; 2) artificial cores with natural materials; and 3) artificial
habitats/substrates. Procedures, methods, and equipment are usually designed to answer the
specific questions at hand. A method to evaluate drinking water called UDI (Under Direct
61
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Influence) was mentioned. The suggestion was made to focus on the algal community as a
surrogate. The algal community and the benthic interface has diagnostic value because there is a
rich literature of algae as bioindicators. The comment was made that including diatoms would be
time consuming and not too practical. A suggestion was made to develop tools to measure activity
first, then measure structure.
b.
Microbial activity/Junction. Again, there is no single ideal method. Methods in use or proposed
include bioassays (such as the Microtox bioassay), determination of metabolic rates and pathways,
describing the dominant terminal electron accepting process (methanogenesis, sulfate reduction,
iron reduction, manganese reduction, denitrification, or aerobic respiration), measuring molecular
hydrogen and testing for metal tolerance.
It may be difficult to generate interest in microbial function—microbes in septic tanks that provide
organic degradation are a familiar example. The transition zone is important for carbon cycling,
nutrient cycling, and a detrital-based food chain. Contamination should not interfere with these
processes and the decomposer community. So, what would be the appropriate method to evaluate
decomposition? Is the desired method to identify the amount of carbon no longer available (tied up in
ligands or refractory) or metabolized?
Another suggestion was to evaluate biological oxygen demand (BOD) and/or chemical oxygen
demand (COD). For example, the presence of soluble reduced metals will result in high COD and
affect interface chemistry. If ground water has high BOD/COD and dissolved oxygen (DO) is present,
that observation is important. However, all agreed that BOD and COD are presently impossible to
resolve across small scales, although fine-scale characterization of DO is possible.
2. Macrobiota/Meiofauna
a.
b.
Community structure. Several methods exist for sampling organisms in the transition zone (see D.
Williams' abstract on page 39 of this report) and various standard metrics can be computed
(community composition, density, species richness). Benthic and ground-water taxa can be
distinguished.
Function. The following were suggested: Tjioaccumulation studies and stable isotope analyses (e.g.
ISN/14N, 13C/12C, and ^S/^S) for food chain relationships. Functional feeding groups can be
evaluated.
It was suggested that these basics (community structure and function at all levels) be understood
first before developing more methods to conduct toxicity testing.
WHAT RESEARCH IS NEEDED TO BETTER UNDERSTAND THE ECOLOGICAL IMPORTANCE OF
AND ASSESS EFFECTS ON THE TRANSITION ZONE?
(1) Basic biological research
Most recommendations centered around basic science needs regarding the transition zone (e.g., life
histories, fauna! surveys, activity measurements) and sampling/evaluation tools. Life history
characteristics of transition zone organisms are generally lacking. Food chain relationships that
describe the linkages among microbial, meiofaunal and macrofaunal organisms also are lacking. A
suggestion was made to develop methods to conduct a subsurface biomass study. Because no large
reference databases exist (compared with surface water data), faunal surveys should be done for major
62
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
riverine ecosystems using a hydroclimatic landscape approach (see T. Winter's abstract on page 46 of
this report). These surveys would be used to develop reference conditions in a national database. If
there are differences between geographic areas, it may be best to look at functional differences rather
than community differences so data can be compared across broad regions. The Chemistry Group also
suggested establishing regional hyporheic study sites.
(2) Macrobiota
Species richness and growth could be evaluated. The physiology of transition zone invertebrates is
poorly known (e.g., O2 uptake rates and mechanisms are often unknown). Respiration studies are
needed as well as information on trophic structure. Stable isotopes of nitrogen might be an effective
way to determine food chain relationships. Dissolved oxygen availability should be accurately
measured. Good biological indicators are as yet uncertain and likely vary for differing flow paths or
discharge zones. One should look at biological impacts but use chemical and hydrologic conditions to
help define sampling zones.
(3) Indicators of ground-water discharge zones
The Chemistry Group discussed the scenario of a plume entering a stream and how to detect
effects in the subsurface. They suggested looking in four dimensions: vertically, horizontally,
temporally, and downstream. In general, a point source will be easier to detect than a diffuse plume.
You will need several transects across the river. What biological components should be measured?
Potential electron acceptors and dissolved hydrogen are good biogeochemically informative
constituents to measure. You can characterize the microbial community in many ways. Culturing
methods normally select for small subsets of the total microbial community. Molecular techniques also
can be used, but presently none of these methods are easily and routinely applied.
Indicator choices depend on the question to be answered. Which attributes are you protecting?
Microbial assays need to be used, even if these assays are not yet perfected. Promising techniques are
currently under development. Morphological measurements in the system are easier to make than
biological measurements. Intensive sampling near the point of discharge plus additional transects
would be useful. Sampling should include "vertical distributions" through the food chain.
(4) Biological indicators of GW discharge zones
Are there any biological attributes that help define ground-water discharge zones? For example,
can you look for benthic algal blooms? Are fish numbers and distributions in context with other
indicators a useful means to locate discharge zones where high quality aerobic ground water is present.
Some species may tend to remain in an area even if contaminated."The mechanisms by which fish and
other species avoid contaminants is very complicated. Distribution of fish does not necessary follow
water quality parameters. Are ostracods good indicators? The consensus was no. It was suggested that
midge larvae might be better indicators for ground-water discharge zones. One documented indicator is
the presence of high biomass benthic algal mats, but this is limited to zones with enhanced nutrient
discharge. Some discharge zones are dead zones, especially where anaerobic, metal-rich ground waters
are discharging. There is an important research need to try and correlate between bottom type and
patchiness with ground-water discharge. In lake ecosystems, these zones may be linked to aggregations
of zooplankton. Acoustic techniques that detect these aggregations may be able to locate ground-water
discharge points in lakes
63
-------
Proceedings of the Ground- Water/Surface- Water Interactions Workshop
July 2000
(5) Chemical/physical indicators of ground-water discharge zones
Temperature and conductivity probes are simple, easily-used, and rugged tools for determining
ground water discharge locations. These methods could be routinely used to guide sampling in many
aquatic ecosystems. Bottom drags with temperature and conductivity probes also can be considered if
site conditions warrant. Protocols are needed to allow better comparisons among sediment samples and
data from temperature and conductivity probes. Although DO probes are somewhat unstable in the
field, investigators could use combined temperature, conductivity, Eh, and DO measuring
instrumentation to look for discharge zones. Oregon State University has a suite of fiber optic sensors/
probes that are commercially available and potentially useful in these transition zones. Redox
measurements in the field are a problem because of a lack of equilibrium in many samples, and redox
potential is often dominated by iron biogeochemistry. Tools needed for improved sampling of ground
water discharge zones include:
• Sampling devices to collect organisms effectively and quantitatively along transition zones;
• Dependable and cost-effective geophysical and tracer tools to delineate transition zones and guide
biological sampling; and
• Routine survey tools to better characterize microbial community structure and activity and assess
water quality and condition.
C6) Scale
Strong gradients in physical and chemical parameters commonly exist in the transition zone. For
example, the distribution of redox sensitive solutes can be very steep. Sampling often must be at the
centimeter scale or finer resolution. All participants agreed that we need better methods to sample
gradients and narrow transition zones. Microcosms or fine-scale bioassays may be approaches to
consider.
(?) Hydrology
Knowledge of hydrologic characteristics of the transition zone is crucial. For example, transpira-
tion rates may be very important to the hydrology of these interface zones, but there are large regional
differences. Chemical and isotopic tracers may be the best methods to determine the effect of the
transition zone on overall stream quality. Some tracers also are sensitive to in-stream processes. Other
participants pointed out the need for subsurface measures in addition to surface water sampling.
Unresolved questions include:
• What techniques are available for measuring the volume of water entrained into the hyporheic
zone?
• What are biological consequences of remediation (pump and treat) that reverse flows in the
transition zone?
Injecting oxygenated water could change the redox chemistry within the hyporheic zone. Highly
regulated rivers (dammed) affect the hydrology of this interface as well.
(8) Signal-to-noise and partitioning sources
Some practical sampling questions were raised about characterizing the transition zone. How is it
different from a place without ground-water recharge or discharge? The responsible party will need to
prove that the background contamination "noise" is greater than their contribution. How can you
64
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
compensate for variability (from the regulator's perspective)? We need screening tools (inexpensive)
to identify the problem and focus the sampling. What methods can distinguish ground water from
sediment sources? Ground-water discharge may become contaminated as it flows through the
hyporheic zone, becoming a "fingerprinting" challenge.
(9) Temporal variability
Temporal variability is important: hourly variability in the hydrology, chemistry, and biology of the
transition zone has been noted. When are the best-case and worst-case times for sampling; which
season or seasons should be sampled? Different life stages have different susceptibilities and
exposures. Ecologists and hydrogeologists need to collaborate. Ecologists can specify time of year and
depths of concern; hydrogeologists can determine the hydrologic regime and geochemistry.
(10) Remote sensing
Field studies combined with remote sensing now can be used to better understand the
heterogeneity and landscape characteristics of transition zones. Hydrology and food resources for
important species are not homogeneously distributed but often highly localized. Remote sensing
provides a tool for assessing landscape-scale patterns of hydrology and biotic distributions. Certain
patterns on the landscape (e.g., localized plankton blooms) may be surficial indicators of processes
occurring in the transition zone. Researchers and managers need to combine extensive and intensive
analyses.
(11) Toxicitv testing
Are there any non-lethal endpoints or tools that could be used to determine or screen for toxicity on
transition zone organisms? Growth studies are generally more sensitive than mortality or fecundity
studies. Are there any ground-water toxicity tests or ground-water bioassays? One suggestion was that
Elmid beetle larvae in the hyporheic zone may be suitable test species.
(12) Nutrients
The role of transition zones in overall nutrient cycling is still poorly known. Nutrient effects need
to be related to species effects, such as effects on sea grasses or corals. The management goal would
be to protect "normal" nutrient cycling. Most people live near coasts, and impacts on transition zones
that affect riverine delivery of nutrients or ground-water discharge of nutrients in estuarine or coastal
waters are critical processes that need to be better understood and monitored. In general, we do not
know the trends in nutrient delivery from these transition zones for rivers, estuaries, or coastal waters.
In addition, the rates and locations for nutrient transformations by'microbial organisms in transition
zones in coastal regions deserves further study. There have been relatively few attempts at quantifying
these processes.
Similarly, nutrient cycling processes in the hyporheic zone should be better studied. Hyporheic
zones receive dissolved oxygen when surface water recharges ground waters. Oxygen participates in
important biogeochemical processes such as aerobic respiration, nitrification, metal oxidation, sulfur
oxidation, and methane oxidation. For example, if ammonium levels are high in ground waters,
nitrification rates can increase and lead to higher concentrations of dissolved nitrate. Where these
processes occur and the seasonally of such processes can affect both surface water and ground water
quality. Can the portion of nutrient loading in a surface water body that is derived from ground water
be distinguished from that derived from surface sources? The U.S. Geological Survey (USGS) has
65
-------
Proceedings of the Ground- Water/Surface- Water Interactions Workshop
July 2000
done some related work on this topic in Massachusetts. The contribution from ground water in polluted
areas is at least as great as the contribution from rivers in many coastal areas. Tools exist for toxicity
testing, but comparable tools do not exist for assessing impacts on nutrient cycling. What methods
exist to test whether nitrate is being removed or if that function is impaired? Researchers are worldng
on these methods, but they are not yet regularly employed in monitoring programs.
fl 3) Dissolved oxygen
The availability of dissolved oxygen plays a major role in the characteristics of ground waters in
transition zones. Not all oxygen depauperate discharge zones are caused by pollution; some are •
naturally low in DO due to hydrologic flow paths (residence time) and rates of microbial metabolism
on sediment organic matter. Anaerobic ground waters may contain increased concentrations of
dissolved metals, sulfur, and methane. Dissolved oxygen is a master variable in processes and
chemical characteristics of transition zones.
(14) Reference comparisons
A disturbed zone needs to be compared to a "normal" reference. How do you identify conditions
for comparison? How can you identify effects of the contaminants? How can biological conditions be
used as a reference? "Acceptable" conditions .need to be defined. Some biotic species (e.g., caddisflies
and mayflies) can be used to define reference conditions. Paleontology tools can be used to determine
prior conditions. Either reference or gradient comparisons can be used to evaluate changes. The group
recommended assessments that allow cross-comparison after remediation (monitoring). The group
considered how to define reference conditions in ground water for a superfund site. One approach
would be to evaluate current approaches for macroinvertebrates. It would be crucial to locate samples
in ground water outside the area of influence. Defining what is meant by reference or reference
condition always is challenging. The area should have the same ground-water characteristics in terms
of hydrology and chemistry, but without the contamination. This is difficult, because the plume may be
a small part of the total ground-water discharge and dispersed contamination may be widespread at a
site. It may be easy to find nearby discharge locations that apparently are not contaminated, but it will
be critical to carefully assess if these aquifer sediments and ground waters are actually not
contaminated.
(15) Correlations between hydrology, sediment, and biology
There have been some correlations described between hydraulic conductivity and ground-water
discharge, but not further linked with the biology. Differences in biota occur between upwelling and
downwelling areas. Silty or clayey soils (sediments) can inhibit the ground-water flux. Most freshwater
macroorganisms do not like turbid water. There may be a juxtaposition of preferred soil type and
discharge zones. Adequate characterization of soil structure, porosity and organic matter content are
necessary. Clogging, percent organics, amount of DO, and other variables need to be measured.
Organisms often preferentially select substrate, so standard artificial substrates sometimes can be used
as a surrogate for enhanced comparability between sites.
(16) Bioaccumulation
Diffuse flows and low concentrations of contaminants are hard to measure. Measuring biota that
receive contaminants from multiple sources will increase the problem of documenting that a problem
exists only from a single ground water source. Bioaccumulation is not always a problem. Lipid bags
may not be a very good method for assessing bioaccumulation, because one of the main biological
66
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
components is accumulation through the food chain. Semi-permeable membrane devices (SPMD) "fat
bags" might be a better method. Another possible method would be to look at higher trophic levels
such as fish. Nitrogen isotope signatures change over time and are dependent on the trophic level of the
animal. Therefore, fish 15N/14N ratios and bioaccumulation analyses can be used in combination to
deduce an impact from contaminant delivery through ground waters to surface waters.
WHAT BIOLOGICAL MEASUREMENTS DO YOU WISH YOU HAD AT YOUR SITES?
During this discussion, participants identified key measurements that biologists, chemists, and
hydrologists would have liked to have had in studies of transition zones:
• Botanical analysis indicative of natural acidic stream condition for studies considering
anthropogenic acidification. Sediment probes and piezometers have been used, but no biological
data have been collected. .
• Sediment and interstitial water toxicity data on Daphnia. Toxicity testing in general would be
valuable as we usually get only chemical information. Would the results from those methods be
any different than from existing bioassays?
• A test where you can measure impacts on nutrient cycling.
• How many replicates can be processed to account for patchiness? How patchy can it get?
• Toxicity tests for biota in the hyporheic zone following their reaction to exposure or accumulation
over time. The tests should be analogous to fish indicators (e.g., hiccuping) or integrative tests such
as bee pollen sampling of contamination over a certain radius.
There was general agreement that it would it be useful to develop a suite of toxicity tests for
microbes and invertebrates. Microtox is the only commonly used test (luminescence is the endpoint),
usually for screening. Certain contaminants lower luminescence and many microbes thrive on
contaminants. Microbial toxicity tests therefore may not show anything. There is a lot of natural
variability spatially and temporally in electron accepting process. Results depend on the location and
timing of sampling. Microtox is usually used for sediment toxicity. One needs to design and interpret
the test based on the endpoint of concern.
OTHER QUESTIONS/SUGGESTIONS
• Is organic carbon available to the food-base (labile organic carbon content) a sensitive indicator of
microbial activity?
• Can microbiota in the transition zone be thought of as sources of primary productivity like
microbial communities in estuarine sediments?
• In Europe, invertebrate organisms are sometimes used as indicators of ground-water quality.
Transition zone organisms in the U.S. also could be evaluatedf or their potential as indicators.
• How can adverse ecological impacts in the transition zone be recognized? Would an indication be
when you do not have the anticipated biodiversity?
• What scale should be used to define adverse impacts? The scale depends on the site's risk
management goal.
• Encourage thinking about the need to better integrate biology, hydrology, and biogeochemistry.
• The workshop report should include references to available methods for microbial, epifauna and
meiofauna sampling. There are methods available for many species.
67
-------
POSTER SESSION ABSTRACTS
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Use of Multi-Parameter Sensitivity Analysis to
Determine Relative Importance of Processes
Involved in Transport of Mining Contaminants
By Jungyill Choi, Judson W. Harvey, and Martha H. Conklin
ABSTRACT
Combining multi-parametric sensitivity analysis (MPSA) with stream transport modeling is
proposed to determine the relative importance of physical and biogeochemical processes controlling
transport of mining contaminants in natural stream systems. The MPSA is based on a large number of
Monte-Carlo simulations to identify the sensitive parameters over a broad range of each parameter.
This combined approach can provide an integrated view of transport processes of contaminants in
natural stream system.
INTRODUCTION
The fate and transport of contaminants in streams and rivers are controlled by a variety of physical
and biogeochemical processes. The physical processes play an important role in determining the fate of
solutes in surface-water environments. These physical processes include advection, dispersion,
hyporheic exchange, and ground-water interaction. In many situations, however, the transport of
contaminants are also greatly affected by biogeochemical processes, such as sorption/desorption,
oxidation/reduction, volatilization, hydrolysis, biodegradation, and other biochemical reactions.
Therefore, transport of contaminants in natural streams and rivers is best described by considering all
of the relevant physical and biogeochemical processes simultaneously (fig. 1).
Stream-tracer
experiments
Transport Model of Contaminants
in Stream-Shallow Groundwater
System
Physical Processes
•Advection
•Dispersion
•Hyporheic exchange
•G.W. interaction
Multi-Parametric
Sensitivity Analysis
Detailed field
and laboratory
measurements
Relative Importance of
involved Physical and
Biogeochemical processes
Biogeochemical Processes
•Aquatic chemistry
(for example, O2 andpH)
'•Precipitation/Dissolution
•Redox processes
•Sorption/Desorption •
•Microbial activities
Figure 1. Coupling MPSA with transport mode! to identify the relative importance of
physical and biogeochemical processes.
To answer the question about relative importance of factors, the sensitivity of a numerical transport
model needs to be tested for the physical and biogeochemical parameters (processes) that are involved
in the forward transport model. However, traditional parameter-sensitivity analysis pertains to a
69
-------
Proceedings of the Ground- Water/Surface- Water Interactions Workshop
July 2000
particular point (localized) in the parameter space, which is defined by all possible combinations of
parameter values. Also, in the localized sensitivity analysis, the importance or sensitivity of a selected
parameter can be affected greatly by the values of other parameters, because the significance of one
selected process is usually dependent on other processes. Typically, the importance of biogeochemical
processes are highly dependent on the physical processes, whereas the physical processes are not
affected by the biogeochemical-processes. For example, the biogeochemical reactions of solutes in the
hyporheic sediments are enhanced by the prolonged retention time of solutes in these sediments.
Therefore, to account for parameter interactions, the relative importance of the physical and
biogeochemical processes of the transport model can be evaluated more accurately by a generalized
(multi)-parameter sensitivity analysis, which encompasses the entire parameter space (fig. 1).
This paper presents the concepts and procedures of multi-parameter sensitivity analysis (MPSA)
that is used to determine the relative importance of transport processes
METHODOLOGY
A numerical transport model may include detailed field measurements as well as ill-defined
parameters that cannot be measured with a high degree of accuracy in the field or in the laboratory.
These ill-defined parameters will severely limit the accuracy of any single simulation and increase the
difficulty of assessing the relative importance. In an attempt to overcome this difficulty and to
recognize the relative significance of parameters involved in the model, the sensitivities of simulations
results to input parameters need to be evaluated by assigning either a range of variation or a degree of
uncertainty to each parameter and implementing a generalized sensitivity analysis (Hornberger and
Spear, 1980; Chang and Delleur, 1992; Choi, etal, 1998; Choi, 1998). This multi-parametric
sensitivity analysis (MPSA) followed the procedure proposed by Chang and Delleur (1992) and Choi,
et al. (1998). The procedure includes the following steps:
• Select the parameters to be tested.
• Set the range of each parameter to include the variations experienced in the field and laboratory
measurement.
• For each selected parameter, generate a series of, for example, 500 independent random numbers
with a uniform distribution within the design range.
• Run the model using selected 500 parameter sets and calculate the objective function values.
• Determine whether the 500 parameter sets are 'acceptable' or 'unacceptable' by comparing the
objective function values to a given criterion (R).
• Statistically evaluate parametric sensitivity. For each parameter, compare the distributions of the
parameter values associated with the acceptable and unacceptable results. If the two distributions
are not statistically different, the parameter is classified as insensitive; otherwise, the parameter is
classified as sensitive. Relative importance can be evaluated statistically if desired.
The objective function values of the sensitivity analysis usually are calculated from the sum of
squared errors between observed and modeled values:
70
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
where/is the objective function value and xc(I) and x0(I) are calculated and observed values,
respectively. Observed values often are obtained from simulations that used the mid-points of the
characteristic range for each parameters. The ranges for each parameter are determined from minimum
to maximum values that are obtained from parameter estimations and field measurements through the
study reaches. If the objective function value obtained from the simulation is less than a subjective
criterion then the result is classified as acceptable, otherwise the result is classified as unacceptable.
Three different objective function values often are tested for a subjective criterion. Those values
typically define the 33, 50 and 66% divisions of 500 sorted objective functions.
The basic concept of MPSA is illustrated by using a hypothetical model with only two parameters
(Figure 2). In addition, the modeling procedure of MPSA described above is summarized using a
flowchart (Figure 3).
Figure 2. Basic concept of multi-parametric sensitivity analysis (MPSA)
using a hypothetical model with only two parameters.
CONCLUSIONS
Comparison of the
distribution between
acceptable and
unacceptable cases
Evaluate the
sensitivity of each
selected parameter
Figure 3. Flow chart illustrating the
procedure of multiparametric sensitivity
analysis (MPSA).
The combined efforts of forward modeling approach and generalized sensitivity analysis can
provide an integrated view and better understanding of contaminant transport processes in natural
stream systems. The multi- parametric sensitivity analysis especially helps identify the relative
importance of physical and biogeochemical processes controlling the transport of contaminants.
Furthermore, this methodology can provide a guide for future data-collection efforts and to order
research priorities.
71
-------
Proceedings of the Ground- Water/Surface- Water Interactions Workshop
July 2000
REFERENCES
Chang, R, and J.W. Delleur, 1992. Systematic parameter estimation of watershed acidification model,
Hydrological Processes, v. 6, pp. 29-44.
Choi, J, 1998. Transport modeling of metal contaminants in a stream-aquifer system; University of
Arizona, Department of Hydrology and Water Resources, unpublished PhD thesis, 225 pp.
Choi, J.Y., Hulseapple, S.M., Conklin, M.H., J.W. Harvey, 1998. Modeling CO2 degassing and pH in a
stream-aquifer system, Journal of Hydrology, v. 209, pp. 297-310.
Hornberger, G.M. and R.C. Spear, 1980. An approach to the preliminary analysis of environmental
system, Journal of Environmental Management, v. 12, pp. 7-18.
72
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Measurements of Plant and Algal
Bioaccumulation of Metals in Final and Pinto
Creeks, Arizona
By Justin C. Marble, Timothy L. Corley, and Martha H. Conklin
Dissolved Mn is an essential element for higher plant systems and is involved in photosynthesis
(the Hill reaction) and activation of different enzyme systems (e.g., superoxide dismutase production)
(Mukhopadhyay and Sharma, 1991). Critical deficiency levels of Mn(II) range between 0.01 to 0.02
microgram Mn per gram (mg Mn(II) g'1) dry weight in dry mature leaves but vary tremendously
between plants (Mukhopadhyay and Sharma, 1991). Vascular plants and algae also require certain
amounts of other trace metals for normal plant growth (Zn, Ni, Cu, Fe, Co, Ca, and Mg).
Although Mn(E) supplements can increase growth yields of plants, large amounts of Mn(E) can
interfere with the uptake of other trace metals (Mukhopadhyay and Sharma, 1991). In addition, excess
concentrations of Zn, Ni, Cu, Fe, and Co can trigger an inherent defense mechanism that plants have
developed that involves production of phytochelatins—polypeptides that bind metals (Ahner, et al.,
1995). Phytochelatin production in response to high metal levels has been identified in land plants,
vascular aquatic plants, fungi, and marine and freshwater algae. This mechanism results in an
accumulation of the excess metals within the plants with the final metal concentration often being
significantly higher than found in water supplied to the plants.
The work reported in this paper focuses on bioaccumulation of metals by aquatic plants, algae, and
moss in Final Creek, an Arizona State Superfund site, near Globe, Arizona, that has been contaminated
by acid-mining activities in the area. The primary purposes of this study were to determine the extent
to which metals were taken up by the diverse plant community at Final Creek and to determine which
plants were particularly effective at bioaccumulation of metals. To further aid in our assessment of the
potential role of plants as a sink for metal contaminants in Final Creek (Figure 1), comparisons of
metals uptake were made with other measurements reported for similar plants in Pinto Creek, also near
Globe, Arizona. A comparison of typical surface
water data for Final and Pinto Creeks is given in
Table 1.
Plant grab samples were collected from several
locations and rinsed with creek water to remove
insects and loosely attached sediment material. At
Final Creek, plant samples were collected from
sites ZO, J2-l, J2-5, J2-15, and Zll (Figure 1). At
Pinto Creek, grab samples were collected from two
USGS stream gaging sites 09498501 (below
Haunted Canyon near Miami, Arizona) and
09498502 (Pinto Creek near Miami, Arizona).
Pringle Diversion Dam
ARIZONA
J2-1&J2-5
pH7.1
J2-15&Z11
pH7.3
Well Group 600
EXPLANATION
Perennial Reach
Intermittent Reach
A Sampling Site
1 KILOMETER
ZO
pH6.0
Coretta Driveway!.
Setka
Ranch
Well Group 500
Figure 1. Study reach and sampling sites at Final
Creek. Median pH values are shown for the study
period.
73
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Parameter
PH
Oxygen
Alkalinity1
TDS
Co(II)
Cu(U)
Fe(H)
Mnai)
Ni(II)
Zn(II)
Final
Value
6.4
6.9
51
2,640
0.410
0.050
<0.130
72.0
0.790
0.500
Pinto
Value
7.8
9.0
180
531
0.003
0.010
0.0053
0.0038
0.010
0.0060
1 As CaCOj.
Table 1. Physical and Chemical Values for Final Creek
(ZO on January 25,1995) and Pinto Creek (near Miami,
Arizona, on June 18,1997, USGS). (mg L'1 except for pH
which is in standard pH units).
After rinsing with creek water, the plant samples were placed in plastic bags and put into a cooler.
Upon arrival at the laboratory, samples were dried at 60°C for 24 hours. Dried samples were ground
and sieved, then digested with nitric acid. Digested plant samples were analyzed by flame or graphite
atomic absorption spectroscopy for different metal concentrations. Results are reported as
bioaccumulation, i.e., mg of metal per kg of dried plant material (mg kg'1). The values reported
represent the average of 2 subsamples with the maximum and minimum values measured being within
±2 percent of the average value.
The aquatic plant species found at Final Creek varied in type and density depending upon the time
of year and the location. Before plant sampling started in 1996, water speedwell (Veronica anagallis-
aquatica) and rabbitfoot grass (Polypogon monspeliensis [L.J [Desf.]) dominated the upstream portion
of Final Creek (J2-l) and algae (e.g., Microcystis, Vaucheria, and Oocystis) dominated in the
downstream section (J2-15). However, over the study period (November, 1996 through June, 1997),
water speedwell, rabbitfoot grass, and algae were found along the entire study reach.
Water speedwell from Final Creek was collected from several field locations (ZO, J2-l, and J2-15)
over a period of 8 months and analyzed for Mn(Di). There was no obvious correlation between
sampling date and bioaccumulation of Mn at J2-! and J2-15 . A subset of the water speedwell samples
from sites J2-l and J2-15 were analyzed for other trace metals (Table 2). No trend with location was
observed for concentrations of Fe, but Zn and Ni were higher at J2-15 than at J2-! and Cu was higher at
J2-! than at JM5. Bioaccumulation of Mn and Co exhibited consistently higher bioaccumulation at J2-
15 compared to J2-!, about a factor of 2 difference.
Metal
Mn
Mn
Fe
Fe
Ni
Ni
Cu
Cu
Co
Co
Zn
Zn
Date
12/13/96
1/31/97
12/13/96
1/31/97
12/13/96
1/31/97
12/13/96
1/31/97
12/13/96
1/31/97
12/13/96
1/31/97
J2-1
6450
7990
4400
2520
109
148
901
1750
80.5
134
516
772
J2-15
18600
16400
1880
2670
151
182
824
1130
158
279
665
801
Table 2. Water speedwell bioaccumulation from Final
Creek collected on December 1996 and January 1997 for
Mn, Zn, Ni, Cu, Co, and Fe (units are mg kg-1).
74
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Rabbitfoot grass samples from J2-l and J2-15 were also analyzed for Mn, Zn, Ni, Co, and Fe (Table
3). Both upstream and downstream sampling sites had similar bioaccumulation values for Zn and Ni,
but Mn, Cu, Co, and Fe values were larger at site J2-l than J2-15. A factor of about 2 between values at
J2-l and J2-15was observed for Mn, Co, and Cu, and a factor of about 10 for Fe. Bioaccumulation of
Mn at both sites was also greater than the other metals. Duckweed (Lemna minor) was less widely
distributed than either water speedwell or rabbitfoot grass, and was typically only found in slow
moving or stagnant water near the banks of the creek. However, a sample collected from P-5 on June
25,1997 (pH 7.1, Mn(H) concentration 47.0 mg L'1) had an Mn concentration of 10760 mg Lr1.
Table 3. Rabbitfoot grass bioaccumulation in samples from Final
Creek collected on January 31,1997 (units are mg kgrl).
Metal
Mn
Fe
Cu
Ni
Co
Zn
J2-l
13600
6890
1640
163
237
581
J2-15
5240
691
828
161
130
534
Algae is prolific at both Final Creek and Pinto Creek and grab samples at both sites included the species
Microcystis, Vaucheria, and Oocystis.. Samples were collected from both creeks to compare
bioaccumulation of Mn (Table 4). Although Final Creek samples had more bioaccumulation, the ratios of
plant concentration to surface water concentration were greater in the Pinto Creek samples.
Table 4. Algae samples from Final Creek and Pinto Creek:
bioaccumulation of Mn (mg kg"1).
Site
Final, Zll
Final, Zll
Final, J?-15
Final, JM5
Pinto, Miami
Pinto, Miami
Date
7/17/96
11/15/96
12/12/96.
1/31/97
6/18/97
6/18/97
Mn
49700
90200
5550
79300
240
1460
Water speedwell collected from Pinto Creek had lower bioaccumulation of Mn than samples collected
from Final Creek (Table 5). However, the ratios of Mn concentrations in water speedwell to the surface
water concentrations were significantly higher in Pinto Creek than found for Final Creek. Similar
differences were apparent from comparisons of the algae samples collected at Pinto and Final Creeks
Table 5. Water speedwell bioaccumulation
(mg kg'1) in Final and Pinto Creeks.
Site
Final, J2-l '
Pinto, Haunted
Canyon
Pinto, Miami
Date
6/25/97
6/18/97
6/18/97
6/18/97
Mn
3870
505
97
47
These studies indicate that water speedwell, rabbitfoot grass, and algae bioaccumulate Mn.
Bioaccumulation of Zn, Ni, Co, Cu, and Fe was also observed in water speedwell and rabbitfoot grass.
Comparisons between water speedwell and algae samples collected from Final Creek and Pinto Creek
suggest that at Final Creek the plant capacity for metal uptake may have been reached and/or that metal
toxicity effects must be considered. Water speedwell and other aquatic plants are prolific in Final Creek
and could play a significant role in determining the fate of metal contaminants entering the stream.
Additional data concerning the total biomass in the system, and the potential release of metals as plants
75
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
die and decay, are required to assess the potential and actual contribution of plants to total metals removal
in this system.
ACKNOWLEDGMENTS
This publication was made possible by grant number P42 ESO4940 from the National Institute of
Environmental Health Science with funding provided by EPA and by grant number EAR-95-23881 from
the NSF. Its contents are solely the responsibility of the authors and do not necessarily represent the
official views of the NIEHS, NIH, or EPA, or NSF.
REFERENCES
Ahner, B.A., Kong, S., and F.M.M. Morel, 1995. Phytochelatin production in marine algae—An
interspecies comparison, Limnology and Oceanography, v. 40, pp. 649-657.
Mukhopadhyay, MJ. and A. Sharma, 1991. Manganese in Cell Metabolism of Higher Plants, The
Botanical Review, v. 57, pp. 17-149.
U.S. Geological Survey, 1997. unpublished data of chemical and physical parameters from Pinto Creek
near Miami, AZ, Station Number 09498502, Lab ID 1830173, June 18.
AUTHOR INFORMATION
Justin C. Marble, Timothy L. Corley, and Martha H. Conklin, The University of Arizona, Department
of Hydrology & Water Resources, Tucson, Arizona.
76
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Tracing Groundwater Flow into Surface Waters
by Application of Natural and Artificial Tracers
By D. Reide Corbett, William Burnett, Jeffrey Chanton, and Kevin Dillon
Submarine groundwater discharge (SGD) is an often overlooked yet possibly significant process in
the geochemical and nutrient budgets of marine nearshore waters. According to Johannes (1980) "SGD
should occur anywhere that an aquifer is hydraulically connected with the sea through permeable rocks
or bottom sediments and where the head is above sea level." Such conditions are met in most coastal
areas. This process may be significant for transport of limiting nutrients in pristine coastal areas or, in
the case of polluted aquifers, could be an important source of contamination to the marine
environments. The problem is how to assess the extent of the groundwater flow and how to link
environmental problems with specific sites of contamination. Due to the extreme temporal and spatial
variability of many of these variables, the exact location of problematic discharges into coastal regions
may be difficult to determine by monitoring standard water quality constituents (e.g., NOX, turbidity).
In this research, subsurface water movement was evaluated with natural and artificial tracers in the
karst limestone of the Florida Keys
(Figure 1).
In the Florida Keys, natural
tracers (222Rn and CH4) were
used to locate areas of
increased groundwater/surface
water interactions by
reconnaissance surveys of the
concentrations of radon and
methane in the bay waters
(Corbett, et al., 1999). These
trace gases function as natural
indicators of submarine
groundwater discharge into
standing bodies of water due to
their significantly higher
concentrations in groundwaters
(Cable, et al., 1996a, b; Bugna,
et al., 1996). General trends in
surface water concentration
were established by contouring
data from each tracer survey
with a kriging method by use
of the software package
Surfer® (Golden Software).
Although kriging interpolates Figure 1. Florida Bay separates the Florida Keys, located off the southern tip of
i, * A* JLtc, ,.«,ariT,«. Florida, from the mainland. Water samples were collected primarily from north of
between data points,- creating Long Key and £ast Q{ Ramingo Groundwater samples from offshore wells were
some artifacts, the general collected where indicated by the circles. Letters refer to locations mentioned in
trends described are the text: A. Carysfort Reef; B. Algae Reef; C. French Reef; D. Molasses Reef; E.
independent of the contouring *£SE£gJ-E* "* """ "" *
77
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
method or a reasonable change
in contouring concentration.
Examination of these contour
plots showed very little
apparent seasonal variation
throughout the study period.
During each period we
sampled, high concentrations
of both tracers were observed
near the Keys. Plots for 222Rn
and CH4 in summer 1997 show
the typical trends observed
(Figure 2). Dkect
measurements of groundwater
flux via seepage meters were
also made in several different
areas of Florida Bay. Radon
and methane concentrations in
water samples collected from
wells, springs, canals, and
Florida Bay showed a
significant correlation, despite
the fact that the two trace
gases have independent source
terms (Figure 3). Natural
abundance of nitrogen
isotopes measured on attached
algae and seagrass also show
greatest 15N enrichment in
areas near the keys. We
observe a strong spatial
gradient in 15N of macrophytes
(seagrasses and macroalgae)
in Florida Bay, with relatively
light (-1 to 4 %o) macrophytes
in western Florida Bay and
relatively heavy (6 to 13 %o)
S^B^mS^lS Figure 2"Contours of radon (A) in -dpm L~1 and methane in nM for
Monda Bay (Figure 4). This samples collected in June/July 1997. Solid crosses indicate sampling
gradient is likely a function of locations. Note the darker contours, indicating higher concentrations of
two processes: (1) progressive both parameters, near the upper Keys. (Figure from Corbett et al., 1999)
denitrification of N brought
into Florida Bay via tidal exchange with the Gulf of Mexico; and (2) entry of 15N-emiched water from
the subsurface adjacent to the Keys in northeastern Florida Bay. Collectively, these results indicate a
greater flow of groundwater along the inside of the keys. Nutrient flux estimates, based on interstitial
nutrient concentrations and groundwater flux measurements, suggest that groundwater in the eastern
area of Florida Bay may provide as much nitrogen (110 ± 60 mmol N m'2 y1) and phosphate (0.21 ±
0.11 mmol PO43- m'2 y1) as surface freshwater sources from the Everglades (i.e., Taylor Slough and C-
111). However, the inputs are clearly not uniform and areas near solution holes/tidal springs may have
a substantially greater nutrient flux into surface waters then these estimates (Corbett, et al., 1999).
- Methane
Ju ne/J u Jy 1997
78
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
2000
1500-
1000
500
Lois Key Spring
Garden Cove Surface
Garden Cove Spring
Porjoe Interstitial
Porjoe Surface
Groundwater Average
i5 In Seagrasses and Maeroalgae
11
Figure 4. Contours of 15N in macroalgae collected
throughout the study period. Solid crosses indicate
sampling locations. Note the darker contours, indicating
higher enrichment, near the upper Keys. (Figure from
Corbettetal.,1999).
100 200 300 400
Excess Radon'(dpm I/1)
500
Figure 3. Radon and methane concentrations in waters
sampled throughout the Keys. The groundwater tracer
concentrations are based on the overall average of all
samples collected. (Figure from Corbett et al., 1999).
Artificial tracers (SF6,1311,32P) were used to
establish a direct link of contaminated ground-
waters to surface waters. Tracers injected
directly into sewage injection wells indicate
rapid flow of groundwater beneath the keys.
Experiments conducted on Long Key indicate
two different types of transport: (1) rapid flow
(0.20-2.20 m/hr), presumably through cracks
and conduits present in the limestone; and (2)
slow diffusive flow (<0.003-0.14 m/hr),
associated with the limestone's primary porosity
(Dillon, et al., 1999). Vertical flow of the wastewater effluent was" comparable to horizontal flow due
to the buoyancy of the relatively fresh wastewater compared to the surrounding saline groundwater.
These experiments showed that solutes injected into the Key's subsurface have the potential to reach
surface waters within a few days (Figure 5). Tracer experiments conducted using both a conservative
tracer (SF6,131I) and nutrients of interest (nitrogen and phosphorous) showed that both nitrate and
phosphate have some non-conservative behavior. Either through microbial alteration or interaction
with the limestone matrix, water from the wastewater injection appears to be polished as it flows
through the subsurface.
In a review of the general subject of SGD, Johannes (1980) stated that "It is...clear that submarine
groundwater discharge is widespread and, in some areas, of greater ecological significance than
surface runoff." I agree with this appraisal and add that from my review of the available literature, I
400-j
5 20°-
5 100-
ft-
• Bay
» Caral +
»
. s ^ -
MDA = 41±18 " " *
• i
"* • " * •
Time (days)
Figure 5.1-131 in surface waters sampled on both sides
of Long Key, Florida Bay (closed square) and the
Atlantic via a canal (closed diamonds). Radioactive
iodine was added to a sewage injection well as a
conservative tracer to track wastewater movement in the
subsurface. Due to the rapid vertical and horizontal
movement of wastewater in the subsurface, injected
tracers appear in surface waters within 3 days after
injection.
79
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
find that it has been largely the biological community which has recognized the potential importance of
SGD. I maintain that the process of SGD may also be an important and overlooked part of the
geochemical cycles of many elements. In the case of Florida Bay and the Florida Keys, SGD has been
completely ignored in previous nutrient budgets in the area and has therefore not been considered as a
potential threat of contaminants (e.g., sewage effluent) to the ecosystem. It is hoped that one of the
main outcomes of this research will be the development of an approach which integrates geochemical
and hydrogeological techniques for assessing directions and rates of subsurface flow and, specifically,
how to quantify the flow into surface waters.
REFERENCES
Corbett, D.R., J. Chanton, W.C. Burnett, K. Dillon, C.M. Rutkowski, and J. Fourqurean, 1999.
Patterns Of Groundwater Discharge Into Florida Bay, Limnol. Oceanog.
Dillon, K., D.R. Corbett, J.P. Chanton, W.C. Burnett, and L. Kump, 1999. Rapid transport of a
wastewater plume injected into saline groundwaters of the Florida Keys, USA. Submitted to
Groundwater,
Johannes, R.E., 1980. The ecological significance of the submarine discharge of groundwater, Mar.
Ecol Prog., Ser. 3, pp. 365-373.
AUTHOR INFORMATION
D. Reide Corbett, William Burnett, Jeffrey Chanton, and Kevin Dillon, Department of Oceanography,
Florida State University, Tallahassee, Florida 32306-4320. email: rcorbett@ocean.fsu.edu- Tel-
850-644-9914, Fax: 850-644-2581
80
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Considerations for Calculating the Mass Loading
of Me la 3 Contaminants to a Marine Embaysnent:
ASARCO Superfund Site, Tacoma, WA
By Gayle Garman and ASARCO Sediments/Groundwater Task Force
INTRODUCTION
The Asarco Superfund Site is located along the southern shore of Commencement Bay, an
industrialized marine embayment in southeastern Puget Sound, approximately 30 miles south of
Seattle, WA. The first industries on the site were sawmills that deposited woodwaste along the
shoreline. A lead smelter that began operations in 1890, was purchased by Asarco in 1905 to process
copper ore from other locations. By-products of copper smelting were further refined to produce
additional products, including arsenic, sulfuric acid, liquid sulfur dioxide, and slag. Smelter operations
ended in 1985 (Hydrometrics, 1996).
Arsenic, cadmium, copper, lead, zinc, and other trace elements were released into soil, air, and
surface water as a result of the smelting and refining operations. There are six upland source areas
where the highest measured concentrations of contaminants in soils are found: The Stack Hill area,
Cooling Pond area, Arsenic Kitchen area, Copper Refinery area, the Fine Ore Bins building, and the
Southeast Plant/DMA area where sulfuric acid spills were frequent. Metals from soil releases and from
slag have migrated to groundwater at the Site (Hydrometrics, 1996).
Many of the smelter buildings and structures are on slag fill. In addition, Asarco extended the
existing shoreline by pouring molten slag into Commencement Bay. The upland area consists of both
gradual and steep slopes extending down to the slag filled shoreline, where slag bluffs extend as much
as 30 ft above the natural sandy substrate. These slag bluffs are very porous, and are subject to twice
daily tides that fluctuate up to 12 ft. vertical (Cross-Section D-D').
The adjacent 23-acre Breakwater Peninsula is composed of massive and granulated slag that were
placed into Commencement Bay between 1917 and 1970. An estimated 15 million tons of slag exist at
the smelter property and slag peninsula (Hydrometrics, 1996).
SEDIMENTS/GROUNDWATER EVALUATION
Sediments as far as 1,000 ft from shore exhibit toxicity in bioassays, and are being evaluated for
remedial action. The preferred alternative is to place a clean sand cap over the contaminated sediment
to prohibit slag contact with marine organisms (Parametrix 1996). Reviewers of the proposed remedial
action asked whether metal contaminants in site groundwater would recontaminate the clean cap. The
Asarco Sediments Groundwater Task Force (ASGTF) was organized to evaluate this question.
Upland geological cross-sections were extended to the shoreline based on boring logs for nearshore
monitoring wells and offshore cores. These cross-sections indicated that the slag formation did not
discharge through sediment to the Bay, but rather, discharged directly to the Bay surface water. In
order to assess the influence of the tides on the hydrologic parameters of the site, 15 slag wells were
monitored over four complete tidal cycles in January 1998 and a multi-well pump-test was conducted
81
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
at new slag well MW-206. This new data was combined with slug-test data from the upland Remedial
Investigation. The distribution of hydraulic conductivities suggested the slag was characterized by four
corridors, indicated as A, B, C and D on the Figures. Later, corridor A was subdivided into corridors
Al and A2 (Figure 1). Discharge rates were calculated for the slag aquifer and the underlying marine
sand aquifer in each corridor by using site data and Darcy's law (ASGTF Group 1,1998):
Q = Kibw
Where:
Q = groundwater discharge rate (ft3/day)
K = hydraulic conductivity (ft/day)
i = hydraulic gradient (ft/ft)
b = aquifer thickness at the shoreline (ft)
w = width of groundwater flow path (ft)
APPROACH TO CONTAMINANT FLUX ESTIMATION
The ASGTF recognized that Darcy's law "provides an estimate of the net groundwater flow
discharging from the site to Commencement Bay." This net groundwater flow originates as recharge in
upgradient water-bearing zones, as infiltration of surface water run-on, and as precipitation onto the
slag. While the twice daily tides, with ranges to 12 vertical feet, are known to cause recurrent inflow
and outflow of seawater in nearshore areas of both the slag and marine sand aquifers, it is assumed that
the net tidal flow is zero (ASGTF, Group 1 Memo).
The mass flux of a contaminant is calculated by multiplying the groundwater discharge rate by the
contaminant concentration. However, when (fresh) groundwater mixes with seawater, there are
changes in geochemistry that alter the solubility, and consequently the mobility, of the metal
contaminants of concern at this site.
Arsenic (As) is the primary contaminant of concern in upland areas of the site. Dissolved arsenic
concentrations to 30 rng/4 have been measured in groundwater near the Fine Ore Bins. The chronic
marine AWQC (Ambient Water Quality Criterion) for arsenic is 0.036 mg/i
DATA AND FIGURES 1
Data are collected at site monitoring wells each Spring and Fall. The ASGTF used data from
March 1994 through September 1998, thus, the Figures show the mean of ten measurements for each
parameter. The Figures are taken directly from the ASGTF Group 4 Technical Memorandum,
(December 1998) and consequently, are not numbered sequentially in this presentation. The
contaminant isopleths were drawn by hand.
ARSENIC ATTENUATION
Figure 10 shows the intrusion of seawater (chloride) for hundreds of feet into the slag formation
along the shoreline. Chloride in upland groundwater is negligible. The landward intrusion of seawater
into the slag is least in corridor D and greatest in corridor Al. The chloride concentrations in both deep
and shallow wells on the breakwater peninsula approximate the chloride concentrations in
Commencement Bay surface water.
1 Editor's note: Figures follow the text. Figures are not consecutively numbered.
82
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Figure 2 shows that the arsenic concentration is an order of magnitude greater in the deeper
breakwater wells (B) than in the shallow breakwater wells (A). The breakwater peninsula is composed
entirely of smelter slag. The ASGTF concluded that the lower oxygen exchange capacity for water
deep within the breakwater peninsula, and the associated lowered redox condition, increase the
solubility of slag arsenic.
Figure 11 shows that dissolved oxygen, in general, diminishes in proportion to distance from the
shoreline and more rapidly in the less permeable corridors, e.g., Corridor D. However, it is difficult to
get accurate field measurements of dissolved oxygen, so the distribution of manganese, which rapidly
precipitates in the presence of dissolved oxygen, also was evaluated.
Figure 6 shows the distribution of dissolved manganese which corroborates the mechanism of
arsenic precipitation, described below. Manganese concentrations decline by an order of magnitude as
site groundwater approaches the shoreline and mixes with oxygen-rich seawater that has intruded into
shoreline slag.
In upland areas of the site (not the slag peninsula, which is entirely slag), the greatest groundwater
flow occurs in the slag formation, which is above the natural geologic formations. The slag, in turn, is
topped by a thin layer of filled soil. Upland groundwater has low oxygen content relative to seawater,
so the geochemistry of the upland groundwater is reducing in comparison to the water of
Commencement Bay. At the shoreline, the tides of Commencement Bay enter the porous seaward face
of the slag formation, forcing seawater into the slag. Thus, as the upland reduced groundwater migrates
toward the shoreline, it gains oxygen by mixing with tidal seawater within the slag. The solubility of
the arsenic then decreases, and most of the dissolved arsenic is precipitated as secondary minerals in
the slag and does not discharge to Commencement Bay. The presence of secondary arsenic minerals
has been confirmed by a mineralogic study of material recovered when MW-206 was installed (US
EPA, 1998). Thus, the changing redox condition of the groundwater explains the attenuation of the
primary contaminant of concern, arsenic. However, understanding the mechanism that controls arsenic
solubility does not answer the question of how to calculate the mass flux of arsenic to the Bay (ASTGF
1998, Group 4).
RELEASE OF COPPER
Figure 3 shows an area in Corridor D where the average dissolved copper concentration in
groundwater is greater than in any other nearshore area. Unlike arsenic, copper is generally more
soluble when there is more dissolved oxygen. However, the shallow (A) wells on the Breakwater
Peninsula, where oxygen is available from seawater and atmospheric exchange, do not have copper
concentrations as great as the wells in Corridor D. The ASGTF concluded there must be another
geochemical parameter causing copper to dissolve from slag in corridor D (ASGTF 1998, Group 4).
Figure 14 shows that acidity may be controlling copper concentrations in Corridor D. Acidity is
measured in logarithmic pH units. The pH of Commencement Bay water is about 8.0. Wells in the
southeast plant/DMA area have average pH values less than 6.0, indicating acid concentrations two
orders of magnitude greater than Commencement Bay. The southeast plant/DMA area is the location
of previous liquid sulfur dioxide and sulfuric acid manufacture. Materials remaining at this location
apparently continue to acidify the groundwater, releasing copper from the slag matrix. Even the
intrusion of seawater does not overcome this effect, as the Corridor D well nearest the shoreline has an
average dissolved copper concentration of 3.6 mg/0, more than an order of magnitude greater than any
other shoreline well. By comparison, the copper acute marine AWQC is 0.0029 mg/£ (a chronic copper
marine AWQC has not been adopted).
83
-------
Proceedings of the Ground-Water/Surface- Water Interactions Workshop
July 2000
Figure 13 shows the distribution of dissolved iron in shoreline wells, which helps corroborate the
mechanism of copper solubility. Like copper, iron is more soluble in a low pH (acid) environment
(e.g., landfill leachate). Higher concentrations of iron are found in the Southeast Plant/DMA source
area where pH was low and copper was high. Thus, both copper and iron exhibit increased solubility
here because of the lower pH (higher acidity). However, unlike copper, the average dissolved iron
concentrations are quite similar in all the wells closest to the shoreline (ASGTF 1998, Group 4). This
suggests that when the dissolved copper and iron in acidified groundwater in corridor D encounter
intruding seawater within the shoreline slag, that the reaction of iron with the oxygen and alkalinity of
the seawater forming an iron precipitate is more rapid than the similar reaction of copper.
CONCLUSIONS
The wells closest to the shoreline have mean dissolved copper concentrations at least three orders
of magnitude greater than the marine acute/chronic AWQC of 0.0029 mg/i The greatest volume of
groundwater discharges through the fractured slag into the marine water column rather than through
contaminated subtidal sediment. Thus, the groundwater process that is having the greatest effect on
marine biota is probably copper discharge to water, not arsenic discharge to sediment.
Mass loading of contaminants in general is a simple calculation that multiplies the average
contaminant concentration by the corresponding average water (volume) discharge rate. This calcula-
tion is valid as long as the contaminant concentration is the concentration in the volume of water that is
discharged. The calculation of contaminant mass loading rates to Commencement Bay is complicated
by the geochemical changes that occur in the shoreline slag as the fresh groundwater mixes with
marine surface water, altering the solubility of the metal contaminants; and by the difficulty in
determining the corresponding volume (discharge rate) of water, which is influenced at the shoreline
by the twice daily tidal flux. Wells nearest the shoreline, where contaminant concentrations are most
representative of discharges to the Bay, are affected by the influx and efflux of tidal water, i.e., the
volume of discharging water characterized by the contaminant concentration measured in the shoreline
wells, is likely greater than the net groundwater flux from the site.
The ASGTF has not yet found a satisfactory method for resolving this problem.
REFERENCES
ASGTF Group 1, 1998. Draft Technical Memorandum on Groundwater Discharge Estimates,
November 1998.25 pp. plus attachments A and B.
ASGTF Group 4, 1998. Draft Technical Memorandum for the Asarco Sediments/Groundwater Task
Force, December 1998. 30 pp. plus Figures.
US EPA, 1998. Mineralogical Study of Borehole MW-206, Asarco Smelter Site, Tacoma, Washington.
US Environmental Protection Agency Region 10, Office of Environmental Assessment, Seattle,
WA, 31 pp. plus Appendices A-D.
Hydrometrics, 1996. Tacoma Smelter Post-Remediation Surface Water Evaluation and Technical
Impracticability Demonstration. Draft Revision 1 for ASARCO, Inc., Tacoma, WA, June 1996.
Parametrix, Inc. 1996. Asarco Sediments Superfund Site Expanded Remedial Investigation and
Feasibility Study, Phase 2 Refinement of Options Report, Volume I. For ASARCO, Inc., Tacoma,
WA, December 1996.
84
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
AUTHOR INFORMATION
Gayle Garman, NOAA Office of Response and Restoration, Coastal Protection and Restoration
Division, 7600 Sand Point Way, NE, Seattle, WA 98115-0070. gayle. garman @noaa. gov
MEMBERS OF THE ASARCO SEDIMENT/GROUNDWATER TASK FORCE (ASGTF)
Marian Abbett, State of Washington, Department of Ecology, Olympia, WA
Thomas Aldrich, Asarco, Inc., Tacoma, WA
Brace Cochran, State of Washington, Department of Ecology, Olympia, WA, David Frank, US EPA,
Seattle, WA
Gayle Garman, NOAA, CPRD, Seattle, WA
James Good, Parametrix, Inc., Kirkland, WA
Douglas Moisten, CH2M-HU1, Inc., Bellevue, WA
Lee Marshall, US EPA, Seattle, WA
Scott Mason, Hydrometrics, Inc., Kalispell, MT
Roger McGinnis, Roy F. Weston, Inc., Seattle, WA
Robert Miller, Hydrometrics, Inc., Tacoma, WA
David Nation, Hydrometrics, Inc., Tacoma, WA
Karen Stash, Roy F. Weston, Inc., Seattle, WA
Carl Stivers, Parametrix, Inc., Kirkland, WA
Donald Weitkamp, Parametrix, Inc., Kirkland, WA
Bernie Zavala, U.S. EPA, Seattle, WA
85
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
fe
o
o
w
to
o
K
o
PH
H
O
S5
O
O
Conceptual Cross-Section D1 -D11
86
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
^
Figure 1
87
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Figure 2
88
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
I
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
^^
D!t (J
Figure 6
90
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Figure 10
91
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
p
|sJ
s§
Figure 11
92
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Figure 13
93
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
,~V»tMtXXKtW::^Sai^faM!f_\ f -jKtt««ttHi,jW .
ii
Figure 14
94
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
The Interaction of Ground Water and Surface
Water within Fall Chinook Salmon Spawning
Areas in the Hanford Reach of the Columbia
River
By David R. Geist
INTRODUCTION
The Hanford Reach is the last unimpounded section of the mainstem Columbia River in the United
States and supports a large run of fall chinook salmon (Oncorhynchus tshawytscha) that returns there
annually to spawn (Dauble and Watson 1997). Previous studies have shown that adult salmon
repeatedly spawn in definite locations within the Reach (Geist 1999; Geist and Dauble 1998; Dauble
and Watson 1997), but the physical characteristics associated with these areas are variable and poorly
understood. More information on the spawning habitat characteristics of fall chinook salmon that
utilize large rivers is needed to recover stocks listed on the Endangered Species Act.
The association between fall chinook salmon spawning and physical habitat characteristics was
previously examined in the Hanford Reach at Locke Island and Wooded Island (Geist 1999). Although
the physical habitat characteristics, e.g., depth, substrate, and water velocity, at the two sites were
similar, only the Locke Island site had extensive salmon spawning. Additional measurements were
taken to determine if the interaction of ground water and surface water within the hyporheic zone could
explain this discrepancy in habitat use between the two sites. Hyporheic discharge was assumed to
affect spawning site selection by providing cues (chemical, temperature, and physical) for pre-
spawning adults to locate spawning reaches (usually 2 to 5 km in length). Once these reaches were
"discovered," hyporheic discharge was assumed to correlate with the distribution of redd clusters (500
to 800 m in length, 120 m in width; Geist 1999) within these river reaches
METHODS
During the fall chinook salmon spawning seasons (October and November; Dauble and Watson
1997) from 1995 to 1997, mini-piezometers (Lee and Cherry 1979) and internal-drive-rod piezometers
(Geist, et al. 1998) were installed within the two sites. Piezometers were installed within the river
channel in groups of three or four, and hyporheic water within thejpiezometers was sampled 2 to 7
times each year for specific conductance (|aS/cm at 25 °C), water temperature (T, °C), dissolved
oxygen (DO, mg/4), and hydraulic head (h, cm). These same parameters were also measured on a
contiguous river sample.
It was assumed that water discharging from the hyporheic zone into the river was a combination of
ground water and surface water. Specific conductance was the primary measure used to differentiate
undiluted ground .water from surface water; specific conductance of undiluted ground water adjacent to
the Hanford Reach averages 300-400 uS/cm while the river water averages -150 uS/cm. The
differences in temperature (T) and hydraulic head (h) were based on the piezometer reading (hyporheic
water) minus the reading from its paired river sample. Differences in hydraulic head between the river
and hyporheic waters were used to calculate a vertical hydraulic gradient (VHG) between the two
95
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
(VHG = h/depth of piezometer). Slug tests were used to estimate the volume of hyporheic discharge
from the sediments into the river channel.
RESULTS AND DISCUSSION
A.
i
o
1
0 08 -r
0.06 -
0.04 -
0.02 -
0.00 -
-0.02 -
: i
I
i
|
!
<
I
Locke Wooded
The results showed that fall chinook salmon
spawning locations were highly correlated with
hyporheic discharge that was composed of mostly
river water and not undiluted ground water.
Hyporheic water that discharged into fall chinook
salmon spawning locations was consistently
greater in magnitude, and had higher dissolved
oxygen and lower specific conductance than
discharge into non-spawning locations. However,
there was no significant difference in temperature
between hyporheic and river water. These results
were true when comparisons were performed
between Locke Island (spawning site) and Wooded
Island (non-spawning site) (Figure 1), and also true
when spawning and non-spawning clusters within
the Locke Island site were evaluated (Figure 2).
Slug tests showed that substrate permeability
decreased with increasing distance below the river
bed at Wooded Island but'did not change over the
depths monitored at Locke Island (Figure 3). This
suggested the mixing zone where river water
penetrated into the river bed was greater within the
spawning site than within the non-spawning site.
Specific discharge calculations gave an average
flux out of the sediments on the order of 9.0 x 10"4
cm/s at Locke Island and 3.0 x 10"4 cm/s at
Wooded Island. Thus, specific discharge of
hyporheic waters was approximately 3 times larger
at Locke Island than Wooded Island.
River water was presumed to have entered
highly permeable riverbed substrate at locations
upstream of spawning areas. Geomorphic bed
features (i.e., islands, gravel bars, riffles) of
alluvial rivers are able to create hydraulic gradients
sufficient to direct surface water into the bed
(Stanford, et al. 1996; Brunke and Gonser 1997).
River water is able to penetrate deeper into
hyporheic habitats if the riverbed is composed of
alluvium that is highly permeable (Vaux 1962,
1968; White 1993). The more permeable the
alluvium, the more that the physiochemical characteristics of the hyporheic waters will resemble
surface water rather than ground water. In contrast, the relative proportion of phreatic ground water in
hyporheic waters will be greater if the riverbed sediments are of low hydraulic permeability because
CD
Specific conduc. ( /jS/cm)
225.0
200.0
175.0
150.0
125.0
; l
; (i
: "
River
Locke Wooded
C.
G"
LT
I
I
Q.
(U
<
0 3
0.2 -
0.1 -
00 -
-0.1 -
-02 -
-03 -
; T
: -r
: " i
: ( i
: -L
Locke Wooded
Figure 1. Physiochemical data collected from
piezometers installed and monitored at the Locke Island
represent the 95% confidence interval of the mean. (A)
Vertical hydraulic gradient (VHG) between hyporheic and
surface waters where positive values indicated potential
upwelling and negative values downwelling, (B) specific
conductance of hyporheic and surface waters, and (C)
differences in water temperatures of hyporheic and
surface waters.
96
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
A.
0.10
"i" o.os
f 0.06
0 °'04
2; 0.02
0.00
T
~
J_
96 S 96
[
nd nd
NS 97
S 97 NS
B.
j 225
^.200
o' 175
1 150
0
o 100
CL
03
.0 -,
0 -
o
n
Q
: T
: River i
; e s.
96 S 96 NS 97 S 97 NS
C.
0 2'° T
LT
!g 15
•c
o 10
a.
-c 05
t oo -
<] -0.5 -
96 S 96
i
1
I
1
NS 97 S 97
!
i
i
L. j
NS
Figure 2. Physiochemical data collected from
piezometers installed and monitored within the Locke
Island site at spawning (S) and non-spawning (NS) sites
during October and November, 1996 and 1997. Bars
above and below the points represent the 95% confidence
interval of the mean. ND = no data.
• Locke Island
o W coded Island
so 100 iso
Piezometer depth (cm)
aoo
Figure 3. Recovery time to 37% of the initial hydraulic head
following a slug-test within piezometers installed at Locke
Island and Wooded Island. The trend line at the Wooded
Island site (dashed line) was significant (P = 0.02, ^=0.78) but
not significant at Locke Island (P = 0.42, r*= 0.08).
river water will not be able to readily enter the
substrate and dilute the ground water (White 1993;
Brunke and Gonser 1997). I concluded river water
that became entrained into the "hyporheic
corridor" had a strong influence on vertical
hydraulic gradients and influenced the use of
salmon spawning habitat. Knowledge of the three-
dimensional connectivity between rivers and
ground water within the hyporheic zone can be
used to improve the definition of fall chinook
salmon spawning habitat.
REFERENCES
Brunke, M. and T. Gonser, 1997. The ecological
significance of exchange processes between
rivers and groundwater. Freshwater Biology v.
37, pp. 1-33.
Bauble, D.D., and D.G. Watson, 1997. Status of
fall chinook salmon populations in the mid-
Columbia River, 1948-1992, North American
Journal of Fisheries Management, v. 17, pp.
283-300.
Geist, D.R., 1999. Redd site selection and
spawning habitat use by fall chinook salmon.
Ph.D. Dissertation, Oregon State University,
Corvallis, Oregon.
Geist, D.R., and D.D. Dauble, 1998. Redd site
selection and spawning habitat use by fall
chinook salmon: the importance of geomorphic
features in large rivers, Environmental
Management, v. 22, pp. 655-669.
Geist, D.R., M.C. Joy, D.R. Lee, and T. Gonser,
1998. A method for installing piezometers in
large cobble-bed rivers, Ground Water
Monitoring and Remediation, v. 18, pp. 78-
82.
Lee, D.R., and J.A. Cherry. 1978. A field
exercise on groundwater flow using
seepage meters and mini-piezometers,
Journal of Geological Education, v.27,
pp. 6-10.
Stanford, J.A., J.V. Ward, W.J. Liss, C.A.
Frissell, R.N. Williams, J.A.
Lichatowich, and C.C. Coutant, 1996. A
general protocol for restoration of
regulated rivers, Regulated Rivers:
Research and Management, v. 12, pp.
391-413.
97
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Vaux, W.G., 1962. Interchange of stream and intragravel water in a salmon spawning riffle. Special
Scientific Report—Fisheries, No. 405. U.S. Fish and Wildlife Service, Bureau of Commercial
Fisheries, Washington, DC
Vaux, W.G., 1968. Intragravel flow and interchange of water in a streambed. Fishery Bulletin 66:479-
489.White, D.S. 1993. Perspectives on defining and delineating hyporheic zones, Journal of the
North American Benthological Society, v. 12, pp. 61-69.
AUTHOR INFORMATION
David R. Geist, Ecology Group, Pacific Northwest National Laboratory, MS K6-85, Post Office Box
999, Richland, Washington, 99352; 509-372-0590; fax: 509/372-3515; david.geist@PNL.gov.
98
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
integrated Acoustic Mapping of Surface Waters:
Implications for Ground-Water/Surface-Water
Linkages
By Chad P. Gubala, Ullrich Krull, Joseph M. Eilers, Mike Montoya, and Jeff
Condiotty
The study of aquatic systems has historically been approached in a traditional scientific manner.
"Representative" sections or components of lakes and rivers have been examined intensively through a
combination of laborious sampling methods. Broader assessments of specific aquatic ecosystems have
then been statistically constructed through the assembly of discrete study elements. Changes in aquatic
ecosystems have then been documented by repeating a similar regimen of sampling at varying time
intervals. Aquatic ecosystem analyses and risk-based management plans have been developed on the
basis of discreet and/or empirical numeric models of aquatic ecosystems, deriving from the original
field investigations.
The efficacy of aquatic assessments and/or risk-based management plans depends upon the
completeness and accuracy of the original data collection and analysis scheme. In order to assemble an
accurate model of an entire aquatic ecosystem, data must be collected in a manner that minimizes the
major components of uncertainty: measurement, spatial and temporal. Most researchers have been able
to adequately minimize measurement error throughout intensive, small-scale research studies or
monitoring exercises. However, precise and accurate measurements distributed over a small section of
a large domain frequently lead to inaccurate conclusions. This phenomenon derives from the
uncertainty of interpolating the conditions of an unknown domain, such as a river reach or lake region,
through interpolation or extrapolation from a limited data-base.
A need exists to develop better monitoring techniques for the dynamic management of aquatic
ecosystems. Combinations of current and emerging technologies, drawn from a variety of application
areas may provide for faster, more cost-effective means of acquiring aquatic systems data and
information. Linking mobile sensors such as hydroacoustic arrays with Global Positioning System
(GPS) navigation have already yielded effective methods for rapidly delineating the bathymetric,
morphometric and hydrologic features of lakes and rivers. Expansion of the role of acoustics has also
permitted the spatial analysis of aquatic biological communities within complex spatial domains. The
coupling of standard aquatic sensor arrays, such as temperature, conductivity or velocity probes, with
GPS will provide multiple parameters for a system in a cost-effective manner. Integration of advanced
sensor probes, such as real-time DNA detectors for identification of aquatic microorganisms will also
greatly enhance the ability to detect and manage change in aquatic ecosystems.
AUTHOR INFORMATION
Chad P. Gubala, SATL (The Scientific Assessment Technologies Laboratory),The University of
Toronto at Mississauga, 3359 Mississauga Road North, Mississauga, ON L5L 1C6; 905-828-3863
(voice); cgubala@credit.erin.utoronto.ca.
Ullrich Krull, University of Toronto, Joseph M. Eilers, JC Headwaters, Inc.
Mike Montoya, Ute Tribal Nation, Jeff Condiotty, Simrad, Inc.
99
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Delineation of VOC-Contaminated Groundwater
Discharge Zone, St. Joseph River, Elkhart,
Indiana
By John H. Guswa, Jonathan R. Bridge, and Michael J. Jordan
A hydrogeologic study was conducted to locate and delineate the portion of the St. Joseph river
within which VOC contaminated groundwater observed in monitoring wells in the study area was
discharging. The principal groundwater contaminants are trichloroethene (TCE) and carbon
tetrachloride (CO,,). Water samples were collected at a depth of approximately two to five feet below
the river bed using a GeoProbe® from a pontoon boat. The samples were analyzed for the purpose of
delineating VOC concentrations in groundwater directly beneath the river bed. The results of this
hydrogeologic study were used to select sampling locations for a benthic macroinvertebrate
investigation to determine if there were any ecological effects resulting from the discharge of VOC-
contaminated groundwater to the river.
Groundwater and surface water samples were initially collected from selected wells and sampling
stations on the river and analyzed for the inorganic analytes listed in Table 1. The purpose of this
sampling was to identify whether there were inorganic analytes that could be used as "tracers" to
ensure that the samples collected from beneath the river bed were groundwater samples and not
induced river water. The concentrations of none of the inorganic analytes proved to be consistently
different between the surface water samples and the groundwater samples. Therefore this group of
inorganic analytes could not be used as "tracers." Other field measured parameters, in particular pH,
temperature and specific conductance were more useful in this regard, and were used for that purpose.
The pH of the river water was approximately one pH unit higher than the groundwater. The river water
temperature was approximately 8°C higher than the groundwater temperature and the specific
conductance of the groundwater was generally higher than the river water.
Table 1. Results of preliminary inorganic analyses, in mg/i
Sampling
Location
CATIONS
calcium
magnesium
sodium
potassium
ANIONS
bicarbonate
carbonate
sulfate
Groundwater
MW-7S
MW-7D
MW-8S
MW-8D
MW-9
MW-10S
94.3
89.9
76.8
81.0
75.9
82.8
25.6
24.3
18.6
19.9
19.5
20.1
7.7
31.7
22.2
31.4
ND
14.5
ND*
ND
ND
ND
ND
ND
280
230
230
230
210
230
ND
ND
ND
ND
ND
ND
28.7
41.2
20.3
26.3
10.9
28.6
Surface Water
SW-01
SW-02
SW-03
83.0
81.1
66.5
23.4
22.8
22.7
15.8
13.7
21.1
ND
ND
ND
180
220
210
ND
ND
ND
2.7
43.5
270
*ND= Not Detected
100
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Groundwater samples were collected from beneath the river bed at 73 locations located along 19
transects. The initial sampling locations were based on evaluation of water quality data from
upgradient monitoring wells and evaluation of groundwater flow direction. Subsequent transect and
sampling locations were selected by reviewing daily analytical results received from an on-site field
laboratory. The water samples were collected by driving a GeoProbe®, from a pontoon boat,
approximately two to five feet below the bed of the river. The one foot long GeoProbe® screen was
then exposed, and water was pumped to the surface using a peristaltic pump. The GeoProbe® was fitted
with a thermocouple to permit in-situ measurement of groundwater temperature. A schematic of the
GeoProbe® sampling device is shown on Figure 1. At some locations the GeoProbe® screen became
clogged with fine sediment from the river bed. When this happened the screen was flushed with
deionized water to clear the screen. The sampling then proceeded using a low-flow sampling protocol.
GEOPROBE®SCREEN
POINT GROUNDWATEI
SAMPLER
Figure 1. Schematic of sub-riverbed sampling equipment
During pumping, the water quality parameters pH, Eh, specific conductance, dissolved oxygen,
temperature, and turbidity were measured in the field. The pH, Eh, specific conductance dissolved
oxygen and temperature were measured utilizing a flow through cell with a YSI model 6820 multi
parameters instrument. Turbidity was measured using an HF model DRT-15CE turbidity meter. A
surface water sample at each sampling location was also analyzed for the field water quality
parameters. The sampling point was purged until the field water quality parameters stabilized. After
the field parameters stabilized the surface water results and the stabilized results from the sampling
point were compared to be certain that groundwater, and not surface water, was being sampled. The
field analyses indicated that pH, temperature and specific conductance were good indicator parameters
for this comparison. The collected sub-river bed samples were then analyzed for the nine VOCs listed
in Table 2. VOC analyses were performed in the field laboratory using a modification of EPA Method
SW-846 8021.
101
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Table 2. VOC Analytes
Trichloroethene
Carbon Tetrachloride
1,1,1 -Trichloroethane
Chloroform
Vinyl Chloride
Tetrachloroethene
1,1 -Dichloroethene
Chloromethane
1,2-Dichloroethene
The results of the sampHng and analyses indicated that TCE contaminated groundwater is
discharging into the river along a 5'500 foot length of the river. The maximum TCE concentration
detected was 1'600 micrograms per liter (ng/0). Groundwater containing CCL4 is discharging to the
river along a 2'500 foot length of the river, and the CC14 discharge zone is contained within the TCE
discharge zone. The maximum CCL4 concentration was 940 [ig/l The sampling locations, and the TCE
and CC14 distribution in the sub-riverbed groundwater, are shown on Figure 2. Based on the results of
this investigation, sampling locations for a benthic macroinvertebrate investigation were selected.
ExpLAHATlOH
SO*. BORBIG t»S (GPS
SJmtrED LOCADOH 1935))
PHASE I ECOLOGICAL ASSESSMENT
SMIPIE LOCAIOI
OSOUKCE ZWC FOR
CftOUCXATtR OMAUSWKO WITH
Iff M« CCH
CCQWtCCZOC FOR
cfiou««rra
KE
Figure 2. Samph'ng grid and contaminant discharge area
AUTHOR INFORMATION
John H. Guswa, Jonathan R. Bridge, and Michael J. Jordan, HSI GeoTrans, Inc. 6 Lancaster
County Road, Harvard, MA 01451
102
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Measuring Enhanced Removal of Dissolved
Contaminants in Hyporheic Zones and
Characterizing Causes and Consequences for
Water Quality
By Judson W. Harvey, Christopher C. Fuller, and Martha H. Conklin
ABSTRACT
Characterizing both the causes and consequences of enhanced oxidation of dissolved manganese
(Mn) in the hyporheic zone at Final Creek basin, AZ required measurements with spatial resolution
varying across five orders of magnitude. Our measurements ranged in scale from that of the
fundamental interactions between surface and ground water (centimeters) to the scale of the perennial
stream that receives ground-water discharge from the entire drainage basin (kilometers). Because of
the lower uncertainty of the stream-tracer approach for estimating the average reaction rate, that
method provided the most reliable basin-scale simulation of the effects of enhanced Mn-removal in
hyporheic zones. The stream-tracer characterization alone, however, could not determine that the
removal of manganese was pH-dependent, or even that the reaction occurred in hyporheic zones (as
opposed to slow-moving zones in surface water). Laboratory and in situ measurements within
hyporheic zones provided the crucial evidence to support interpretations about the causal processes.
INTRODUCTION
Hydrologic exchange of streamwater and ground water back and forth across channel beds of
rivers and streams enhances chemical transformations in shallow groundwater beneath the streambed
(hyporheic zone). The hyporheic zone is defined hydrologically by flow paths that route streamwater
temporarily through the subsurface and chemically by subsurface water that can be shown to receive
greater than 10% of its water from the surface (Triska and others, 1993). Steep chemical gradients in
dissolved oxygen, dissolved organic carbon, and pH in hyporheic zones enhance biogeochemically
mediated transformations of solutes, such as nitrification and demtrification (Grimm and Fisher, 1984;
Triska and others, 1993), oxidation of metals (Benner and others, 1995), and biodegradation of volatile
organic compounds (Heekyung and others, 1995). Hyporheic flow paths are typically small in their
spatial dimensions, but if chemical reaction rates are fast enough, and if enough exchange occurs
between flowing water and sediment, then the effects can accumulate downstream and affect water
quality (Harvey and Fuller, 1998).
This short paper considers three types of measurements at different spatial scales of resolution. The
three measurement types are: (1) laboratory-batch experiments that quantify solute-sediment
interactions at the millimeter-scale, i.e. the scale of individual sediment grains, (2) in situ
measurements in hyporheic flow paths at the scale of centimeters beneath the streambed, and (3)
stream-tracer experiments that quantify removal rates at the scale of experimental subreaches in the
perennial stream (approximately 500 meters) or at the scale of the perennial stream that receives
ground-water discharge from the entire drainage basin (3 kilometers).
A number of physical and chemical measurements of the hyporheic zone have been made as part
of our investigations, including the hyporheic-zone depth, hydrologic residence time in the hyporheic
103
-------
Proceedings of the Ground- Water/Surface- Water Interactions Workshop
July 2000
zone, net removal-rate constant for dissolved manganese (Mn), and percent removal of Mn in
hyporheic flow paths. Previously, we found good agreement across scales of measurement based on a
relatively limited data set (Harvey and Fuller, 1998). In this paper, we update with new data the means
and standard deviations for manganese removal-rate constants and compare them among the three
measurement types. Field methods, analyses, and modeling calculations for reach-scale and in-situ
measurements are presented in Harvey and Fuller (1998), Fuller and Harvey (1999), and Duff, et al.,
(1998). Laboratory methods and analyses are given by Marble and others (1999) and Harvey and Fuller
(1998).
COMPARISON OF REMOVAL-RATE CONSTANTS ACROSS SCALES
In situ rate constants (cm-scale) in the hyporheic zone were determined at a total of eleven sites in
1994,1995, and 1997. Rate constants were determined for sub-reaches of the perennial stream (500-m
scale) by averaging results from stream-tracer injections in 1994 and 1995. The basin-scale estimates
(3-km scale) were computed by averaging the mean rate constant from the four subreaches in 1994
with the mean for the five subreaches in 1995. Laboratory rate constants estimates were computed
using data from the subset of unpoisoned experiments conducted between pH 6 and 6.9, which
matches the range of pH's that were measured in situ. We chose the coefficient of variation (standard
deviation divided by the mean) as a measure of uncertainty.
The mean rate constant for the three field estimates (e.g. in situ, sub-reach, and basin-scale) was
approximately 2.3 x 10"4 per second. The mean rate constant determined in the laboratory was
approximately 30% lower. An average rate constant of 2.3 x 10"4 per second for removal of manganese
corresponds to a time constant (inverse of rate constant) of approximately 1.3 hours, which is
comparatively fast in a drainage basin where the hydrologic residence time in surface water of Final
Creek is approximately 1 day. Although the mean estimates for each field technique varied little (4%
coefficient of variation), standard deviations varied by approximately a factor of four. The in situ
estimate of the removal-rate constant was most uncertain with a coefficient of variation of 107%.
Estimates made at the kilometer-scale based using the stream-tracer approach were least uncertain,
with a coefficient of variation equal to 26%. The coefficient of variation for laboratory and sub-reaches
had intermediate values of 84% and 56%, respectively.
DISCUSSION
Rate constants for removal of manganese differed little between laboratory experiments, in-situ
field measurements, and measurements based on stream-tracer experimentation. The advantage of
laboratory experiments was the isolation of the effects of microbial colonies and pH. Marble and
others (1999) discuss pH and other factors affecting Mn-oxidation reactions. One problem of the
batch-laboratory experiments is extrapolating results to sediment-water ratios that more accurately
approximate field conditions. Following Harvey and Fuller (1998), we scaled laboratory-rate constants
by multiplying them times the ratio between the average sediment concentration (grams/liter) in the
streambed at Final Creek and the sediment concentration used in laboratory experiments. That
adjustment assumes that grain-size variations, which are likely to affect sediment-surface area
available for oxidation of manganese, are the same in laboratory experiments and in the streambed.
Another possible problem of the laboratory experiments is controlling for variation in activity levels of
microbial colonies. For example, Marble and others (this volume) report a significant time lag before
removal in Mn begins in sediment samples that were stored before usage in experiments. Either of
those possible problems might explain the lower Mn removal-rate constant compared with in-situ and
stream-tracer estimates
104
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
In situ sampling within hyporheic flow paths addresses the problem of realistic field conditions by
quantifying rates of removal without disturbing the sediments or natural hydrologic fluxes. But this
method has practical limitations, however. In situ sampling has the disadvantage that the
measurements are difficult and time consuming to make in the field, which limits sample sizes. In
addition there is also the problem that ancillary physical and chemical factors cannot be varied except
though careful site selection. The principal advantage of in-situ field measurements is that interactions
between flow and biogeochemical processes are preserved, which potentially could reveal findings that
would be difficult to detect in a laboratory setting.
Stream-tracer experiments provided the most reliable reach-averaged rate constants for modeling
the basin-scale consequences of enhanced chemical reactions in hyporheic zones. Nevertheless, there
remains a major disadvantage of the stream-tracer approach for quantifying hyporheic-zone processes.
On the basis of stream-tracer experiments alone, we cannot be sure that the removal of reactive solutes
actually occurs in hyporheic-zones, or on the leaves of aquatic vegetation in slowly-moving surface
water at channel margins or behind channel obstructions. Another problem with stream-tracer methods
is that the detection sensitivity for hyporheic zones is not equal across the multiple types of hyporheic
zones that may be present in a given system (Harvey and others, 1996). Only direct sampling of
hyporheic zones using in-situ methods can provide the independent confirmation needed to support
physical interpretations at larger spatial scales.
SUMMARY AND CONCLUSION
Mean rate constants for the removal of dissolved manganese agreed closely between three scales of
resolution in the field, ranging from centimeter-scale field measurements acquired in situ in hyporheic
zones to kilometer-scale estimates determined using stream tracers. The laboratory estimate of the Mn
removal-rate constant was approximately 30% lower than field estimates. In situ and laboratory rate
constants had relatively large coefficients of variation (107% and 84%, respectively), which may be
too large to be used reliably in transport simulations. Stream-tracer experiments provided estimates of
the rate constant with lower uncertainties; 56% when averaged at the reach-scale (approximately 500
meters) and 26% when averaged at the basin-scale (3 kilometers). Our experience at Final Creek basin
leads us to conclude that a multi-scale approach is a necessity for characterizing enhanced
biogeochemical reactions in hyporheic zones.
REFERENCES
Benner, S.G., Smart, E.W., and J. N. Moore, 1995. Metal behavior during surface-groundwater
interaction, Silver Bow Creek, Montana, Environmental Science and Technology, v. 29, pp. 1789-
1795.
Duff, J.H., Murphy, R, Fuller, C.C., Triska, F.J., Harvey, J.W:, and A.P. Jackman, 1998. A mini
drivepoint sampler for measuring pore water solute concentrations in the hyporheic zone of sand-
bottom streams, Limnology and Oceanography, v. 43, no. 6, pp. 1378-1383.
Fuller, C.C. and J.W. Harvey, 1999. The effect of trace-metal reactive uptake in the hyporheic zone on
reach-scale metal transport in Final Creek, Arizona, in Morganwalp, D.W. and Buxton, H.T., eds.,
U.S. Geological Survey Toxic Substances Hydrology Program—Proceedings of the Technical
Meeting, Charleston, South Carolina, March 8-12,1999—Volume 1—Contamination from Hard
Rock Mining: U.S. Geological Survey Water-Resources Investigations Report 99-4018A, this
volume.
Grimm, N.B. and S.G. Fisher, 1984. Exchange between interstitial and surface water: implications for
stream metabolism and nutrient cycling, Hydrobiologia, v. Ill, pp. 219-228.
105
-------
Proceedings of the Ground-Water/Surface- Water Interactions Workshop
July 2000
Harvey, J.W. and C.C. Fuller, 1998. Effect of enhanced manganese oxidation in the hyporheic zone on
basin-scale geochemical mass balance, Water Resources Research, v. 34, pp. 623-636.
Harvey, J.W., Wagner, B.J., and Bencala, K.E., 1996. Evaluating the reliability of the stream-tracer
approach to characterize stream-subsurface water exchange, Water Resources Research, v. 32, no
8, pp. 2441-2451.
Heekyung, K., Hemond, H.F., Krumholz, L.R., and B.A. Cohen, 1995. In-situ biodegradation of
toluene in a contaminated stream, 1, Field Studies, Environmental Sciences and Technology, v. 29
pp. 108-116.
Marble, J.C., Corley, T.L., Conklin, M.H., and C.C. Fuller, 1999. Environmental factors affecting
oxidation of manganese in Final Creek, Arizona., in Morganwalp, D.W. and Buxton, H.T., eds.,
U.S. Geological Survey Toxic Substances Hydrology Program—Proceedings of the Technical
Meeting, Charleston, South Carolina, March 8-12, 1999—Volume 1—Contamination from Hard
Rock Mining: U.S. Geological Survey Water-Resources Investigations Report 99-4018A, this
volume.
Marble, J.C., 1998. Biotic contribution of Mn(H) removal at Final Creek, Globe, Arizona, unpublished
M.S. thesis, University of Arizona, Department of Hydrology and Water Resources, Tucson. 91
pp.
Triska, F.J., Duff, J.H., and R.J. Avanzino, 1993. The role of water exchange between a stream
channel and its hyporheic zone on nitrogen cycling at the terrestrial-aquatic interface,
Hydrobiologia, v. 251, pp. 167-184.
106
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Bioassessrnent of Hyporheic Microbial
Communities Using a Specially-designed
Sediment Colonization Chamber
By Susan P. Hendricks (
Streambed sediments are often very heterogeneous in particle size distribution and permeability,
creating mosaics of hyporheic habitats and biotic communities. Biotic patchiness and difficulties
sampling the physical envkonment present challenges for researchers investigating hyporheic
microbial transformation processes. Heterogeneous sediments may not be conducive to conventional
sampling techniques (Fraser and Williams 1997, Mauclaire, et al. 1998). For example, streambeds
composed of large gravel and cobble are prohibitive to mechanical or hand-coring of sediments.
Freeze-core sampling techniques often used in heterogeneous sediments for invertebrates are not
appropriate because freezing alters microbial activity (Humpesch and Niederreiter 1993, Claret 1998a,
1998b).
Some sampling difficulties may be overcome using artificial chambers. Many investigators have
designed chambers for various specific monitoring purposes, including water chemistry,
macroinvertebrates, and in situ microbial metabolism measurements (Danielopol and Niederreiter
1987, Dodds, et al. 1996, Shati, et al. 1997). There also have been several site- and/or question-specific
chamber designs, particularly for sediment microbial studies (e.g., Fischer, et al. 1996, Frazer, et al.
1996, Eisenmann, et al. 1997, Claret 1998a, 1998b). The purpose of this paper is to describe a multi-
purpose sediment microbial colonization chamber that combines attributes of several previous designs.
The chamber has been used successfully in heterogeneous cherty western Kentucky and Tennessee
streams and can be used not only for chemical and microbial monitoring, but also for experimental
manipulations in situ.
DESIGN CONCEPT
Gravel bars are conspicuous geomorphic features of many mid-reach streams (Figure 1).
Conservative tracer experiments in the study streams have shown that both gravel bars and the
hyporheic zone are important transient storage zones where dissolved organic matter, nutrients and
contaminants may be retained for periods of time and transformed before re-entering the surface
stream environment. Microbial activity within these subsurface regions is important in mediating
nutrient and carbon cycling (Hendricks and White 1991,1995; Hendricks 1993, 1996) and potentially
important in transforming contaminants.
The chamber was designed for maximum flexibility in placement and function within both the
hyporheic zone and within gravel bars lateral to the stream (Figure 1). The design (Figure 2) has
allowed us to 1) collect interstitial water samples for nutrient chemistry and dissolved oxygen, 2) sub-
sample sediments for various microbial assays (e.g., bacterial productivity, phosphatase activity), and
3) carry out time-course in situ experiments for determining transformation rates of various substances
(e.g., respiration rates, nitrification rates). Additionally, the sediment volume within the chamber
enables 1) collection of small test-tube sized cores from each of the depth intervals for further
laboratory microbial analyses (e.g., microbial diversity studies using amplified rDNA restriction
analysis, fatty acid analysis, perfusion experiments) and 2) transplant experiments between sites and
streams and monitor subsequent changes in activities and other effects. Our initial experiments have
107
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
been comparisons between an
agriculturally impacted third-order
stream (Ledbetter Creek) and a pristine
third-order stream (Panther Creek)
with similar sediment heterogeneities
and watershed characteristics.
METHODS
Sediments collected from each
stream were sieved (£ 3 mm size
fraction), autoclaved, and placed into
chamber baskets (Figure 2). The
baskets were stacked such that each
represented ahyporheic depth interval
(0-10 cm, 15-25 cm, and 30-40 cm).
Chambers were placed just below the
top of the water table along previously
determined subsurface flow-paths
within the gravel bars and below the
sediment-water interface within
hyporheic regions (Figure 1). Ports
between inner and outer cylinders were
aligned in the open position to allow
interstitial flow and colonization with
microflora for approximately 8-10
weeks.
„ x,
LetibetteM
Cneek
;/ Paste
Ereeft
Following colonization, the Figure 1. Study sites at Ledbetter and Panther Creeks. Black
chamber caps were removed and arrows indicate groundwater inflow: Gray arrows indicate
interstitial water was withdrawn for flowpaths of surface water into gravel bars and hyporheic zone
dissolved O2, nutrients, and DOC from beneath stream.
each of the stacked sediment baskets by attaching a syringe to the tubing embedded in the basket
(Figure 2). Baskets were then retrieved from the inner cylinder by pulling up on the central stem.
Sediment sub-samples were collected from each basket, placed into sterile containers, and transported
to the laboratory for phosphatase activity, bacterial productivity and microbial diversity assays.
In situ experiments also were conducted. For example, interstitial dissolved O2 and NO3-N samples
were collected from each depth interval within the chamber as'pre-incubation references (i.e., ports on
both cylinders aligned in the open position). Twisting the inner cylinder in the opposite direction
resulted in closing of the ports and isolation of sediments from interstitial flow. Following incubation
in the closed position for a chosen period of time (e.g., 24-72 hours), interstitial water was re-sampled.
Oxygen consumption (respiration rate) and NO3-N accumulation (nitrification rate) were estimated as
the difference between dissolved O2 or NO3-N concentrations before and after incubation over time
and expressed as change in concentration g"1 wet weight or I/1 sediment h"1.
108
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Hyporheic/Gravel Bar Sediment Chambe:
Stem to facilitate basket removal
Chambers are buried in
streambad to just below the
sediment surface and in
gravel bars to just below the
top of the water table
Chamber constructed of PVC
pipes - Outer chamber is
casing for inner chamber,
inner chamber contains
sediment baskets made of
screening.
The inner cylinder holding U
sediment baskets can be
turned to align the openings
with outer cylinder exposing
sediments to interstitial floi
or closed off to isolate
sediments from flow.
RESULTS
Some examples of subsurface
microbial activities important in P,
N and C cycles are summarized in
Tables 1-3. The tabulated data are
means calculated from all depths (3)
within replicate chambers (2) within
a site (gravel bar=3, hyporheic=2)
for each stream. Alkaline
phosphatase activity (Sayler, et al.
1979) was higher in the gravel bar
than in the hyporheic zone at
Ledbetter Creek during both summer
and spring sampling periods, and
higher in general than in Panther
Creek (Table 1), indicating
differences in phosphorus demand
by the microbial community
between the two streams and among
sites within the streams.
The Ledbetter Creek gravel bar
generally showed higher bacterial
productivity (methods modified
from Findlay 1993) than the Panther
Creek gravel bar (Table 2). Bacterial
productivity was highest where
interaction between the subsurface
gravel bar and stream surface water
was greatest as indicated by
dissolved O2 gradients in both gravel bars (data not shown) along subsurface flow-paths (Figure 1).
Nitrification rate (methods of Jones, et al. 1995), reported as the increase in ug NO3+NO2 L"1
sediment h'1, was higher in Ledbetter Creek than in Panther Creek (Table 3). The Ledbetter Creek
sediment bacterial community is composted of taxa, which appear adapted to high levels of NH4NO3
fertilizer applications.
Table 1. Alkaline phosphatase activity (APA = jim nitro-phenylphosphate
reduced g"1 sediment dry wt) in hyporheic and gravel bar chambers at
Ledbetter and Panther Creeks. AG = agriculturally impacted, P = pristine.
Figure 2. Schematic of gravel bar/hyporheic chamber design.
Stream
Site
Spring
Summer
Ledbetter (AG)
Gravel Bar
Hyporheic
1020.5
700.9
1942.9
205.4
Panther (P)
Gravel Bar
Hyporheic
687.7
no data
370.9
no data
109
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
My 2000
Table 2. Subsurface gravel bar bacterial productivity (ug C m'2 h'1) at Ledbetter and
Panther Creek as estimated from incorporation of 3H-thymidine into bacterial DNA. AG =
agriculturally impacted, P = pristine.
Gravel Bar Position
Stream
Ledbetter (AG)
Panther (P)
Month
Jan
Jun
Sep
May
Aug
Upstream
3.13
64.5
37.7
72.1
147.0
Mid-Bar
11.7
73.6
67.1
58.6
51.6
Downstream
32.3
225.9
67.1
44.9
48.0
Table 3. Nitrification rates measured as the increase in NO3+NO2 in sediments (ug NO3+NO2
produced I/1 sediment h'1) in Ledbetter and Panther Creek chambers. AG = agriculturally
impacted, P = pristine.
Rate
Stream
Site
ANO3 (ug L'1) % Increase sediment h'1)
Ledbetter(AG)
Panther (P)
Gravel Bar
Hyporheic
Gravel Bar
Hvoorheic
801.2
238
4.5
8.0
+340 •
+260
+5
+8
11.13
3.30
0.06
0 11
DISCUSSION
It is well known that agricultural practices increase sedimentation and greatly alter the chemistry of
surface waters. However, land-use effects on microbial and biogeochemical processes at the
groundwater-surface water interface (hyporheic zone) mediated by increased nutrient, carbon, and
sediment loads are largely unknown. Contaminants reaching streams from subsurface sources such as
groundwater are expected to be processed/transformed at the groundwater-surface water interface
depending on heterogeneity and permeability of sediments and subsurface flow-path complexity.
Methods and data presented here have focused primarily on delineating differences in hyporheic zone
function that mediate agricultural and suburban runoff between and within streams. Results presented
above are limited examples of data which might be obtained from colonization chambers. It is feasible
to examine other processes which indicate disturbance or alteration of function by other contaminants
entering streams from either point or non-point sources (e.g., contaminated groundwaters).
CONCLUSIONS
The sediment microbial colonization chamber described in this paper appears to be a reasonable
device for examining microbial activities and biogeochemical transformations within the hyporheic
zone or at the groundwater-surface water interface within streambeds.
Advantages of using the sediment colonization chamber described here are 1) construction .
materials are inexpensive and the design is flexible for a variety of streambed types, 2) chambers are
multi-purpose in that both interstitial water chemistry and sediment sampling can be done, 3) time
course incubations can be carried out, 4) transplanting of chambers for inter- and intra-site
110
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
comparisons may be carried out, and 5) either natural sediments or more homogeneous artificial
particles (e.g., glass or ceramic beads) may be used in the chambers, 6) replication is quite good and
data are consistent between replicate chambers.
Some disadvantages may include 1) installation that may require 2 or more people, 2) colonization
periods may be long (6 weeks minimum, 8-10 weeks preferred), 3) chamber sediments may not reflect
actual particle size distributions found in streambeds, 4) chambers may prohibit infiltration of natural
CPOM (microbial fuel), and 5) chambers may alter local subsurface hydraulics. Continued monitoring
of the chambers over time and space will help evaluate their ultimate usefulness in stream ecosystems.
REFERENCES
Claret, C., 1998a. A method based on artificial substrates to monitor hyporheic biofilm development,
Internat. Rev. Hydrobiol, v. 83, pp. 135-143.
Claret, C., 1998b. Hyporheic biofilm development on artificial substrata, as a tool for assessing trophic
status of aquatic systems: first results. Annls. Limnol., v. 34, pp. 119-128.
Danielopol, D.L. and R. Niederreiter, 1987. A sampling device for groundwater organisms and oxygen
measurement in multi-level monitoring wells, Stygologia, v. 3, pp. 252-263.
Dodds, W.K., C.A. Randel, and C.C. Edler, 1996. Microcosms for aquifer research: application to
colonization of various sized particles by ground-water microorganisms, Ground Water, v. 34,
pp.756-759.
Eisenmann, H., W. Traunspurger, and E.I. Meyer, 1997. A new device to extract sediment cages
colonized by microfaunafrom coarse gravel river sediments, Arch. Hydrobiol., v.139, pp. 547-561.
Findlay, S., 1993. Thymidine incorporation into DNA as an estimate of sediment bacterial production,
in: Kemp, P., B. Sherr, E. Sherr, and J. Cole (eds), Handbook of methods in aquatic microbial
ecology, Lewis Publishers, Boca Raton, FL.
Fischer, H., M. Pusch, and J. Schwoerbel, 1997. Spatial distribution and respiration of bacteria in
streambed sediments, Arch. Hydrobiol., v.137, pp. 281-300.
Fraser, E.G. and D.D. Williams, 1997. Accuracy and precision in sampling hyporheic fauna, Can. J.
Fish. Aquat. Sci., v. 54, pp. 1135-1141.
Fraser, E.G., D.D. Williams, and K.W.F. Howard. 1996. Monitoring biotic and abiotic processes
across the hyporheic/groundwater interface. Hydrogeology Journal, v. 4, pp. 36-50.
Hendricks, S.P. 1996. Bacterial biomass, activity, and production within the hyporheic zone of a north-
temperate stream, Archivjur Hydrobiologie, v.135, pp. 467-487.
Hendricks, S.P. and D.S. White, 1995. Seasonal biogeochemical patterns in surface water, subsurface
hyporheic, and riparian groundwater in a temperate stream ecosystem, Archivfiir Hydrobiologie, v.
134, pp. 459-490.
Hendricks, S.P. and D.S. White, 1991. Physicochemical patterns within a hyporheic zone of a northern
Michigan river, with comments on surface water patterns, Can. J. Fish. Aquat. Sci., v.48, pp. 1645-
1654.
Hendricks, S.P., 1993. Microbial ecology of the hyporheic zone: a perspective on the integration of
hydrology and biology, J. No. Amer. Benthol. Soc., v. 12, pp. 70-78.
Humpesch, U.K., and R. Niederreiter, 1993. Freeze-core method for sampling the vertical distribution
of the macrozoobenthos in the main channel of a large deep river, the River Danube at river
kilometre 1889, Arch. Hydrobiol. Suppl, v. 101, pp. 87-90.
Jones, J.J., S.G. Fisher, and N.B. Grimm, 1995. Nitrification in the hyporheic zone of a desert stream
ecosystem, /. N. Am. Benthol. Soc., v. 14, pp. 249-258.
Mauclaire, L., P. Marmonier, and J. Gibert, 1998. Sampling water and sediment in interstitial habitats:
a comparison of coring and pumping techniques, Arch. Hydrobiol., v.142, pp. 111-123.
Ill
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Sayler, G.S., M. Puziss, and M. Silver, 1979. Alkaline phosphatase assay for freshwater sediments:
application to perturbed sediment systems, Appl. Environm. Microbiol, v. 38, pp. 922-927.
Shati, M.R., D. Ronen, and R. Mandelbaum, 1996. Method for in situ study of bacterial activity in
aquifers, Environm. Sci. Tech., v. 30, pp. 2646-2653.
AUTHOR INFORMATION
Susan Hendricks, Hancock Biological Station, Murray State University, Murray, KY 42071;
susan.hendricks@murravstate.edu.
112
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Fundamentals of SPSWD Sampling, Performance,
and Comparability to Baomonitoring Organisms
By J.N. Huckins, J.D. Petty, H.F. Prest, J.A. Lebo, C.E. Orazio, J. Eidelberg, W.L.
Cranor, R.W. Gale, and R.C. Clark
INTRODUCTION
Passive monitoring devices have long been used by the chemical industry and governmental
agencies to ensure compliance to OSHA standards for time-weighted-average (TWA) concentrations
of organic vapors in the work environment. With the recent development of lipid-containing
semipermeable membrane devices (SPMDs) and samplers based on diffusive gradients across thin
polymeric films (DGTs), the passive in situ monitoring approach can now be applied to determining
TWA concentrations of both hydrophobic organics (SPMDs) and heavy metals (DGTs) in aquatic
environments (1-3). In this work, we focus on the fundamentals of SPMD technology, and the potential
utility of the approach for monitoring organic contaminants in groundwater (includes the hyporheic
zone).
SPMD SOURCE, DESIGN AND THEORY
The SPMD technology is the subject of two government patents and the devices are
commercially available from Environmental Sampling Technologies, 1717 Commercial Drive, St.
Joseph, MO 64503. A standard SPMD consists of a thin walled (75-95 m) layflat tube of low density
polyethylene (LDPE) containing a thin film of 95% pure triolein (Figure 1). The ends of the LDPE, are
welded by heat-sealing. The length or size of the standard device can be customized to fit an
investigator's sampling needs. The membrane surface area-to-lipid-volume ratio of a standard SPMD
is 450 cm2/m
-------
Proceedings of the Ground- Water/Surface- Water Interactions Workshop
July 2000
constant volume of air or water extracted per unit time (i.e., linear uptake of analytes at constant
ambient concentrations), and the following equation applies
Cw = CSPMD VSPMD/RS t (1)
where Cw is the TWA analyte concentration in water (ng/0), CSPMD is the concentration in the whole
device (membrane + lipid) in ng/«, VSPMD is the volume of the SPMD (L), Rs is the sampling rate for a
chemical in L/d, and t is exposure time in days (d).
Huckins et. al. (4) have shown that R. values for PAHs and OCs ranged from 1 to 8 L/d (exposure
conditions: 10-26 "C, water velocity < 1 cm/sec), when using a standard SPMD with VSPMD = 5mL.
Thus, in the linear region of analyte uptake, an SPMD with VSPMD = 1 mC will daily extract dissolved
contaminants from 200 to 1,600 mfl of water. If the target compounds have relatively low octanol-water
partition coefficients (i.e., log K^ < 4), such as VOCs, equilibrium between the device and the
surrounding water is often achieved in < 1 week. In that case, the following simple model can be used
for water concentration estimates:
Cw = CSPMD/ KSPMD (2)
where KSPMD is the equilibrium SPMD-water partition coefficient. For compounds with log K^ of 4.0
and > 4.0, KSPMD is 0.75 K,,w and 0.3 K^, respectively.
APPLICABILITY OF THE APPROACH
Standard SPMDs are designed to sample nonionic hydrophobic compounds. The total volume of
water extracted by an SPMD at equilibrium is estimated by 0.75 Kow VSPMD (log K^s 4.0) and 0.3 Kow
VSPMD (log KOWS > 4.0). For example, if the target analyte has a K^ of 300 the maximum (equilibrium)
volume of water extracted by a standard 1 m« triolein SPMD (VSPMD 5 mL) is only a little more (i.e.,
1.1L) than a 1L grab sample. Thus, standard SPMDs are useful only for sampling compounds with
K,,ws > 300, unless an adsorbent is dispersed in the triolein (4).
Trace levels of a variety of classes of organic contaminants have been successfully determined in
aquatic systems using SPMDs. These include but are not limited to the following: polycyclic aromatic
hydrocarbons, polychlorinated -biphenyls and -terphenyls, organochlorine pesticides, polychlorinated-
dibenzodioxins and -dibenzofurans, chlorinated and brominated diphenyl ethers, chlorinated -
benzenes, -anisoles and -veratroles, certain alkylated and chlorinated phenols, heterocyclic aromatics,
pyrethroid and nonpolar to moderately polar organophosphate pesticides, and nonionic organometals.
DEPLOYMENT CONSIDERATIONS
Because SPMDs readily sample a broad spectrum of chemicals from air, exposure to organic
vapors is minimized by transport to and from the sampling site in clean gas-tight metal cans. Other
precautions are similar to those used for standard grab sampling methods.
The appropriate exposure duration is dependent on the physicochemical properties (e.g., K^s) of
the target analytes, analytical sensitivity needed, choice of sampling approach (i.e., integrative or
equilibrium), environmental conditions, and the potential for vandalism. When calibration data are
114
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
available (4,6) or can be estimated for contaminants of concern, the following model can be used to
determine the time required to reach half of the SPMD equilibrium concentration (ti/2):
ti/2 - fa 0-5 KSPMD VSPMD/RS
(3)
Knowledge of these half-times or half-lives is useful because the uptakes of analytes are linear during
1 t1/2.
Thus, within 1 11/2 equation 1 can be used for water concentration estimates. If an equilibrium
sampling approach is used, exposure time should be > 4 t^s and equation 2 is applicable. SPMD
sampling rates are affected by temperature, water velocity, and biofouling. Fortunately, temperature,
and to a lesser extent hydraulic conductivity (velocity), are generally more constant in ground water
systems than in surface waters. In surface waters, biofouling impedes analyte uptake, often limiting the
utility of long exposure times. However, biofouling is much reduced in groundwater, thereby
permitting significantly longer exposures (months instead of weeks) for compounds with high Kows
(i.e., 6.0).
Even with the effects of the aforementioned environmental variables on sampling rates, Ellis et. al.
(5) have shown that river water concentrations (dissolved phase) of trace organic contaminants can be
estimated from SPMD concentrations within two-fold accuracy. Huckins, et al. (1,4) have suggested
that the use of permeability reference compounds (PRCs) may further reduce errors in water
concentration estimates. PRCs are analytically noninterfering compounds, such as deuterated PAHs
with log K^s < 5.0, that are added to SPMD lipid before deployment. By determining PRC loss rates
VSPMDor k2 ) from SPMDs during an environmental exposure and comparing them to PRC
k2s measured during the experimental determination of sampling rates (Rss), the laboratory derived Rss
of analytes can be adjusted to account for the effects of field exposure conditions.
COMPARABILITY TO BIOMONITORING ORGANISMS
The accumulations of organic contaminants by SPMDs and aquatic organisms have been compared
in a number of studies (2,4-6). Some studies have shown that the concentration patterns and uptake
rates of several classes of chemicals by SPMDs and fishes are similar (6,7). However, it is unrealistic
to expect SPMDs to mimic the uptake of all organic contaminants by all aquatic species because large
differences exist among species in regard to diet/source of energy, metabolic activities (xenobiotics),
and lipid composition and percentages. For example, few aquatic organisms contain 20% lipid by
weight as SPMDs do. Thus, the capacities of most aquatic organisms to retain accumulated residues
are nearly always less than SPMDs (i.e., organism k2»SPMD k^, which suggests that organisms used
as biomonitors may not retain detectable levels of some residues several days after an episodic
contaminant release.
GROUNDWATER APPLICATION
Although SPMDs are used extensively in surface waters (2,4) and in the atmosphere (8), their
application to ground water systems has been limited to a few studies. This is surprising because
laboratory studies performed to determine SPMD sampling rates (4) more closely simulate the
relatively constant conditions existing in some ground water systems.
Herein, we highlight a pilot study on the use of SPMDs by EPA Region 9 personnel for sampling
dieldrin in groundwater at the George Air Force Base, California. The devices were deployed in two
115
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
ground water monitoring wells where dieldrin residues were previously determined. The exposure
period was 29 days. Based on laboratory calibration studies (4), each standard SPMD sampled a total.
volume of 58 0 of groundwater during the 29-day exposure. Figure 2 illustrates the results of the
analysis (high resolution gas chromatography-electron capture detection [BCD]) of SPMD extracts
from well water at one of the sites, and of an associated SPMD control. Microgram quantities of
dieldrin were concentrated in each of two replicate SPMDs and the SPMD-derived TWA water
concentration was estimated at 69 ng/« (recovery-corrected). Earlier, a grab sample at the same site was
found to contain 110 ng/£ of dieldrin (includes sorbed fraction). Much lower levels (140 to 2,800-fold
less) of fifteen other chlorinated pesticides (e.g., chlordane components, endrin, endosulfan II and
sulfate, etc.) were also detected in the SPMDs. These trace contaminants were not detected using
EPA's (CLP) low-level pesticide method. In summary, SPMDs appear to be well suited for monitoring
trace hydrophobic organics in ground water systems.
Slim *• 2 »t 1 .Q nil.
A
Sit* M 2 at 100 mL
T"|''T"|''T''''^
SPMD Control at 100 ml.
Figure 2. ECD Chromatograms of an SPMD sample extract and a control SPMD. SPMDs were deployed for
29 days in a ground water monitoring well at George Air Force Base, CA. Dieldrin and 15 other organo-
chlorlne pesticides were detected in the sample. The bottom two chromatograms were diluted 100-fold to
keep dieldrin on scale, and octachloronaphthalene (OCN) was used as an instrumental internal standard.
REFERENCES
(1) Huckins, J.N., Manuweera, O.K., Petty, J.D., Mackay, D., and J.A. Lebo, 1993. Environ. Sci.
Technology, v. 27, pp. 2489-2496.
(2) Huckins, J.N., Petty, J.D., Lebo, J.A., Orazio, C.E., Prest, H.F., Tillitt, D.E., Ellis, G.S., Johnson,
B.T., and O.K. Manuweera, 1996. In Techniques In Aquatic Toxicology; O.K. Ostrander (Ed.),
CRC-Lewis Publishers, Boca Raton, FL, pp. 625-655.
(3) Davison, W. and H. Zhang, 1994. Nature, pp. 367,545.
(4) Huckins, J.N., Petty, J.D., Prest, H.F., Orazio, C.E., and R.C. Clark, 1999. Guide for the Use of
Semipermeable Membrane Devices (SPMDs) as Samplers of Waterborne Hydrophobic Organic
Contaminants, Reports for the American Petroleum Institute (API), API: 1220
(5) Ellis, G.S., Huckins, J.N., Rostad, C.E., Schmitt, C.J., Petty, J.D., and P. MacCarthy, 1995.
Environ. Toxicol. Chem., v.14, pp. 1875-1884.
116
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
(6) Meadows, J.C., Echols, K.R., Huckins, J.N., Borsuk, F.A., Carline, R.F., andD.E. Tillitt, 1998.
Environ. Sci. Technol., v. 32, pp. 1847-1852.
(7) Peven, C.S., Uhler, A.D., and FJ. Querzoli, 1996. Environ. Toxicol. Chem., v. 15, pp. 144-149.
(8) Ockenden, W.A., Prest, H.F., Thomas, G.O., Sweetman, A., and K.C. Jones, 1998. Environ. Sci.
Technol., v. 32, pp. 1538-1543.
AUTHOR INFORMATION
J.N. Huckins and J.D. Petty, Columbia Environmental Research Center, BRD, USGS, 4200 New
Haven Road, Columbia, MO. 65201.
H.F. Prest, J.A. Lebo, and C.E. Orazio, Institute of Marine Science, Long Marine Laboratory,
University of California Santa Cruz, Santa Cruz, CA 95060.
J. Eidelberg, W.L. Cranor, R.W. Gale, and R.C. Clark, U.S. EPA, 75 Hawthorne Street, San Francisco,
CA 94105.
117 .
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Acid Mine Drainage—The Role of Science
By Briant Kimball
Thousands of abandoned and inactive mines are located in environmentally sensitive mountain
watersheds. Cost-effective remediation of the effects of metals from mining in these watersheds
requires knowledge of the most significant sources of metals. The significance of a given source not
only depends on the concentration of a toxic metal, but also on the total mass of metal added to the
stream. This discussion deals with accounting for the mass of metal that enters the stream, which is
called the mass loading. It is calculated as the product of metal concentration and stream discharge.
Without discharge measurements, the overall effect of high metal concentrations on streams and
aquatic organisms is unclear. A traditional discharge measurement is obtained by dividing a stream
into srhall sections and measuring cross-sectional area and average water velocity in each section.
Because the channel bottom in mountain streams is not smooth and much of the flow is among the
streambed cobbles, accurate discharge measurements are difficult to obtain, even under the best
conditions.
AN APPROACH FOR MOUNTAIN STREAMS
A recent study by the U.S. Geological Survey Toxic Substances Hydrology Program illustrates a
practical approach to obtaining and using discharge measurements in mountain streams. Chalk Creek,
a tributary of the Arkansas River in Colorado, receives mine drainage from the Golf Tunnel adit.
Metal-rich mine drainage from the Golf Tunnel is routed around waste rock and a capped tailings pile
into a constructed wetland. From the wetland, the mine drainage enters Chalk Creek from small
springs and seeps along the stream. Regulatory and land management agencies have asked three basic
questions about Chalk Creek. First, is there more than one source of mine drainage that affects the
stream? Second, does a remediation plan need to account for drainage from more than one source?
Finally, have past remediation efforts been successful? To address these questions, we employed a
tracer-dilution study to determine discharge and synoptic sampling to obtain detailed chemical
composition from many locations. The synoptic samples are collected during a short period of time,
typically a few hours, providing a "snapshot" of the changes along a stream at a given point in time.
ADDING A TRACER: DISCHARGE BY DILUTION
Discharge in mountain streams can be measured precisely "by adding a dye or salt tracer to a
stream, measuring the dilution of the tracer as it moves downstream, and calculating discharge from
the amount of dilution. Because we know the concentration of the injected tracer and the rate at which
it is added to the stream, we know the mass added to the stream. By measuring the concentration of the
tracer upstream and downstream from the injection point, we can calculate the discharge by dilution of
the tracer in the stream. To define discharge in Chalk Creek, a sodium chloride tracer was added at a
constant rate for 24 hours at a point upstream from the mine drainage. The chloride injection was
monitored at several sites downstream from the injection point, documenting the incremental increase
of discharge due to water entering the stream. The difference in discharge between two stream sites
gives the total amount of inflow from surface- and ground-water sources in that small reach.
118
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
SYNOPTIC SAMPLING: A "SNAPSHOT" IN TIME TO COMPARE METAL LOADING OF SOURCES
Synoptic samples provided metal concentrations, giving a detailed profile of zinc concentrations in
both the stream and inflows along the stream reach. To evaluate these concentrations, a mass-loading
profile was calculated from the concentrations and the discharge values. The concentrations and the
mass-loading profile help answer the basic questions about the sources of metals and the effectiveness
of remediation. First, there appears to be more than one source of mine drainage, because the high
concentration at 252 meters could not be from the Golf Tunnel. Second, despite the higher
concentration of zinc in water from the second source, the loading profile shows that the high-
concentration water only contributes about 8 percent of the zinc load and would not require a separate
remediation plan. Finally, there are still effects on metals in the stream where old tailings were
removed downstream from 300 meters; these effects will likely decrease with time. The example of
Chalk Creek shows that the highest inflow concentrations do not always result in the most significant
sources of metal loading. Our results show that a site can be investigated in great detail to help make
decisions by using tracer injections and synoptic sampling.
AUTHOR INFORMATION
Briant Kimball, U.S. Geological Survey; bkimball@usgs.gov.
(Mr. Kimball's article is adapted, by permission, from an article published by the USGS. in January
1997. More information on the Toxic Substances Hydrology Program can be obtained at
http://toxics.usgs.gov/toxics.)
119
-------
Proceedings of the Ground-Water/Surface- Water Interactions Workshop
July 2000
Temporal and Spatial Trends in Biogeochemical
Conditions at a Groundwater-Surfacewater
Interface
By John M. Lendvay and Peter Adriaens
BACKGROUND
The Bendix, Allied Signal National Priority List (NPL) site at St. Joseph, Michigan, has been
extensively characterized for contaminant distribution and biogeochemical conditions between the
contaminant source and zone of emergence in Lake Michigan [Tiedeman and Gorelick, 1993;
Semprini, et al., 1995; Lendvay, et al., 1998a; Lendvay, et al., 1998b]. The source of the contaminant
plume at the site consists of trichloroethene (TCE) and minor contamination with hydrocarbons.
Contaminant hydrocarbons and natural organic matter have stimulated sufficient indigenous microbial
activity in the groundwater to result in anaerobic conditions, predominantly sulfate-reducing and
methanogenic. Under these terminal electron accepting processes (TEAPs), TCE has been reductively
dechlorinated to predominantly cw-l,2-DCE, chloroethene, and ethene with minor production of 1,1-
DCE, trans-l,2-DCE, and ethane.
Since the source of contamination is located approximately 750m up-gradient of Lake Michigan,
the flow of groundwater toward the lake raised concern about the potential contamination of Lake
Michigan with reductive dechlorination products, particularly chloroethene. Between 1994 and 1996,
three transects of temporary bore-holes were established on the beach and approximately 100 meters
from shore under the lake bottom to monitor the contaminant distribution and oxidation-reduction
conditions at the GSI [Lendvay, et al., 1998a]. This study of the GSI suggested that the predominant
TEAPS under Lake Michigan and in the zone along the beach was sulfate-reduction interspersed with
methanogenic conditions. Furthermore, reductive dechlorination was the predominant contaminant
transformation where these TEAPs predominated. However, in shallow regions of the contaminant
plume near the lake shoreline, iron-reducing conditions predominated, which are conducive to either
dechlorination [McCormick and Adriaens, 1998] or oxidation [Bradley and Chapelle, 1996] reactions.
The more oxidized iron-reducing zone along the top of the GSI was hypothesized to result from re-
oxygenation of the plume by surface water run-up, infiltration, and wave activity. Furthermore, it was
hypothesized that re-oxygenation of the aquifer to hypoxic conditions resulting from wave activity
might provide a suitable environment for aerobic commensalic or cometabolic biodegradation
processes in the shallow zone along the beach.
The goals of this current study were to: i) evaluate the temporal effects of increased wave
activity on the TEAPs and contaminant distribution; if) determine the most likely microbial processes
affecting intrinsic remediation of the contaminants at the GSI; Hi) present laboratory data to
corroborate field observations; and iv) determine field oxidative flux of chloroethene at the GSI.
METHODS
Multi-level arrays were placed to capture spatial variations in contaminant distribution and
predominant TEAPs, seasonal changes and effect of wave activity, and spatial infiltration of lake water
into the GSI [Lendvay, et al., 1998b; Lendvay, et al., 1999b]. The arrays were semi-permanent to
evaluate temporal effects of lake-activity on contaminant distribution and TEAPs during a six month
120
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
period (July - December)
encompassing seasonal changes in
weather and lake activity. Seasonal
changes were evidenced by changes in
Lake Michigan wave height, as
measured by National Oceanographic
and Atmospheric Administration
(NOAA) buoy 45007, averaged 0.53m
for August and September and 1.17 m
for November and December with
height increasing steadily between
June and December (Figure 1)
[Lendvay, etal., 1998b].
Jin JJ Ag Ssp OS ttv DBC
n/brth1996
Figure 1: Average wave height for Lake Michigan as measured by
NOAA buoy 45007 during the sample time study (left). Error bars
represent one standard deviation. Picture of GSI after a major storm
event (right). Notice the scatering of large debris, not present prior to
the storm.
Groundwater was sampled five
times, in four to six week intervals,
from each sample point between July
and December [Lendvay, et al., "
1998b]. Specifically, temperature, pH, reduction potential (redox), specific conductance, and dissolved
oxygen were measured using a QED flow cell [Ann Arbor, MI]. Dissolved oxygen, aqueous ferrous
iron, and aqueous sulfide were determined colorimetrically using a Chemetrics [Calverton, VA] field
sampling kit. Dissolved hydrogen gas i» the groundwater was determined as previously described
[Lovley, et al., 1994]. Samples for contaminants, methane, sulfate, and short chain organic acids were
collected, preserved, and analyzed using headspace gas chromatography, ion chromatography, or
HPLC as previously described [Lendvay, et al., 1998b].
Biomass was separated from aquifer solids to evaluate transformation potential by indigenous
methane-oxidizing microorganisms as previously described [Lendvay and Adriaens, 1999a]. The
resulting liquid suspension was plated and grown in an atmosphere of 50% methane and 50% air at
ambient temperature (22° ± 1 °C). Methane-oxidizing colonies were washed off the plates and grown
in liquid culture. Finally, cells were harvested by centrifugation and washing then resuspended in
liquid culture to obtain a cell density of 2.89 mg of cells/ml! of suspension. Aqueous batch
transformation experiments were conducted at ambient temperature to evaluate transformation kinetics
of czs-DCE, chloroethene and ethene by resting cells.
To evaluate the impact of oxidation on the flux of chloroethene into Lake Michigan, laboratory and
field results were transformed to an aquifer oxidative flux (Equation 1) as previously described
[Lendvay and Adriaens, 1999a].
dC_ A[VC]Field
dt day
(1)
The temporal change in field concentration of chloroethene (dC/dt) was assumed to equal the
observed rate of change for chloroethene concentration in the field, and the initial concentration of
chloroethene (C0) was assumed to be the average chloroethene concentration upgradient of the GSI
[Weaver, et al., 1995]. The first order decay constant (X,) was then determined. For this calculation,
only the sample points that provided evidence of chloroethene oxidation were considered to contribute
to the field oxidative flux.
121
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
To evaluate the impact of oxidation on the flux of chloroethene into Lake Michigan, laboratory and
field first order decay constants were transformed to an aquifer oxidative flux at the GSI using
Equation 2 [Lendvay and Adriaens, 1999a]. The porosity (n) was assumed to be 0.32, a nominal value
for sandy aquifers. The horizontal distance of the zone where chloroethene is oxidized (x) is the
horizontal coverage of ML-2 and ML-3. The vertical coverage of the oxidative zone (zshallow) is ratioed
to the total depth (ztot J of 6.3 meters to only consider the oxidative zone. Finally, the storm activity of
the lake is assumed to be sufficient to oxidize chloroethene for only 180 days/year.
dC _ A[VC]Field _ c
dt day °
RESULTS AND DISCUSSION
(2)
Vertical profiles of contaminants, geochemically relevant species, specific conductance, reduction
potential, and dissolved hydrogen gas concentrations were previously reported [Lendvay, et al., 1998b]
with vertical profiles for chloroethene, oxygen and methane presented here (Figure 2). In addition to
these profiles, a lack of quantifiable contaminant concentration, high dissolved oxygen, and specific
conductance measurements corresponding to typical lake water values provided direct evidence of lake
water infiltration at the shallowest sample point as wave activity increased suggesting that lake water
penetrated the top 4 meters of the aquifer. Products of electron acceptor reduction, ferrous iron and
sulfide, provided evidence for iron-reducing conditions in the shallow zone and iron- and sulfate-
reducing conditions in the deep zone of the plume. The methane concentration profile decreased with
time in the shallow zone as was noticeable for the December sampling (Figure 2C).
Reduction potential measurements increased in value with time at all elevations suggesting re-
oxidation occurred across the entire depth profile between August and December [Lendvay, et al.,
1998b]. With the caveat that reduction potential measurements are biased towards the iron couple
Parcelona, et al., 1989; Barcelona and Holm, 1991], observed values were indicative of denitrifying to
iron-reducing conditions. Dissolved hydrogen values were indicative for iron-reducing to sulfate-
reducing conditions in the shallow zone, and iron-reducing to methanogenic in the deep zone [Lendvay
et al., 1998b]. Furthermore, a temporal decrease in dissolved hydrogen concentrations (less reducing
TEAPs) corroborated temporal trends in redox measurements.
To be able to discern temporal effects in the contaminant plume, selected contaminant and
methane concentrations were compared at the same location in the plume over the time period of
interest, using quantile-quantile plots (Figure 3) [Lendvay, et al., 1998b]. These plots show that the
concentration of methane was
lower for the November and
December data compared to
the August and September
data (Figure 3A). Additionally,
these plots showed that
chloroethene concentrations
decrease with time only in the
shallow zone, and slightly
increased in the deep zones of
the plume (Figure 3C). In
contrast, the concentration of
cis-DCE increased in both the
shallow and deep zones of the Figure 2: Vertical concentration profiles for chloroethene (A), oxygen (B), and
contaminant plume with time methane (C) at ML-3 for three different time periods.
Chloroethene
Oxygen
2 169 -•:
H lee
0.00 0.02 0.04 0.06 0.08 0.10
[Oxygon] (mM)
-July . .A . Soplombor —x
Methane
173 1
0.2 0.4
[Molhann] (mU)
—July - -a . September —x
122
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
(Figure 3B). Combining the results for chloroethene and cis-DCE suggests the decrease in
chloroethene concentration in the shallow zone was not a result of dilution by infiltration of lake water.
Tampon! Virlition In H.lhmr
To elucidate specific field microbial processes responsible for chloroethene transformation,
scattergrams comparing chloroethene or methane with oxygen in the shallow zone were evaluated for
statistical correlations [Lendvay, et al., 1998a]. In the case of both chloroethene and methane, peak
concentrations occurred at low oxygen
concentrations and low chloroethene and
methane concentrations occurred at high
oxygen concentrations. Combined, these trends
suggest that chloroethene may be co-oxidized
by methane-oxidizing microorganisms in the
shallow zone of the GSI as lake activity
increases.
To test this field-derived hypothesis and
corroborate the field measurements, laboratory
studies were conducted using groundwater and
aquifer solids collected from the GSI. The
mixed cultures were grown on methane as a
sole source of carbon and energy under aerobic
conditions. A resting cell oxidation experiment
was conducted for cis-DCE, chloroethene, and
T«mpor«l tftrlnltsn In Chli
lVCIAug./S.p.inM)
jgshanoiv QDaep |
Figure 3: Quantile-quantile plots for the data indicated. For
each case, all data (ML-2 and ML-3) are considered for
ethene in triplicate (Figure 4). Separate controls November and December samples on the ordinate, and
of either 220 mg/£ sodium azide or 0.3% . August and September samples on the abscissa.
(voL/vol. gas phase) ethyne were effective at
suppressing all transformation of contaminants. Considering Figure 4, transformation of both
chloroethene and ethene is evident relative to controls, with the rate of chloroethene oxidation being
2.9 nmoles/(day-mg protein) and ethene oxidation being 0.9 nmoles/(day-mg protein). No
transformation of cis-DCE was observed relative to controls. These results support the field findings,
that chloroethene and possibly ethene are co-oxidized by methane-oxidizing microorganisms while cis-
DCE is not.
To evaluate the impact of chloroethene oxidation on the flux of chloroethene into Lake Michigan,
£
[Cls-DCE
cis-DCE Oxidation
A
^ f I
0 20 40 60
Time (hours)
• Active cls-DCE H Ethyne Control
A Azide Control
80
Chloroethene Oxidation "
12 1
10 f~~^ s a F?
s
a.
c
» 6
o
o
t 2.
"^\^
*^\+
y=-6.2E-02x*1.1E+01
RZ = 9.9E-01
0 20 40 60 80
Time (hours)
• Active Chloroathane ^Ethyne Control
A Aztde Control
[Ethene] (^M)
Ethene Oxidation
C
i
»
y = -2.QE-02X + 9.9E+00
R2 = 9.6E-01
0 20 40 60
Time (hours)
• Active Ethane M Ethyne Control
A Azide Control
80
Figure 4: Resting Cell Oxidation of cis-DCE (A), chloroethene (B), and ethene (C) by methane oxidizing
microorganisms.
123
-------
Proceedings of the Ground-Water/Surface- Water Interactions Workshop
July 2000
Table 1: Reported value for chloroethene flux at transect-5 (upgradient transect) using a MOD-
FLOW model [Weaver, et al., 1995], and values for chloroethene flux at the GSI using
calculated field and laboratory rates. Percent values represent the percent of chloroethene flux
at transect-5 that is oxidized by the reported value.
Defined Value
Flux (g/(vear-m2))
Chloroethene Flux at
Transect-5
Oxidative Flux of
Chloroethene by Field
Measurement
Oxidative Flux of
Chloroethene by Laboratory
Measurement
0.86
0.063 (-7%)
0.0007 (-0.1%
of 0.86 or -1%
of Field Flux)
laboratory and field results were transformed to a field oxidative flux [Lendvay and Adriaens, 1999a].
Using the calculated field and laboratory chloroethene oxidation rates, field flux calculations were
possible (Equation 2). The calculated oxidation fluxes were compared to the chloroethene flux
upgradient as calculated by a MOD-FLOW model (Table 1) [Wilson, et al., 1994].
Field calculations of the oxidative flux for chloroethene suggest that only about 7% of the annual
chloroethene flux into Lake Michigan is mitigated by measurable biogeochemical processes at the GSI
as a result of increased storm activity. Of this observed field flux, only 1% could be associated with a
particular microbial culture namely methane-oxidizing microorganisms. The remaining catalytic
activity affecting the contaminant plume may be due to other physiological types in the aquifer solids
such as non-culturable methane-oxidizing, heterotrophic [Wackett, et al., 1989; Davis and Carpenter,
1990], autotrophic fVannelli et al, 1990], ethene oxidizing [Freedman and Herz, 1996], iron-reducing
[Bradley and Chapelle, 1996], or fermentative microorganisms [Bradley, et al., 1998]. This study has
contributed to our understanding of the dynamics of GSIs with respect to microbial activity and
geochemistry, and points towards a possible role of methane-oxidizers in mitigating chloroethene
imparted toxicity.
REFERENCES
Barcelona, MJ. and T.R. Holm, 1991. Oxidation-Reduction Capacities of Aquifer Solids,
Environmental Science and Technology, v. 25, pp. 1565-1572.
Barcelona, M.J., Holm, T.R., Schock, M.R., and O.K. George, 1989. Spatial and Temporal Gradients
in Aquifer Oxidation-Reduction Conditions, Water Resources Research, v. 25, pp. 991-1003.
Bradley, P.M. and F.H. Chapelle, 1996. Anaerobic Mineralization of Vinyl Chloride in Fe (IH)-
Reducing, Aquifer Sediments, Environmental Science and Technology, v. 30, pp. 2084-2086.
Bradley, P.M. and F.H. Chapelle, 1998. Effect of Contaminant Concentration on Aerobic Microbial
Mineralization of DCE and VC in Stream-Bed Sediments, Environmental Science and Technology,
v. 32, pp.553-557.
Davis, J.W. and C.L. Carpenter, 1990. Aerobic Biodegradation of Vinyl Chloride in Groundwater
Samples, Applied and Environmental Microbiology, v. 56, pp. 3878-3880.
124
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Freedman, D.L. and S.D. Herz, 1996. Use of Ethylene and Ethane as Primary Substrates for Aerobic
Cometabolism of Vinyl Chloride, Water Environment Research, v. 68, pp. 320-328.
Lendvay, J.M., Sauck, W.A., McCormick, M.L., Barcelona, M.J., Kampbell, D.H., Wilson, J.T., and
P. Adriaens, 1998a. Geophysical Characterization, Redox Zonation, and Contaminant Distribution
at a Groundwater/Surface Water Interface, Water Resources Research, v. 34, pp. 3545-3559.
Lendvay, J.M., Dean, S.M., and P. Adriaens, 1998b. Temporal and Spatial Trends in Biogeochemical
Conditions at a Groundwater-Surface Water Interface: Implications for Natural Bioattenuation,
Environmental Science and Technology, v. 32, pp. 3472-3478.
Lendvay, J.M. and P. Adriaens, 1999a. Laboratory evaluation of temporal trends in biogeochemical
conditions at a groundwater-surface water interface, Physics and Chemistry of the Earth.
Lendvay, J.M., Dean, S.M., Barcelona, M., Adriaens, P., and N.D. Katopodes, 1996b. Installing Multi-
Level Sampling Arrays to Monitor Groundwater and Contaminant Discharge to a Surface Water
Body, Groundwater Monitoring and Remediation.
Lovley, D.R., Chapelle, F.H., and J.C. Woodward, 1994. Use of Dissolved H2 Concentration to
Determine Distribution of Microbially Catalyzed Redox Reactions in Anoxic Groundwater,
Environmental Science and Technology, V..28, pp. 1205-1210.
McCormick, M.L. and P. Adriaens, 1998. Tetrachloroethylene Transformation in an Iron Reducing
Enrichment Culture, Abstract, 98th General Meeting of the American Society for Microbiology,
Atlanta, GA, p.453, American Society for Microbiology, Washington, DC, 1998.
Semprini, L., P.K. Kitanidis, D.H. Kampbell, and J.T. Wilson, 1995. Anaerobic Transformation of
Chlorinated Aliphatic Hydrocarbons in a Sand Aquifer Based on Spatial Chemical Distributions,
Water Resources Research, v. 31, pp.1051-1062.
Tiedeman, C. and S.M. Gorelick, 1993. Analysis of Uncertainty in Optimal Groundwater Contaminant
Capture Design, Water Resources Research, v. 29, pp. 2139-2153.
Vannelli, T., Logan, M., Arciero, D.M., and A.B. Hooper, 1990. Degradation of Halogenated Aliphatic
Compounds by the Ammonia-Oxidizing Bacterium Nirosomonas europaea, Applied and
Environmental Microbiology, v. 56, pp.1169-1171.
Wackett, L.P., Brusseau, G.A., Householder, S.R., andR.S. Hanson, 1989. Survey of Microbial
Oxygenases: Trichloroethylene Degradation by Propane-Oxidizing Bacteria, Applied and
Environmental Microbiology, v. 55, pp. 2960-2964.
Weaver, J.W., J.T. Wilson, D.H. Kampbell, and M.E. Randolph, 1995. Field Derived Transformation
Rates for Modeling Natural Bioattenuation or Trichloroethylene and its Degradation Products, in
Proceedings of the Next Generation Environmental Models and Computational Methods, Bay City,
MI.
Wilson, J.T., Weaver, J.W., and D.H. Kampbell, 1994. Intrinsic Bioremediation of TCE in Ground
Water at an NPL Site in St. Joseph, Michigan, EPA/540/R-94/515, pp.154-160.
125
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Natural Attenuation of Chlorinated Solvents in a
Freshwater Tidal Wetland, Aberdeen Proving
Ground, Maryland
By Michelle M. Lorah and Lisa D. Olsen
Ground-water contaminant plumes that are flowing toward or currently discharging into wetland
areas present unique remediation problems because of the hydrologic connections between ground
water and surface water and the sensitive habitats in wetlands. Because wetland sediments typically
have a large diversity of microorganisms and redox conditions that could enhance biodegradation, they
are ideal environments for natural attenuation of organic contaminants. "Natural attenuation" is a
general term that includes all naturally occurring physical, chemical, and biological processes that can
reduce contaminant concentrations or toxicity without human intervention. Biodegradation generally is
considered the most important of these processes for ground-water remediation purposes because it is a
destructive process, unlike sorption, dilution, or volatilization. Natural attenuation is a treatment
method that would leave the ecosystem largely undisturbed and be cost-effective. In this research, the
natural attenuation of chlorinated volatile organic compounds (VOCs) was studied in a contaminant
plume that discharges from a sand aquifer to a freshwater tidal wetland at Aberdeen Proving Ground,
Maryland (Lorah, et al., 1997).
Biodegradation processes of two of the major contaminants, trichloroethylene (TCE) and 1,1,2,2-
tetrachloroethane (PCA), are the focus of this combined field and laboratory study. The fate of PCA in
the wetland is particularly emphasized in this research because the occurrence and dominant pathways
of PCA degradation in ground water or soil were largely unknown. The few previous studies on PCA
degradation were laboratory experiments that were constructed with anaerobic mineral medium or
glass beads and seeded with mixed cultures from municipal sludge waste, or with abiotic aqueous
mixtures of transition-metal coenzymes (Lorah, et al., 1997). The general field approach used in this
study included (1) installing nested drive-point piezometers to characterize the ground-water chemistry
and contaminant distribution along two transects through the wetland (one shown in Figure 1), and (2)
using porous membrane sampling devices (peepers) to obtain centimeter-scale resolution of
contaminant distribution in the wetland porewater. The general laboratory approach included
conducting batch microcosm experiments with wetland sediment and porewater under methanogenic,
sulfate-reducing, and aerobic conditions to confirm field evidence of biodegradation pathways,
investigate potential controlling factors on biodegradation, and estimate biodegradation rates.
Field evidence collected along the two ground-water flowpaths shows that anaerobic
biodegradation of TCE and PCA is enhanced in the wetland compared to the aquifer sediments (Lorah,
et al., 1997; Lorah and Olsen, in press). The enhanced biodegradation is associated with the natural
increase in dissolved organic carbon concentrations and decrease in redox state of the ground water
along the upward flow direction in the wetland sediments. The aquifer typically is aerobic. Iron-
reducing conditions are predominant in the lower wetland sediment unit composed of clayey sand and
silt, and methanogenesis was predominant in an upper unit composed of peat (Figure 2). A decrease in
concentrations of TCE and PCA and a concomitant increase in concentrations of anaerobic daughter
products occurs along upward flowpaths through the wetland sediments (Figures 1 and 2). The
daughter products 1,2-dichloroethylene (1,2-DCE), vinyl chloride (VC), 1,1,2-trichloroethane (1,1,2-
TCA), and 1,2-dichloroethane (1,2-DCA) are produced from hydrogenolysis of TCE and from PCA
126
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
VBfllCAL EXAGGERATION « 3.67X
EXPLANATION
-50 LWE OF EQUAL CONCENTRATION OF TraCHLOHOHHYLENE (TCE) -
Intervol in mlcrogroms per Brer, Is variable; clashed where
approximatey bcated.
63. PIEZOMETER SCREEN-Number
3£4 Is concentration of trfchtoroelhyjene.
In mlcrograms per Iter.
Figure 1. Concentrations of TCE in ground water along section A-A', June-October 1995.
degradation through hydrogenolysis and dichloroelimination pathways. Total concentrations of TCE,
PCA, and their degradation products, however, decrease to below detection levels (generally less than
0.5 ug/£) within 0.15 to 0.30 m of land surface. Natural attenuation in the wetland sediments seems to
be effective even where relatively high concentrations of VOCs are discharging upward through very
thin (less than 2 m) layers of wetland sediment and when microbial activity probably decreases during
cooler seasons (Lorah, et al., 1997; Lorah and Olsen, in press). Field evidence indicates that highly
reducing conditions are not necessary for dichloroelimination of PCA to 1,2-DCE, or for
hydrogenolysis of PCA to 1,1,2-TCA and then to 1,2-DCA. Maximum concentrations of VC, however,
Upper peat (B)
unit SITE
WB-26
Upper peat unit
Upper peat
unit
0.00 0.25 0.50 0.75 1.00
CONCENTRATION, IN MICROMOLES PER LITER
EXPLANATION
| 1,1,2,2-TETRACHLOROETHANE (PCA) ^ VINYL CHLORIDE (VC)
| TRCHLOROETHYLENE (TCE) t3 1,1,2-TRICHLOROETHANE(112TCA)
| 1,2-DICHLOROETHYLENE (12DCE) Q' 1 -2-DICHLOROETHANE (T 2DCA)
0 200 400 600
CONCENTRATION, IN MICROMOLES PER LITER
EXPLANATION
H FERROUS IRON
Q SULFIDE
• METHANE
Figure 2. Vertical distribution of (A) the parent contaminants TCE and PCA and possible anaerobic daughter
products and (B) selected redox-sensitive constituents at site WB-26, June-October 1995.
127
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
coincided with the highest concentrations of methane in the wetland porewater, indicating that
continued anaerobic degradation of 1,2-DCE to VC and of VC to the non-toxic end-product of
ethylene may require the highly reducing conditions of methanogenesis.
Laboratory microcosms that were prepared using wetland sediment and ground water from the site
confirmed field evidence of biodegradation pathways and allowed estimation of biodegradation rates
(Lorah, et al., 1997). TCE biodegradation occurred through hydrogenolysis to 1,2-DCE (predominately
the cis isomer) and VC under methanogenic (Figure 3) and sulfate-reducing conditions. For PCA
degradation under methanogenic conditions, the cis and trans isomers of 1,2-DCE and VC were the
predominant persistent daughter products in one set of microcosm experiments (Lorah and Olsen,
1999) (Figure 3). In two sets of later experiments, however, 1,2-DCA was the predominant persistent
daughter product from PCA and 1,1,2-TCA degradation under methanogenic conditions. The
differences between the experiments seemed to result from differing pathways of 1,1,2-TCA
degradation, rather than from differences in the initial PCA degradation pathway. 1,1,2-TCA was
produced simultaneously with 1,2-DCE early in the time course in all experiments. The 1,1,2-TCA
produced from hydrogenolysis of PCA was degraded by dichloroelimination to VC in the first
experiment, whereas it was degraded by continued hydrogenolysis to 1,2-DCA in the second and third
experiments. Natural temporal or spatial variations in the microbial populations in the wetland
sediments may have caused the differing degradation pathways in these experiments.
In all PCA-amended microcosms, 1,1,2-TCA occurred simultaneously with 1,2-DCE early in the
time course, indicating that hydrogenolysis and dichloroelimination of PCA can occur simultaneously
(Lorah and Olsen, 1999). Significantly lower ratios of cw-l,2-DCE to trans-l,2-DCE were produced
by dihaloelimination of PCA than by hydrogenolysis of TCE (Figure 3). Only one other study, which
was conducted in the laboratory using anaerobic municipal sludge, has reported evidence for both of
ixi
0.8
0.6
0.4
' 02
0.0
(A)
Uve:
.cfe-l^E '.,V L^>
J*-0/ \2DCEX "--- \
^p^^^^n-rr-yfe^s;^":"^:-;:
Sterile: 12DCE
0 5 10 IS 20 25 30 35"
NUMBER OF DAYS AFTER TCE ADDITION
0.0
0 5 10 15 20 25 30 35
NUMBER OF DAYS AFTER PCA ADDITION
0.8
0.7
0.6
0.5
0.4
0.3
0,2
^ 0.1
OX)
(C)
V Sterile: 12DGE/'
\
35
Uve:cfe-12DCE
" Uve?\
/
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
these PC A degradation pathways (Chen, et al., 1996). Recognition of the dichloroelimination pathway
for PCA is important because TCE is a common co-contaminant with PCA at this site and other
hazardous-waste sites. The 1,2-DCE isomer distribution potentially could assist in determining whether
PCA degradation is occurring at sites where TCE is a co-contaminant. In the PCA-amended
microcosms with the wetland sediment, TCE production commonly was less than 5 % of the initial
PCA concentration in the live and sterile microcosms, showing that abiotic dehydrochlorination is not
a significant degradation pathway for PCA in this environment.
Under methanogenic conditions, first-order biodegradation rates of TCE ranged from 0.30 to 0.37
day"1 (half-life of about 2 days), showing extremely rapid biodegradation in these organic-rich wetland
sediments. Although the TCE biodegradation rate was an order of magnitude slower under sulfate-
reducing conditions (0.032 day"1) than methanogenic conditions, the rate was still two orders of
magnitude higher than those reported in the literature for anaerobic TCE biodegradation in microcosms
constructed with sandy aquifer sediments (Rifai, et al., 1995). The slow production and degradation of
1,2-DCE and VC in TCE-amended microcosms that were incubated under sulfate-reducing conditions
or with the addition of an inhibitor of methanogenic activity confirmed field evidence that
methanogenic activity is important in continued anaerobic degradation of these daughter products.
First-order rate constants for anaerobic degradation of PCA and 1,1,2-TCA ranged from 0.15 to
0.58 day"1 (half-lives of 1.2 to 4.6 days), again showing that biodegradation of highly chlorinated
VOCs is extremely rapid in the wetland sediments. Similar PCA degradation rates were observed
under methanogenic and sulfate-reducing conditions, although methane production and sulfate
reduction occurred simultaneously during the first 15 days of incubation in the microcosms that were
amended with sulfate to stimulate sulfate-reducing conditions. The addition of an inhibitor of
methanogenic activity to PCA-amended microcosms decreased the methane production rates by a
factor of 10 and caused a nearly 50% decrease in the PCA degradation rate compared to microcosms
without the inhibitor. In addition, the production and subsequent degradation of the daughter products
1,2-DCA, 1,2-DCE, and VC were slower when methanogenic activity was inhibited in the PCA-
amended microcosms. Both the TCE-amended and PCA-amended microcosm experiments, therefore,
indicate that complete anaerobic degradation of the chlorinated VOCs is most rapid when
methanogenic activity is high.
Although the wetland sediments have predominantly anaerobic conditions, aerobic conditions may
be present in surficial sediments near the air-water interface and in subsurface sediments near plant
roots, providing a suitable environment for methanotrophs that can degrade chlorinated VOCs through
cometabolic oxidation (Lorah, et al., 1997). Although coupling of anaerobic and aerobic degradation
processes has been suggested as the best possible bioremediation method for chlorinated VOCs such as
TCE, few studies have investigated the degradation of TCE under ioth anaerobic and aerobic
conditions for a natural subsurface setting. In aerobic microcosm experiments with the wetland
sediment, biodegradation of cis- 1,2-DCE, trans- 1,2-DCE, and VC only occurred if methane
consumption occurred, indicating that methanotrophs were involved. Aerobic biodegradation rates for
cis- 1,2-DCE, trans-l,2-DCE, and VC were in the same range as those measured for TCE and PCA
under anaerobic conditions. Production of these anaerobic daughter products of TCE and PCA,
therefore, could be balanced by their consumption where methanotrophs are active in the wetland
sediment, including near land surface and in the rhizoplane and root tissues of aquatic plants in
wetlands (Lorah, et al., 1997).
In summary, biodegradation through both anaerobic and aerobic processes is a significant natural
attenuation mechanism for chlorinated hydrocarbons in these wetland sediments, causing a reduction
129
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
in contaminant concentrations and toxicity before surface-water receptors are reached. This combined
field and laboratory study provides a general approach and scientific basis for investigating the
feasibility of natural attenuation as a remediation alternative for other sites where ground-water plumes
discharge to wetlands and similar organic-rich environments at ground-water/surface-water interfaces.
Because conditions in the wetland sediments are naturally conducive to biodegradation of the
chlorinated VOCs in the discharging ground water, biodegradation could potentially be sustained
indefinitely, unlike many ground-water environments where biodegradation commonly is limited by
the supply of organic substrates. The results of this study also increase our understanding of the fate of
chlorinated VOCs in the environment, providing information beneficial in characterizing contaminant
behavior and in implementing bioremediation systems for other ground-water contaminant plumes.
REFERENCES
Chen, Chun, Puhakka, J. A., and J.F. Ferguson, 1996. Transformations of 1,1,2,2-tetrachloroethane
under methanogenic conditions, Environmental Science and Technology, v. 30, no. 2, pp. 542-547.
Lorah, M.M., and L.D. Olsen, 1999. Degradation of 1,1,2,2-tetrachloroethane in a freshwater tidal
wetland: Field and laboratory evidence, Environmental Science and Technology, v. 33, p. 227-234.
Lorah, M.M., and L.D. Olsen, in press. Natural attenuation of chlorinated volatile organic compounds
in a freshwater tidal wetland: Field evidence of anaerobic biodegradation: Water Resources
Research.
Lorah, M.M., Olsen, L.D., Smith, B.L., Johnson, M.A., and W.B. Fleck, 1997. Natural attenuation of
chlorinated volatile organic compounds in a freshwater tidal wetland, Aberdeen Proving Ground,
Maryland: U.S. Geological Survey Water-Resources Investigations Report 97-4171, 95 pp.
Rifai, H.S., Borden, R.C., Wilson, J.T., and C.H. Ward, 1995. Intrinsic bioattenuation for subsurface
restoration, in Intrinsic Bioremediation. R. E. Hinchee and others (eds.), Columbus, Ohio, Battelle
Press, pp. 1-29.
AUTHOR INFORMATION
Michelle M. Lorah and Lisa D. Olsen, U.S. Geological Survey, 8987 Yellow Brick Road, Baltimore,
MD 21237.
130
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Discharge of Contaminated Ground Water to
Surface Water: An Ecological Risk Assessment
Perspective
By Mary Baker Matta and Tom Dillon
INTRODUCTION/OVERVIEW
The ecological risk associated with the discharge of contaminated ground water to surface water
may be difficult to evaluate at complex hazardous waste sites..However, at many sites, groundwater
discharges provide a significant pathway for contamination to reach ecological receptors. Although
EPA has guidance for generic risk assessment, and has specific guidance for ecological risk
assessments at CERCLA sites, specific guidance on how to evaluate ecological risk of groundwater
discharges is lacking.
The hyporheic zone connects the ecological communities in ground water, sediment, and surface
water. Although benthic macroinvertebrate communities are often considered in ecological risk
assessments (and they may be adversely affected by contamination in the hyporheic zone), microbial
communities of the hyporheic zone are rarely considered receptors of concern in ecological risk
assessments. Because groundwater discharges contribute to sediment and surface water contamination
they also have the potential to adversely affect fish and other aquatic species, particularly where
bioaccumulative compounds are released or where sensitive life stages are in close proximity to
groundwater discharge points.
The nature and extent of the contamination present in groundwater and characteristics of the
ecosystem will determine what and where to sample, and what tests should be conducted to evaluate
ecological risk and develop protective cleanup levels if risk is significant. Natural attenuation is
increasingly considered as a remedial option at many sites. Natural attenuation is influenced by
microbial activity and physico-chemical characteristics of the groundwater and its movement.
Therefore, if natural attenuation is to be considered as a viable remedial option, the microbial
community in the hyporheic zone should be protected, and its role in chemical fate and transformation
should be assessed. At sites where ecological risk is significant, more active cleanup measures may be
necessary to protect natural resources.
NOAA's experience in aquatic ecological risk assessment is applied in this poster to provide some
recommendations for evaluating ecological risk of groundwater discharge to surface waters. This
poster focuses on data needs and how to answer them, structured around the EPA risk assessment
process.
RISK ASSESSMENT STEPS 1 AND 2: SCREENING LEVEL ECOLOGICAL RISK ASSESSMENT
NOAA prepares Coastal Hazardous Waste Site Reviews using available information to screen sites
based on:
• proximity of the site to the coast or inland water bodies that support anadromous fish populations
131
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
• site history, contaminants that are likely to be or have been detected at the sife
• potential contaminant migration pathways (including groundwater)
• presence of sensitive habitats or species near the site
This same information is required in screening level risk assessments.
Maximum contaminant concentrations in groundwater are screened by NOAA using a value ten
times greater than chronic ambient water quality criteria to account for potential dilution when
groundwater enters surface waters and to protect all aquatic species. Therefore, detection limits for
chemical analysis of unfiltered groundwater should be less than 10 times chronic ambient water quality
criteria to conduct this assessment. The most recent' edition of Hazardous Waste Site Reviews (NOAA,
1997) indicates that groundwater discharges are a widespread problem for natural resources:
• At 10 of the 11 sites evaluated, groundwater is a potential pathway for contaminants to reach
natural resources.
• At 8 of the 11 sites evaluated, concentrations in groundwater exceed screening levels.
Sites Reviewed in December, 1997
Region 1 Beede Waste Oil, Plaistow, NH
Region 2 V&M/Albaladejo Farms, Vega Bgja, Puerto Rico ->
Region 3 Fort George G. Meade, Anne Arundel Co, MD
Norfolk Naval Base, Norfolk, VA
Salford Quarry, Lower Salford Townshp, PA
Region 4 Brunswick Wood Preserving, Brunswick, GA
MRI Corp, Tampa, FL
*Terry Creek Dredge Spoil/Hercules, Brunswick, GA
Tyndall Air Force Base, Bay Co, FL
Region 6 Madisonville Creosote Works, Madisonville, LA
Region 10 Oeser Company, Bellingham, WA'
*Groundwater not a significant pathway to natural resources
RISK ASSESSMENT STEP 3: BASELINE RISK ASSESSMENT PROBLEM FORMULATION
Steps:
• Refine list of contaminants based on screening
• Summarize toxicological effects
• Consider likely fate and transport of contaminants
• Determine receptors likely to be at risk
• Determine complete exposure pathways
• Develop conceptual site model with risk questions
• Develop assessment endpoints
Considerations specific to sites with groundwater contamination:
* Potential biogeochemical alterations of contaminants.
-persistent organic contaminants might be released as an LNAPL or DNAPL and
bind to sediments as they are released to surface water.
132
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop July 2000
- Metals might be released in dissolved form and complex with other compounds at
the discharge point, groundwater discharge to marine water may alter solubility of metals.
- Groundwater discharge may leach metals from sediments.
- Organic contaminants released in a solvent plume may be more bioavailable
• Conceptual model for the site should include pathways for exposure and mechanisms of toxicity
for the contaminants and receptors specific to the site
• Consider characteristics of discharge specific to the regime:
-groundwater discharge into intertidal zone at marine sites;
- discharge into tidal creeks in salt marshes; ,
- hyporheic zone in alluvial rivers as a function of channel morphology, bed
roughness, and permeability (Triska, et al. 1989).
• Consider hyporheos as receptors of concern (macrofauna and microbes)
Assessment Endpoints of General Concern to NOAA (with some specific examples)
• Protection of benthic community structure and function
(Protection of stonefly populations from direct toxic effects)
• Protection of aquatic community structure and function
(Protection offish eggs and larvae from direct toxic effects, including ecologically relevant
sub-lethal effects) .
(Protection of hyporheic microbial community from direct toxic effects)
• Protection of fish populations and communities
(Protection of fish from reproductive effects)
(Protection of fish from immune system disruption)
(Protection of fish from reductions in survival and growth)
• Protection of specific habitat functions (for example, nutrient cycling)
(Protection of hyporheic microbial community from direct toxic effects)
*
• Protection of fishery resources from contamination
(Protection of human health from exposure to bioaccumulated contaminants)
RISK ASSESSMENT STEPS 4, 5, AND 6: STUDY DESIGN/DATA QUALITY OBJECTIVES/FIELD
VERIFICATIONS/SITE INVESTIGATIONS
Steps
• Evaluate uncertainty in existing data
• Develop measurement endpoints to evaluate assessment endpoints
• Develop work plan to evaluate exposure and effects
-methods
-statistical considerations
-sampling locations, timing, frequency
133
-------
Proceedings of the Ground-Water/Suiface- Water Interactions Workshop
July 2000
• Verify sampling design
• Site investigation
• Data analysis
Considerations specific to sites with groundwater contamination:
•Evaluating exposure potential is a major concern; need to determine nature and extent
of contamination and factors that may control bioavailability and toxicity:
-Discharge locations
-Flux of contaminants
-Form/speciation of contaminants
-Presence of non aqueous phase layers (NAPL)
-Potential contaminant interactions, for example enhanced transport of hydrophobic
(biomagnifying) contaminants in a dissolved solvent plume
• NOAA recommends a weight of evidence approach based on:
-Groundwater models predicting a vector discharge to surface water
-Remote sensing and geological analysis: identification of paleochannels using ground-
penetrating radar; location of surface depressions, abandoned meander channels to locate
springbrooks through aerial photography (Stanford and Ward, 1993).
-Developing a water budget - upstream/downstream gauging, tracer studies
-Changes in water quality parameters due to groundwater discharge (temperature, pH,
conductance, nutrients, DOC, oxygen (Triska, et al., 1989).
-Direct measurements
•Specific effects measurements will be a function of receptors of concern; contaminants; and
exposure pathway
-Benthic macroinvertebrates-generallya concern for direct toxicity via surface water or
sediment contamination as a result of groundwater discharges
-Microbes-generally a concern for direct toxicity of groundwater, pore water, or surface water
-Fish- may include a concern for direct toxicity of pore water or surface water to early life
stages (for example, where dissolved metals are discharged) or a concern for indirect
toxicity after contamination of sediment and food organisms.
• Groundwater and NAPL can contaminate sediments and surface water
• Ecological risk associated with sediment contamination is usually evaluated via
-chemical analysis of sediment (focus on persistent contaminants)
-toxicity testing
-benthic community evaluations
-chemical analysis of biota for bioaccumulative contaminants
-comparisons of tissue concentrations to literature effects thresholds
-food web modeling for persistent bioaccumulative contaminants
• Ecological risk associated with water contamination is usually evaluated via
-chemical analysis of water (focus on persistent contaminants or continuous releases of less
persistent contaminants)
134
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
-aquatic toxicity tests
-comparisons of water concentrations to literature or AWQC benchmarks.
• Sampling and analysis concerns:
-number and locations of samples should be adequate to evaluate exposure potential
and minimize uncertainty (plumes should be characterized, and locations where sensitive
receptors would be exposed should be sampled)
-seasonal considerations affecting groundwater flow or the seasonal presence of
sensitive receptors may require multiple samplisg events
-analyze filtered and unfiltered water (organisms are exposed to both)
-analyze pore water where appropriate for the receptor (fish eggs, for example)
-collect ancillary data that can be used to interpret bioavailability, toxicity, and
potential for natural attenuation (sediment grain size, sediment total organic carbon, acid
volatile sulfides, dissolved organic carbon, nutrients, alkalinity, dissolved oxygen).
-use detection limits low enough to compare to benchmarks
-consider metal speciation and effects on bioavailability and toxicity
-selection of reference sites is critical in evaluating exposure and effects (reference
sites should be similar to the study site, but located away from point sources of
contamination)
RISK ASSESSMENT STEP 7: RISK CHARACTERIZATION
Steps;
" • Interpreting data
• Tracking sources
• Modeling mass flux
• Modeling food web effects
• Reducing and incorporating uncertainty
• Applying protective assumptions
• Interpreting the weight of evidence
• Developing cleanup levels
Considerations specific to sites with groundwater contamination:
• Interpreting data
-agree on interpretive methods before collecting data
-compare results to benchmarks, control samples, and reference site results
-consider normalizing data to nutrients, grain size, or other factors
• Source tracking (relative contribution of groundwater and surface releases) is important to ensure
success of the remedy (in terms of reducing risk to ecological receptors). Consider the potential
for recontamination through ongoing groundwater discharge.
• Models on groundwater discharge rates and contaminant loading are usually
highly uncertain, therefore, to be protective, conservative assumptions must be made until
better techniques, data, and models are available.
•Conduct specific studies to determine the potential for natural attenuation, which is driven largely
by microbes and their physico-chemical environment (consider the specific types of microbes
required to degrade contaminants at the site; some require aerobic).
135
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
•Water quality and sediment goals for receptors can be empirically derived from toxicity tests but
back calculating for groundwater quality goals requires a good model
Examples:
At the Metal Bank of America site in Philadelphia, PA, the primary pathway for contaminants to
reach natural resources was through discharge of contaminated groundwater and non-aqueous phase
layer. PCBs accumulated in sediment, clams, and fish near the site to concentrations that were
determined to pose significant risk.
RISK ASSESSMENT STEP 8: RISK MANAGEMENT
Considerations specific to sites with groundwater contamination:
•Potential for recontamination- uncontrolled groundwater discharges have the
potential to interfere with a sediment remedy
•Monitoring-the effects that drove the selection of the remedy should be incorporated into
monitoring, significant ongoing exposures should be monitored over time and re-evaluated
CONCLUSIONS/RECOMMENDATIONS
• Potential for groundwater to provide a contaminant pathway to aquatic species should be
considered, it seems to be a problem throughout the country in all regions examined.
• NOAA screens GW for potential concern using 10 times chronic AWQC values
• Detection limits for unfiltered groundwater should be low enough to conduct this screening
• Consider potential biogeochemical alterations and complex interactions between contaminants.
• Consider characteristics of discharge specific to the regime:
• Consider hyporheos as receptors of concern (macrofauna and microbes)
• Use multiple methods and a weight of evidence to evaluate exposure potential (discharge
locations and contaminant fluxes)
• Evaluate potential for bioaccumulation and food web effects
• Specific sampling and analysis methods will depend on characteristics of contaminants and
receptors at the site (but consider seasonal variations and sample in such a way as to provide
information relevant for the behavior of the organisms at the site)
• A critical need is to reduce uncertainty in risk conclusions: better methods are needed to
locate and quantify contaminant flux from groundwater to surface water
• Until better data is available, protective assumptions should be used to evaluate risk to natural
resources
REFERENCES
NOAA, 1997. Coastal hazardous waste site reviews, December 1997, Seattle, WA, 109 pp.
Stanford, J.A. and J.V. Ward., 1993. An ecosystem perspective of alluvial rivers: connectivity and the
hyporheic corridor, Journal of the North American Benthological Society, v. 12, pp. 48-60.
136
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Triska, F.J., V.C. Kennedy, RJ. Avanzino, G.W. Zellweger, and K.E. Bencala, 1989. Retention and
transport of nutrients in a third order stream: Hyporheic processes. Ecology, v. 70, pp. 1893-1905.
U.S. EPA, 1997. Ecological Risk Assessment Guidance for Superfund: Process for Designing and
Conducting Ecological Risk Assessments. Interim Final, Edison, NJ, EPA 540-R-97-006.
U.S. EPA, 1992. Framework for Ecological Risk Assessment, Washington, DC, EPA/630/R-92/001.
AUTHOR INFORMATION
Mary Baker Matta and Tom Dillon, NOAA Coastal Protection and Restoration Division, Seattle,
Washington.
(NOAA Hazardous Waste Site Reviews are available from John Kaperick, NOAA Office of Response
and Restoration Bin C15700 Seattle, WA 98115).
137
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Defining Groundwater Outcrops in West Neck
Bay, Shelter Island, New York Using Direct
Contact Resistivity Measurements and Transient
Underflow Measurements
By Ronald Paulsen
ABSTRACT
Mapping out groundwater discharge zones can be a formidable task due to the highly variable nature of
the discharge zone, temporal and spatial variability of seepage rates within the zone, and complications
from tidal loading. An integrated approach incorporating direct contact resistivity logging and transient
seepage rate measurements was undertaken to delineate the seepage zone and monitor discharge on a
bay-wide scale. Conductivity values for the saturated sediments ranged from 280 S/cm in freshwater
zones to 12,800 S/cm in zones with high salinity. The discharge zone at West Neck Bay, Shelter Island
was observed to extend to 10-75 feet offshore. The groundwater seepage within the discharge zone was
measured using a time transient seepage meter that was developed with ultrasonic technology. Seepage
velocities in the study area ranged from 1.27 x 10"3 cm/s to 3.94 x 10"5cm/s, equivalent to a mean value
of 16 {/m2/d. Integrating over the horizontal extent of the seepage zone, the total daily discharge was
estimated to be 1.72 x 106 S/day for the north-east section of West Neck Bay. This estimate of the total
discharge due to underflow is comparable to the recharge in the contributing area, estimated to be 1.50
x 106 {/day for this section of the bay.
DIRECT CONTACT RESISTIVITY MEASUREMENTS
To characterize near-shore sediments in a simple and rapid manner, we used geophysical logging
to determine the electrical resistivity of surface sediments off-shore. Electrical measurements have
been used for some time to characterize the lithology and hydraulic characteristics of geological
structures. The basic concept of resistivity logging dates back to 1927 when C.M. Schlumberger made
the first well log near Paris (Goldberg, 1997). Conductivity generally increases with increasing
porosity (Gueguen and Palciauskas, 1994). Archie (1942) invoked laboratory measurement of
conductivity to infer amounts of water and hydrocarbons in the pore space. The electrical conductivity
of saturated sediment is commonly analyzed in terms of the formation factor F as a function of the
porosity where s is the electrical conductivity of the saturated bulk sediment, „ is that of the interstitial
solution, and the Archie exponent n ~ 1-2. Archie's law is applicable when the conductivity of the
interstitial solution is much higher than that of the sediment particles, so those surface conduction
phenomena are insignificant.
At off-shore locations where groundwater discharge is negligible, resistivity measurements of the
sea water and sediments saturated by water of identical salinity can be used to determine the formation
factor and infer the porosity from Archie's law (Aller 1982). In sediments where freshening of the pore
spaces has occurred due to groundwater discharge, the measurements usually show a decrease of
electrical resistivity with depth, which provides important qualitative constraints on the increase of
salinity in the pore fluid and the depth range over which the transition from fresh to sea water occurs.
138
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Several different electrode configurations are commonly used to measure electrical resistivity. Our
project employs a direct contact probe (Figure 1) arranged in a Wenner array, which is a non-linear
array with the potential electrodes placed close together and evenly spaced. This configuration
provides discrete measurements even if good contact is not always maintained. Electrode A and B are
the positive and negative electrodes that measure the current /, and M and N are the electrodes that
measure the voltage drop V (Figure 2).
Shielded cable
1" (25mm) O.D. steel
Engineering type
- provides mechanical
and electrical
Electrode M
Electrode N
HectrodeB
—> Contact
V^Tapered
enhance soil
Figure 1.
Figure 2.
The off-shore horizontal extent of the interface was delineated by direct contact resistivity
measurements. The resistivity probe was driven manually into the bay bottom by scuba divers at six-
inch increments. The unit's string pot (that was originally designed for use with a Geoprobe percussion
drill keeps track of the depth measurement automatically and also trigger the electrical measurement)
had to be modified accordingly. The string pot was mounted on a jig and manually moved along a
displacement that would coincide with the depth that the probe was being driven into the bottom
sediments. Resistivity measurements were also simultaneously triggered manually. After the resistivity
was logged, the diver then drove the probe to the next six-inch level. This continued until a freshwater
zone was contacted or the probe had been driven to a maximum depth of 4 ft. The diver then moved on
to the next off-shore position at a horizontal spacing of ~30 feet, and the manual probing and logging
operations were repeated.
Cross-sectional plots as shown in Figure 3 where prepared from the field measured resistivity of
bay bottom sediments off shore. The blue (dark) areas indicate the location of fresh water outcrops off-
shore and the red (light) areas indicate only saltwater is present.
DESCRIPTION OF GROUNDWATER UNDERFLOW DEVICE
In recent years ultrasonic flow meters have been developed and used to measure relatively low
flow rates in a variety of water and wastewater industries. In this study we take advantage of this
advancement in ultrasonic technology to develop a seepage meter for continuous measurement of
submarine groundwater discharge.
139
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Fresh/ Salt Water Interface at West Neck Bay:
cross-sectional profile from, electrical logging
tUSTAMCE FHW M^AN TIQEMARK
data collection at horizontal spacing
of 25' and depth interval of 0.5'
electrical conductance of
sea water: - 33,000
Figure 3.
A transient-time flow meter uses the effect of the flow on the travel time of an ultrasonic signal as
the bases for determining the flow rate. Figure 4 shows a cross section of the meter with the path of the
flow tube sonic beam and non-intrusive transducers. A multi-pulse sonic signal is transmitted through
the flow tube in both directions by transducers located at opposite ends. When there is no flow the
signal will arrive at each transducer at the same time. However, when there is flow in the tube the
upstream flow will cause the signal to arrive ahead of the signal traveling downstream. The difference
in transit times (t) between the two signals is proportional to the liquid's flow velocity Vf. The constant
of proportionality depends on the average of the upstream and downstream transit times and length L
of the tube. The specific discharge from the
seepage surface q is inferred from the flow
velocity by multiplying Vfby the ratio between
the areas of the flow tube and the collection
funnel.
Our seepage meter system is based on the
widely used technique of placing a funnel (24" x
24") into the seepage surface on the seabed to
Flow Outlet
Bi-Directional Sonic Beam Path
(fasler from left due to flow)
Figure 4.
capture submarine groundwater discharge (Figure 5). This seepage flow is then directed via tubing
through the ultrasonic meter, which is connected to a data logger. The sampling frequency and duration
are programmed into the logger by the investigator. This meter can resolve seepage rates on the order
of 10"6cm/s, detect reversals in flow, and it includes a totalizer which acquires data on cumulative
volume of water passing through the meter which (when after normalized by the collection funnel area)
provides the specific discharge.
CONCLUSION
This study has demonstrated the feasibility of using a methodology that integrates geophysical
logging and transient seepage measurement to map out the spatial distribution of seepage and measure
the underflow discharge in real time. Conductivity values for the saturated sediments were observed to
range from 280 u.S/cm in freshwater zones to 12,800 p,S/cm in zones with high salinity. The discharge
zone at West Neck Bay, Shelter Island was observed to extend to 10-75 feet offshore. Electrical
140
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
CAPTURE
FUNNEL
AND
FLOW
TUBE
conductivity profiles of the coastal system obtained by direct-
contact resistivity logging delineates in cross-section the subset
of pore water that has been subjected to significant freshening,
and in turn they provide important constraints on the geometry
of the fresh/salt water interface and the mechanisms of mixing.
While the resistivity logging is very effective in identifying key
areas with pronounced seepage, continuous measurements using
the ultrasonic seepage meter provide high-resolution data on the
discharge in real time. Relatively high seepage velocities
ranging from 1.27 x 10'3 cm/s to 3.94 x 10'5cm/s (with a mean
value equivalent to 16 £/m2/d) were measured in the study area
(Figure 6). The input of underflow to the hydrological budget
was evaluated. Integrating over the projected area of the
seepage zone, the total daily discharge was estimated to be 1.72
x 106 I/day for the northeast section of West Neck Bay. This
estimate of underflow discharge is comparable to the recharge
in the contributing area, estimated to be 1.50 x 106 £/day for this
section of the bay (Schubert, 1998).
Figure 5.
The techniques used for this
investigation were effective in defining
the spatial extent of the off shore
groundwater seepage zone and in
measuring the flux within that zone
(Shaw and Prepas, 1989). The
investigator working within this
dynamic and spatially variable zone are
advised to use extreme caution in their
interpretation of any field measure-
ments. Adequate numbers and
replications of groundwater seepage
measurements need to be taken in order
to in order to address the spatial
variability of the bay bottom and
SEEPAGE VELOCITY VS TIDE STAGE -WEST NECK BAY 8A27/97
-TTDESTAGE
-SEEPAGE vHjt>crrY(Cf#S)
14 IS I*
TIME(HR)
Figure 6.
changes in near shore hydraulic gradients associated with tidal flux and precipitation.
REFERENCES
Aller, R.C., ,1982. Diffusion coefficients in near shore marine sediments, Limnol. Oceanogr., v. 27 no.
3, pp. 552-556.
Archie, G.E., 1942. The electrical resistivity log as an aid in determining some reservoir
characteristics, Trans. Am. Inst. Min. Metall. Pet. Eng., v. 146, pp. 54-62.
Gueguen, Y. and V. Palciauskas, 1994. Introduction to the phvsics of rocks, Princeton Univ. Press,
Chapter 8.
Paulsen, R.J., C.F. Smith, and T.-f. Wong, 1997. Development and evaluation of an ultrasonic
groundwater seepage meter, in Geology of Long Island and Metropolitan New York, pp. 88-97.
141
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Shaw, R.D. and RE. Prepas, 1989. Anomalous short term influx of water into seepage meters,
Limnology and Oceanography, v. 34, no. 7, pp. 1343-1351.
Shaw, R.D. and E.E. Prepas, 1990. Groundwater-lake interactions: I. Accuracy of seepage meter
estimates of lake seepage, Journal of Hydrology, v. 119, pp. 105-120.
Schubert, S. E., 1998. Areas contributing groundwater to the Peconic estuary and groundwater budgets
for North and South Forks and Shelter Island, Eastern Suffolk County, New York, U.S. Geological
Survey Water Resources Investigation ,97-4136.
142
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Influence of Stream Orientation on Contaminated
Ground-Water Discharge
By Don A. Vroblesky
The discharge zones of contaminated ground water are often beneath surface-water bodies. Such
discharge zones have been identified by using analyses of bottom-sediment gas bubbles (Vroblesky
and Lorah, 1991) and have been mapped using diffusion samplers in shallow Coastal Plain aquifers
(Vroblesky and others, 1991; 1992) and in a fractured-rock aquifer (Vroblesky and others, 1996). In
this paper, analyses of diffusion-samplers buried beneath a gaining stream were used to investigate the
relation between stream orientation and ground-water contaminant discharge.
Many environmental factors have the potential to influence the locations of contaminant discharge
to surface water. These factors include preferential ground-water flow from one side of the creek,
channeling of contaminants by fractures or other permeability heterogeneities, or capture of
contaminants by vegetation near the stream. The purpose of this report is to present preliminary data
suggesting that stream-channel orientation relative to the ground-water flow direction also is an
influential factor affecting contaminant discharge from ground water to surface water.
The approach used to delineate areas of contaminated ground-water discharge to surface water
involved the installation of diffusion samplers in the bottom sediment of Huntington Downs Creek, in
Greenville, South Carolina. The diffusion samplers consisted of 40-m(! (milliliter) glass vials enclosed
in scalable polyethylene bags. Preparation, burial, recovery, and analysis of the diffusion samplers is
described in an earlier study (Vroblesky and others, 1996). Samplers were buried approximately 1 ft
(0.3 m) deep in the creek-bottom sediment, which consisted of sand or weathered rock. The origin
point of the diffusion-sampler array in Huntington Downs Creek was a spring at the origin of the creek
(Figure 1). The total length of the array was approximately 400 ft (122 m), ending at the discharge
culvert to Huntington Downs Pond. The distance between samplers in each line was approximately 10
ft (3.048 m). The site identification numbers for the diffusion-sampler locations were the distance, in
feet, along the creek channel downstream from the spring at the origin of the creek.
The diffusion samplers were installed in the creek bottom sediment on May 25,1994, and
recovered on June 1, 1994. Immediately upon removal of each diffusion sampler from the bottom
sediment, the outer polyethylene membrane covering the vial opening was cut open, leaving the inner
polyethylene membrane intact. A cap was screwed onto the diffusion sampler over the inner
polyethylene membrane (the polyethylene was between the glass vial and the Teflon1 septum of the
cap). The samples were analyzed for volatile organic compound (VOC) content within 48 hours of
sample collection. A 50-(j.L vapor sample was slowly withdrawn and immediately analyzed on a
Photovac1 10S55 field gas chromatograph (GC), equipped with a capillary and a packed column. The
total VOC analysis was accomplished by calibrating the GC against a 10-ppm vapor standard of
trichloroethene (TCE). In this investigation, the difference between total VOC concentrations detected
in diffusion samplers and their respective replicates ranged from 5 to 21 percent.
'The use of tradenames does not imply endorsement by the U.S. Geological Survey.
143
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Water-level measurements from an observation well near the source of the creek and an
observation well downgradient from Huntington Downs Pond, and extrapolation of water-table
contours based on measurements made in wells approximately 350 to 550 ft south of the creek (Rust
Environment and Infrastructure, 1995) show that ground water near the creek moves generally north to
northwestwardly, discharging to ponds and streams (Figure 1). The water-table distribution derived
from these measurements is used in this investigation to show the probable dominant directions of
ground-water flow in the vicinity of Huntington Downs Creek (Figure 2).
The shallow aquifer is composed of a silty clay saprolite containing relict metamorphic structures
and rock fabrics. Estimates of hydraulic conductivity in the saprolite aquifer in the study area range
from less than 2 ft/d (Rust Environment and Infrastructure, 1995) to about 3.2 ft/d Kubal-Furr and
Associates, 1996). The estimated rate of ground-water flow in the saprolite aquifer (average thickness
of approximately 40 ft) is about 76 ft/yr (Kubal-Furr and Associates, 1996). Steeply dipping relict
Regional
ground-watcr-
flow direction
Figure 1. Ground-water flow directions near Huntington Downs
Creek (Modified from Rust Environment and Infrastructure, 1995).
fractures and foliation planes are present in the saprolite and appear to influence the direction of
contaminant transport (Rust Environment and Infrastructure, 1995). Based on lithologic observations
made during an excavation of the spring in August 1994, the spring appears to be the discharge zone of
water-bearing relict fractures in the saprolite.
Analysis of the vapor in the diffusion samplers for total VOCs showed that the concentrations, as
vapor, ranged from 0.3 to 21.2 ppm relative to TCE. Analysis of the vapor in diffusion samplers from
144
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
B .
Jg 20
§E"
'3-2 15
SH (D
5l 10
•a -g
6 a s
^ „
. , . . ,
-
-
_
-I
ll.H
....,,.
1
,1 III
|
I . , . ^ | . , . | . . . .
/Yi
* y *""
l& fai?zi
MM-h 4
•
II 1
& X;*
j-^1-^
l.rl,1
1
-§ 400 350 300 250 200^ ^15g 100 50
E-< Distance DownstreamTrom Spring (feet)
.
•
0
To
Huntmgton
Downs
Pond
EXPLANATION
200 Diffusion-Sampler
i Location and Distance
Downstream From
Spring, in Feet
Inflow from
Parking Lot
Drainage
'Spring
Figure 2. Concentrations of volatile organic compounds (VOCs) detected in
vapor diffusion samplers beneath Huntington Downs Creek (A), and locations of
samplers in creek (B), June 1994.
locations 21, 72, 110, 200, and 270 ft downstream from the spring, using the gas chromatograph
coupled to a capillary column, indicated the presence of TCE and tetrachloroethene (PCE). A similar
analysis on the vapor in the diffusion sampler from 349 ft downstream from the spring showed no
detectable concentrations of TCE or PCE. Because the only known source for VOC contamination is
south of the creek, the TCE and PCE detected in the diffusion samplers are considered to represent
contaminated ground-water discharge from the southern side of the stream.
The highest concentrations of VOCs (16.6 to 21.2 ppm) detected in the diffusion samplers beneath
the stream were found near the origin of the creek, approximately 16 ft downstream from the spring
(Figure 2). The presence of a spring indicates that there is an upward hydraulic gradient from the
ground water into the surface water. Therefore, the relatively high concentration of total VOCs in the
ground water beneath the stream immediately downstream from the spring implies that contaminant
discharge to the creek involved seepage of contaminated ground water upward through the stream bed
as well as movement of contaminants into the stream via the spring.
The distribution of VOC concentrations in the diffusion samplers beneath the stream showed a
general downstream decrease in the maximum concentrations detected (Figure 2). The concentrations
of TCE and PCE in diffusion samplers also decreased downstream. Thus, the downstream decrease in
145
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
maximum detected concentrations of total VOCs probably reflects a decrease in concentrations of
ground-water contaminants along a flowpath away from the source. The overall distribution of VOC
concentrations in the diffusion samplers, however, showed an irregular pattern, with areas of relatively
low concentrations between areas of relatively high concentrations (Figure 2). For example, the
concentrations of total VOCs in diffusion samplers in the reach between sampling points 140 and 190
were consistently less than 3 ppm; however, concentrations of total VOCs in diffusion samplers
immediately upstream and downstream from that reach were greater than 10 ppm. In fact,
concentrations of total VOCs in diffusion samplers between sampling points 140 and 180 were below
the apparent background concentration measured at location 349, implying that there was no
substantial discharge of chlorinated VOCs to the creek in that reach. A unique feature of the creek
reach between sampling points 140 and 180 is that it is oriented approximately parallel to the probable
dominant direction of ground-water flow. In contrast, the creek reach encompassing sampling points
50 to 90 and 200 to 250 are oriented at a sharper angle to the probable dominant direction of ground-
water flow (Figure 2B). VOC concentrations in most diffusion samplers from those reaches were
higher than in the reach encompassing sampling points 140 to 180 (Figure 2A).
A variety of factors potentially affect the distribution of contaminant discharge to a stream. Not all
of the factors were identified in this preliminary investigation. For example, potential influences
include bed-sediment heterogeneities, relict fractures, vegetation, and hyporheic zones. However, the
uniformly low concentrations of total VOCs in the channel reach oriented approximately parallel to the
probable dominant direction of ground-water flow (between diffusion samplers 140 and 190)
compared to adjacent reaches oriented at a sharper angle to ground-water flow, strongly suggest that
orientation of the creek exerts a major influence on contaminant discharge. Reaches of the creek that
transect the axis of the contamination plume receive greater contaminant and ground-water discharge
than the reach oriented along the axis of contaminant transport.
In conclusion, diffusion samplers placed beneath creek-bed sediments were used to detect the
distribution of VOC contamination discharging from ground water to Huntington Downs Creek, in
Greenville, South Carolina. The uniformly low concentrations of total VOCs in the channel reach
oriented approximately parallel to the probable dominant direction of ground-water flow, between
diffusion samplers 140 and 190, compared to adjacent reaches oriented at a sharper angle to ground-
water flow, strongly suggest that orientation of the creek is a major influence on contaminant
discharge. Reaches of the creek that transect the axis of the contamination plume appear to receive
greater contaminant discharge than a reach oriented along the axis of contaminant transport. These data
imply that site investigators attempting to locate zones of ground-water contaminant discharge to
surface water in meandering streams should put particular emphasis on reaches transecting the
dominant direction of contaminated-ground-water transport.
REFERENCES
Kuball-Furr and Associates, 1996.1995 Annual ground-water monitoring report, Consultants report to
General Electric Company, Greenville, South Carolina, 35 pp.
Rust Environment and Infrastructure, 1993. Analyses of surface water samples from the North End,
October 12,1993 through April 6,1994, Consultant's data reported to the General Electric
Company, Greenville, South Carolina.
Rust Envkonment and Infrastructure, 1994. Data from the April 1,1994 sampling of the North End,
Consultant's data reported to the General Electric Company, Greenville, South Carolina.
Rust Environment and Infrastructure, 1995. North end off-site area environmental investigation report,
Consultant's report to the General Electric Company, March 1995,25 pp.
146
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Vroblesky, D.A. and M.M. Lorah, 1991. Prospecting for zones of contaminated ground-water
discharge to streams using bottom-sediment gas bubbles, Ground Water, v. 29, no. 3, pp. 333-340.
Vroblesky, D.A., M.M. Lorah, and S.P. Trimble, 1991. Mapping zones of contaminated ground-water
discharge using creek-bottom- sediment vapor samplers, Aberdeen Proving Ground, Maryland,
Ground Water, v. 29, no. 1, pp. 7-12.
Vroblesky, D.A., J.F. Robertson, Mario Fernandez, and C.M. Aelion, 1992. The permeable-membrane
method of passive soil-gas collection, in Proceedings of the Sixth National Outdoor Action
Conference: National Water Well Assoc., May 5-13,1992, Las Vegas, NV, pp. 3-16.
Vroblesky, D.A., L.C. Rhodes, J.F. Robertson, and J.A. Harrigan, 1996. Locating VOC contamination
in a fractured-rock aquifer at the ground-water/surface-water interface using passive vapor
collectors:, Ground Water, v. 34, no. 2, pp. 223-230.
AUTHOR INFORMATION
Don A. Vroblesky, U.S. Geological Survey, Columbia, S.C.
147
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Factors Controlling Hyporheic Exchange in a
Southern Ontario Stream: Modeling Riffle-Scale
Patterns in Three Dimensions Using MODFLOW
By R.G. Storey, D.D. Williams, and K.W.F. Howard
INTRODUCTION
The general pattern of exchange flow between stream surface waters and their hyporheic zones
(Fig.l) has been observed in numerous field situations (e.g., Valett, et al., 1994), and in laboratory
flume studies (e.g., Thibodeaux and Boyle, 1987). These latter studies have shown that the
downwelling/upwelling sequence is produced by a rise in hydraulic head of the surface water as it rises
over the surface of a riffle, and a rapid drop in head as surface water flows down the downstream slope
of the riffle. Thus from a surface water point of view, the forces controlling the riffle-scale pattern of
exchange flow have been adequately explained.
However, in the field situation hyporheic zones are subject not only to the forces applied by
surface waters, but also to hydraulic gradients associated with the underlying aquifer. Stream reaches
are rarely hydraulically neutral, but either receive net ground water inflow or export water to the
High surface head
Downwelling zone
Low surface head
Upwelling zone
Pool
Fig. 1. Subsurface flow paths associated with pool/riffle sequences in rivers.
aquifer. How then do riffle-scale exchange flows persist without being overwhelmed by net ground
water movements?
The aims of this study were first to determine the local geological and hydrogeological conditions
necessary to produce riffle-scale exchange flows within a given regional hydrogeological environment.
148
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop July 2000
A secondary aim was to determine how hyporheic exchange flows would be affected by natural
changes or artificial disruptions to aspects of the aquifer, stream flow or stream bed.
METHODS
The Speed River, in Southwestern Ontario, is a low-gradient stream, approximately 6 m wide and
0.15-0.35 m deep at the study site. The primary aquifer, a dolomite bedrock, is overlain by 20 m of low
permeability glacial kame deposits. The stream itself lies in a bed of modern alluvium 1-1.5 m deep,
and extending 30-40 m wide on each side of the stream, which has a very high permeability.
An area of the river catchment, 1,000 m x 500 m, was modeled using MODFLOW (USGS), a
three-dimensional finite difference ground water flow model. The two lateral boundaries of the model
were defined by the Speed River catchment boundaries, and the upper and lower boundaries followed
ground water flowlines. The model focused on a single riffle of the stream, a 10 m long section.
Stream stage was defined by constant head boundaries in the top layer of the model, with a rise and fall
of a few cm over the length of the riffle site.
RESULTS AND DISCUSSION
Initially the model was run without inserting the high conductivity alluvial deposits. In these runs
hyporheic flows everywhere were upwelling, and towards the stream laterally, even when aquifer
heads were low in summer.
When the zone of high permeability, representing alluvial deposits, was inserted around the stream,
flow patterns changed dramatically. Hydraulic gradients within the near-stream zone became very low
(<1 cm per m) and small differences of <3 cm in surface hydraulic head between upstream and
downstream ends of the riffle produced downwelling and lateral exchange flows in the subsurface.
This occurred even though hydraulic gradients in the material surrounding the alluvium were strongly
towards the stream.
These results show that in a low gradient stream system with strong hydraulic gradients from the
catchment towards the stream, exchange flows can still occur in a zone of high- permeability alluvium.
The essential feature of this alluvium was that it lowered hydraulic gradients within the hyporheic zone
to within the range of variation shown by the stream surface heads as they flowed between pools and
riffles. Thus these variations were able to alter flow paths up to 1.5 m deep in the stream bed.
In reducing the vertical hydraulic gradient, the alluvium changed the dominant hydraulic gradient
beneath the stream bed from almost vertical to almost horizontal. This meant that flows within the
alluvium were driven by hydraulic heads downstream rather than by those in the underlying aquifer
and surrounding catchment. The large flux of water flowing downstream within the alluvial sediments
was able to supply or withdraw sufficient water to support lateral and vertical exchange flows between
the stream and alluvial sediments, independently of hydraulic heads beneath and to the sides of the
alluvium.
Sensitivity of exchange flows to changes in system conditions
In this simulation, the highly permeable alluvial sediments allowed exchange flows in the
hyporheic zone to operate somewhat independently of the heads in the aquifer. Thus exchange flows
were relatively insensitive to changes in aquifer heads; however field data showed that a large increase
149
-------
Proceedings of the Ground- 'Water/Surface- Water Interactions Workshop
July 2000
of several meters in-aquifer heads from summer to fall did result in loss of exchange flows. Exchange
flows are sensitive, however, to a decrease in permeability of the alluvium, as may result from siltation
of the river bed.
Importance of alluvial sediments to exchange flows in other stream systems
Do all streams that receive net input of ground water from their catchments, require a zone of high
conductivity alluvium in order for exchange flows to occur? This depends on the steepness of the
stream, which determines the hydraulic gradient between upstream and downstream ends of a riffle,
and the hydraulic gradient in the aquifer. In a steeper stream, or one with weaker hydraulic gradients
between aquifer and stream, exchange flows can occur in less permeable near- stream sediments.
CONCLUSIONS
Modeling on a small scale has shown that surface water head-differences of a few centimeters
between riffles and pools can produce exchange flows within permeable alluvial sediments, despite net
discharge of ground water to the stream. This model reveals local interactions between surface water
and ground water which would not be predicted by larger scale models, but which have important
chemical and biological consequences for the stream and ground water systems.
REFERENCES
Thibodeaux, LJ. and J.D. Boyle, 1987. Bedform-generated convective transport in bottom sediment,
Nature, v. 325, pp. 341-343.
Valett, H.M., Fisher, S.G., Grimm, N.B., and P. Camill, 1994. Vertical hydrologic exchange and
ecological stability of a desert stream ecosystem, Ecology, v. 75, pp. 548-560.
AUTHOR INFORMATION
R.G. Storey and D.D. Williams, Division of Life Sciences, University of Toronto at Scarborough, 1265
Military Trail, Scarborough Ontario, MIC 1A4, Canada.
K.W.F. Howard, Division of Physical Sciences, University of Toronto at Scarborough, 1265 Military
Trail, Scarborough Ontario, MIC 1A4, Canada.
_
150
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Solute and Solid Phase Relationships in the
Surface Hyporheic Zone of a Metal Contaminated
Stream, Silver Bow Creek, MT
By Johnnie N. Moore and William W. Woessner
INTRODUCTION
Transformations occurring within the hyporheic zone impact the chemistry of both the overlying
surface waters and the underlying ground water systems. The hyporheic zone is a key ecological niche
crucial to the health of stream biota, as well as a major site of exchange, metabolism, and storage of
particulates and solutes in rivers. There is, therefore, a critical need to understand the nature of the
geochemistry that governs the transformation of materials through the hyporheic zone, especially
metals and metalloids that are toxic to aquatic organisms utilizing the hyporheic zone.
The complexities of the substrate in most streams/rivers make determining the relationships
between solute and solid phases difficult. To address this issue we have used ceramic beads as an
artificial substrate to examine what solid phases are in equilibrium within the shallow hyporheic zone.
Combined with "mini-tube wells," this inexpensive and rapid method can be used to examine solute
and solid phase components in essentially any stream/river.
Site Conditions
Silver Bow Creek, at the headwaters of the Clark Fork River, has received contaminated surface
water and sediments from the Butte gold, silver and base metal mining and processing region for over
a century (Moore and Luoma, 1991). Over 100 million tons of tailings and mining wastes were
released into Silver Bow Creek, a portion of which were deposited at the Miles Crossing Research Site
located 18 Km down stream (Figure 1).
At the Miles Crossing Site, Silver Bow Creek has an average discharge of 850 L/s. Its water is near
neutral, high in dissolved oxygen and relative low in dissolved metals (Table 1). The fluvial plan is
covered with up to 2 m of metal rich mine tailings highly elevated in arsenic, cadmium, copper, iron,
lead, manganese and zinc.
Groundwater flow is generally in the direction of the sloping fluvial plain (Woessner, 1998; Smart,
1995; Shay, 1997) (Figure 1). Groundwater within the fluvial plan is acidic and contaminated (Table
1). On the large scale, groundwater exchange with the creek occurs as the stream stage falls below the
fluvial plain water table. In the study area reaches of flow through channel (Site 1 and 3) and zero
exchange (parallel flow)(Site 2) were observed.
151
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
(604.5
Water Table Elk (m)
Figure 1. Location map of the Miles Crossing Research Site including Silver Bow Creek
and the associated fluvial plain (zone of stream deposited higher hydraulic conductivity
sediments. Groundwater flow is parallel to the fluvial plain and down valley. Bed tubes
and mini tube wells were installed in three portions of the stream bed, Sites 1, 2 and 3.
Definition of the Hyporheic Zone
The hyporheic zone is generally defined as the saturated zone beneath and associated with the
stream channel that shares some biological, chemical or physical characteristics with the surface water
Williams, Triska, et al., 1989; Valett, et al., 1990; Hendricks and White, 1991; Valett, 1993). Our work
uses geochemistry to define this transition zone between 100 % surface water and 100% groundwater.
We further focus part of our effort on the. "surface" hyporheic zone defined as the transition zone
within 30 cm of the streambed at our site. Benner (1995) and Benner, et al. (1995) described a
geochemical transition zone extending to a depth of one or more meters at Site 1. Nagorski (1997)
continued work at the Site 1 and two additional sites focusing her effort on conditions within 30 cm of
the streambed.
152
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Table 1.
Water Quality
Constituent
pH
D.O.
Cond.
Alkal.
Cl
NO3-N
SO4
Al
As
Ca
Cd
Cu
Fe
Mg
Mn
Na
Pb
Zn
Silver Bow Creek
(Nagorski, 1997)
Std. Units and mg/^
7.7
7.2
0.424
1.3E+02
14
1.3
78.4
(<0.07)
0.010
47
(<0.01)
0.136
0.22
10.4
0.90
23
(<0.1)
0.645
Fluvial Plain
Groundwater (Benner,
1995; Shay, 1997)
Std. Units and mg/t
4.2-4.9
<1
2.0
0
20
1-2
1487
33
0.010
141
0.55
19
365
33
28
42
(<0.06)
54
METHODOLOGY
Bead Tube Samplers
Bead tube samplers are 40-175 cm long polycarbonate tubing (1cm OD, 0.6 cm ID) slotted with a
1 mm width ban sawed on two sides at 3 mm intervals were filled with aluminosilicate beads (2 mm
average diameter). Plastic dividers were inserted into the columns at 10-cm intervals to minimize
vertical migration of water in the samplers. Completed bead tubes were then acid cleaned in 20%
reagent grade HC1 for two hours and rinsed repeatedly with sterilized deionized water until a pH of 5
was reached.
Bead tube samplers were inserted into the bed by driving a dual tube steel rod into the stream
sediments, removing the center solid rod, inserting the bead tube and then removing the outer steel
tube. Sediments were allowed to collapse around the sampler. The tubes were installed so that about 10
cm of the sampler extended above the stream bed. Bead tubes were retrieved after 42 to 52 days.
Retrieved tubes were rinsed in the field to remove excess sediment, labeled, photographed,
wrapped in plastic and stored. In the laboratory, columns were oven dried at 70 C and sectioned into 4
to 7 segments, depending on the amount of visible coating. Approximately one gram of beads were
placed in an acid-washed centrifuge tube to which 10 m« of 40% metal-grade HC1 was added. The
sample was shaken for 1 hour and centrifuged for 10 minutes. Solutions were analyzed for major
metals using ICAPES using standard procedures.
153
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
"Mini Tube Wells"
Small diameter tube wells were constructed using 0.95 cm diameter polyethylene with 5 cm of the
tip slotted and covered with a nylon mesh screen. These groundwater sampling devices were installed
as single instruments or nests using the same driven dual tube method used to install the bead tubes.
These small diameter tubes were sampled using a 60 cc acid-washed syringe after purging at lease
one tube volume and one syringe volume. Tubes were then tightly capped to keep them full of bed
water. The syringe-collected samples were pushed through a 0.45 urn acid-washed filter and into two
acid washed sample bottle. Bed water samples for cation analyses were preserved with trace metal
grade HNO3 and samples for anion analyses were untreated. Samples were placed on ice and returned
to the lab for standard 1C and ICAPES analyses. DO, pH, specific conductance and alkalinity were
determined in the field.
RESULTS
At Site 1, surface water is alkaline, oxygenated and contains relatively high concentrations of
nitrate and low concentrations of sulfate; ground water is acid and has low concentrations of oxygen,
and high sulfate. The bead tube data allows for a higher resolution of the complexities of the hyporheic
zone (transition zone) than tube wells as interfaces can be resolved by examining the continuous bead
tube. Iron is a major control due to the precipitation of Fe-oxyhydroxides at the interfaces between the
surface water and ground water with the hyporheic zone. Solute Fe concentrations are relatively low in
the surface water and hyporheic zone water, with a concomitant elevation in the solid phase Fe on the
beads. Two sets of bead tube sections show that Fe precipitates at the surface water-hyporheic zone
boundary and at the ground water-hyporheic zone boundary. High values of solid phase Fe seen in the
surface water beads resulted from fine sediment particles attached to the bead surfaces, not from Fe-
oxyhydroxides precipitates.
Certain elements (As, Cu, Mo, P, Pb, Sr) are strongly related to Fe precipitation, but others are
offset (e.g., Mn and Zn). Surface water and ground water contain relatively low concentrations of As
and Fe, however, the surface hyporheic zone has elevated concentrations of As and Fe. We believe
these elevated concentrations result from the dissolution of Fe-oxyhydroxides that contain As at the pH
and dissolved oxygen levels found only in the surface hyporheic zone. At some sampling points Fe
precipitates were either absent of showed a complex interfingering.
CONCLUSIONS
Bead tubes and tube wells provided detailed geochemical data'in the near surface hyporheic zone
(30 cm). The presence of iron hydroxides and co precipitating As and metals was observed and
quantified. The bead tubes provided continuous sampling of the hyporheic zone and were useful in
establishing the presence of geochemical interfaces. These instruments are best suited for sampling of
groundwater and surface water systems with geochemical contrasts.
We can identify relationships at the 1-2 cm scale in the solid phase and at the 10 cm scale in the
solute phase using these methods. We think that improvements in resolution could be obtained by
constructing bead samplers with smaller beads and using high-resolution "peepers" (dialysis membrane
samplers) for solute sampling. This modification should resolve changes within the hyporheic zone at
the cm or possibly the sub-cm scale. However, the use of peepers would increase the cost and
154
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
complexity of the technique. For many problems the method we present should offer a rapid and
relatively easy technique to examine the fine scale relationships within the hyporheic zone.
REFERENCES
Benner, S.G., 1995. Geochemical processes in a transition zone between surface water and acidic,
metal-rich groundwater. M.S. Thesis, Univ. of Montana, Dept. of Geology, Missoula, 71 pp.
Benner, S.G., Moore, J.N., and E. Smart, 1995. Geochemical processes in a transitions zone between
surface water and acidic, metal-rich groundwater. ES&T, v. 29, pp. 1789-1795.
Moore, J.N. and Luoma, S.N., 1991. Hazardous wastes from large-scale metal extraction. ES&T, 24,
pp. 1279-1284.
Hendricks, S.P. and D.S. White, 1991. Physiochemical patterns within a hyporheic zone of a northern
Michigan river, with comments on surface water patterns. Can. J. Fish. Aquat. ScL, v. 48, pp.
1645-1654.
Nagorski, S.A., 1997. Impacts by acidic, metals-rich groundwater on the hyporheic zone of an
intermontane stream. M.S. Thesis, Univ. of Montana, Dept. of Geology, Missoula, 137 pp.
Shay, D.T., 1997. An Investigation of the hydrogeology and geochemistry of a floodplain aquifer
system impacted by mine tailings, Silver Bow Creek, Montana. M.S. Thesis, University of
Montana, Dept. of Geology, Missoula, 147 pp.
Smart, E.W., 1995. Surface water and groundwater interaction in a shallow unconfined alluvial aquifer
and small mountain stream Silver Bow Creek, Montana. M.S. Thesis, University of Montana,
Dept. of Geology, Missoula, 170 pp.
Triska, F.J., Kennedy, V.C., Avanzino, R.J., Zellweger, G.W., and K.E. Bencala, 1989. Retention and
transport of nutrients in a third-order stream in northwestern California: hyporheic processes.
Ecology, v. 70, pp. 1893-1905.
Valett, H.M., 1993. Surface-hyporheic interactions in a Sonoran Desert stream: Hydrologic exchange
and diel periodicity. Hydrobio., v. 259, pp. 133-144.
Valett, H. M., Fisher, S. G. and E.H. Stanley, 1990. Physical and chemical characteristics of the
hyporheic zone of a Sonoran Desert stream. /. N. Am. Benthol. Soc., v. 9, no. 3, pp. 201-205.
Williams, D.D. and H.B.N. Hynes, 1974. The occurrence of benthos deep in the substratum of a
stream. Freskwat. Biol, v. 4, pp. 233-256.
Woessner, W.W., 1998. Changing views of stream-groundwater interaction. Eds. J. Van Brahana, U.
Eckstein, L. K. Ongley, R. Schneider and J. E. Moore. Proceedings of the Joint Meeting of the
XXVUI Congress of the International Association of Hydrogeologists and the Annual Meeting of
the American Institute of Hydrology: Gambling With Groundwater, Las Vegas, Nevada, Sept. 28-
October 2, 1998, AIH, St. Paul, MN., p. 1-6.
AUTHOR INFORMATION
Johnnie N. Moore and William W. Woessner, Department of Geology, University of Montana,
Missoula, MT 59801.
155
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
APPENDICES
156
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Appendix A: Workshop Participants List
Tom Aalto .
U.S. EPA, Region 8
999 18th St.
Mail Code: 8P2-HW
Denver, CO 80202-2413
Phone: (303) 312-6949
Fax: (303) 312-6064
aalto .torn @ epa. go v
James Bartolino
U.S. Geological Survey
4501 Indian School Rd., Ste 200
Albuquerque, NM 87110-3929
Phone: (505) 262-5336
Fax: (505) 262-5398
jbartol@usgs.gov
Katherine Baylor
U.S. EPA, Region 9
RCRA Corrective Action Office
75 Hawthorne St.
Mail Code: WST-5
San Francisco, CA 94105
Phone: (415) 744-2028
Fax: (415) 744-1044
baylor.katherine @epa. gov
Ned Black
U.S. EPA, Region 9
75 Hawthorne St.
Mail Code: SFD-8B
San Francisco, CA 94105
Phone: (415) 744-2354
Fax: (415) 744-1916
black.ned @ epa. gov
Randy Breeden
U.S. EPA, Region 8
999 18th St., Suite 500
Mail Code: 8P2-HW
Denver, CO 80202-2466
Phone: (303) 312-6522
Fax: (303)312-6064
breeden.randy@epa.gov
David Burris
Air Force Research Laboratory
139 Barnes Dr.
Mail Code: AFRL/MLQR
TyndallAFB,FL 32403
Phone: (850) 283-6035
Fax: (850) 283-6090
david.burris@mlq.afrl.af.mil
Allen Burton
Wright State University
Inst. for Environmental Quality
3640 Colonel Glenn Hwy.
Dayton, OH 45435-0001
Phone: (937) 775-2201
Fax: (937) 775-4997
aburton @ wright.edu
Judy Canova
South Carolina DepL of Environ-
mental Health and Conservation
2600 Bull St.
Columbia, SC 29201
Phone: (803) 896-4046
Fax: (803) 896-4292
canovail@columb34.dhec.state.sc
Lisa Capron
U.S. EPA , Region 5
77 W. Jackson Blvd.
Mail Code: DE-9J
Chicago, IL 60604-3507
Phone: (312) 886-0878
Fax: (312) 353-4342
capron.lisa® epa.gov
David Charters
U.S. EPA, Environmental
Response Team
2890 Woodbridge Ave.
Edison, NJ 08837-3679
Phone: (732) 906-6825
Fax: (732) 321-6724
charters .davidw @ epa. gov
Jungyill Choi
U.S. Geological Survey
430 National Center
12201 Sunrise Valley Dr.
Mail Code: MS-431
Reston,VA 20192
Phone: (703) 648-5472
Fax: (703) 648-5484
jchoi@usgs.gov
Brewster Conant Jr.
University of Waterloo
Earth Sciences Department
Waterloo, Ontario N2L 3G1
Phone: (519) 888-4567, x2973
bconanti ©sciborg.uwaterloo.ca
Martha Conklin
University of Arizona
Department of Hydrology and
Water Resources
P.O. Box 210011
Tucson, AZ 85721-0011
Phone: (520) 621-5829
Fax: (520) 621-1422
martha@hwr.arizona.edu
D. Reide Corbett
Florida State University
Department of Oceanography
P.O. Box 4320
Tallahassee, FL 32306-4320
Phone: (850) 644-9914
Fax: (850) 644-2581
rcorbett@ ocean.fsu.edu
157
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Cliff Dahm
University of New Mexico
Department of Biology
Albuquerque, NM 87131
Phone: (505) 277-2850
Fax: (505) 277-5355
cdahm @ sevilleta.unm.edu
Kathy Davies
U.S. EPA, Region 3
1650 Arch St.
Mail Code: 3HW41
Philadelphia, PA 19103-2029
Phone: (215) 814-3315
Fax: (215) 814-3015
davies.kathv@epa.gov
Joe Dlugosz
U.S. EPA, ORD/MED
6201 Congdon Blvd.
Duluth, MN 50408
Phone: (218) 529-5215
Fax: (218) 529-5003
dlugosz.joseph@epa.gov
Maureen Dudley
Denver Department of
Environmental Health
1391 Speer Blvd., Suite 700
Denver, CO 80204
Phone: (303) 285-4063
Fax: (303) 285-5621
dudlevm@ci.denver.co.us
Bruce Duncan
U.S. EPA, Region 10
1200 Sixth Ave.
Mail Code: OEA-095
Seattle, WA 98101
Phone: (206) 553-8086
duncan.bruce@epa.gov
Rene Fuentes
U.S. EPA, Region 10
1200 Sixth Ave.
Mail Code: OEA-095
Seattle, WA 98101
Phone: (206) 553-1599
Fax:(206)553-0119
fuentes.rene@epa.gov
Gayle Garman
National Oceanic and
Atmospheric Administration
7600 Sand Point Way, NE
Mail Code: Bin C15700
Seattle, WA 98115-0070
Phone: (206) 526-4542
Fax: (206) 526-6865
gayle. garman @ noaa. go v
Kevin Garon
DuPont Engineering
6324 Fairview Rd.
Charlotte, NC 28210
Phone: (704) 362-6635
Fax: (704) 362-6636
Kevin.p. garon @ usa.dupont.com
David Geist
Battelle Pacific Northwest
National Laboratory
P.O. Box 999
Mail Code: MS K6-85
Richland, WA 99352
Phone: (509) 372-0590
Fax: (509) 372-3515 - -
david. geist @ pnl. gov
Ron Gouguet
National Oceanic and
Atmospheric Administration
c/o U.S. EPA, Region 6
1445 Ross Ave.
Mail Code: 6SF-L
Dallas, TX 75202-2733
Phone: (214) 665-2232
Fax: (214) 665-6460
Ron gouguet crc6@
hazmatnoaa. gov
Chad Gubala
University of Toronto
The Scientific Assessment
Technologies Laboratory
3359 Mississauga Rd. North
Mississauga, Ontario, Canada
L5L 1C6
Phone: (905) 828-3863
Fax: (905) 828-5273
cgubala@credit.erin.utoronto.ca
Jack Guswa
HSI GeoTrans
6 Lancaster County Rd.
Harvard, MA 01451
Phone: (918) 772-7557
Fax: (918) 772-6183
jguswa@hsigeotrans.com
Mark Hartle
Pennsylvania Fish and Boat
Commission, Division of
Environmental Services
450 Robinson Ln.
Bellefonte, PA 16823
Phone:(814)359-5116
Fax: (814) 359-5175
mhartle @ fish. state.pa.us
S.M. Harrison
Wyoming State Department of
Health
Hathaway Bldg.
Cheyenne, WY 82002
Phone: (307) 777-6186
sharril @missc.state.wv.us
158
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Judson Harvey
U.S. Geological Survey
12201 Sunrise Valley Dr.
Mail Code: MS 430
Reston,VA 20192
Phone: (703) 648-5876
Fax: (703) 648-5484
jwharvey@usgs.gov
Susan Hendricks
Murray State University
206 Hancock Biological Station
Murray, KY 42071
Phone: (502) 474-2272
Fax: (502) 474-0120
susan.hendricks@murravstate.edu
Briant Kimball
U.S. Geological Survey
1745 W. 1700 S., Rm. 1016
Salt Lake City, UT 84104
Phone: (801) 975-3384
Fax: (801) 975-3424
bkimball@usgs.gov
David Lee
Atomic Energy of Canada, Ltd.
Environmental Research Branch
Mail Code: 51A
Chalk River, Ontario KOJ1PO
Phone: (613) 584-8811, x 4710
Fax: (613) 584-1221
leed@aecl.ca
James Huckins
U.S. Geological Survey
4200 New Haven Rd.
Columbia, MO 65201
Phone: (573) 875-5399, x!879
James Huckins@usgs.gov
Richard Jack
Wyoming Department of
Environmental Quality, Solid
Waste Permitting and Corrective
Action, Solid and Hazardous
Waste Division
250 Lincoln St.
Lander, WY 82520
Phone: (307) 332-6924
rjack@missc.state.wy.us
Jeff Johnson
U.S. EPA, Region 7
726 Minnesota Ave.
Mail Code: ARTD/RCAP
Kansas City, KS 66101
Phone: (913) 551-7849
Fax: (913) 551-7947
iohnson.ieff @ epa. gov
John Lendvay
University of Michigan
219 EWRE Building
Environmental Engineering
1351BealAve.
Ann Arbor, MI 48109-2125
Phone: (734) 764-6350
Fax: (734) 763-2275
lendvay@engin.umich.edu
Michelle Lorah
U.S. Geological Survey
8987 Yellow Brick Rd.
Baltimore, MD 21237
Phone: (410) 238-4301
mmlorah@usgs.gov
Vince Malott
U.S. EPA, Region 6
1445 Ross Ave.
Mail Code: 6SF-AP
Dallas, TX 75202
Phone: (214) 665-8313
Fax: (214) 665-6660
malott.vincent@epa.gov
Mary Matta
National Oceanic and
Atmospheric Administration
7600 Sand Point Way, NE
Mail Code: Bin C15700
Seattle, WA 98115
Phone: (206) 526-6315
Fax: (206) 526-6865
Mary Matta@hazmat.noaa.gov
Garry McKee
Wyoming PHL
Hathaway Bldg.
Cheyenne, WY 82003
Phone: (307) 777-7431
Gayle Miller
Wyoming Department of Health
2300 Capitol Ave., Rm. 427
Cheyenne, WY 82002
Phone: (307) 777-5596
Fax: (307) 777-5573
gmille@missc.state.wy.us
Scott Miller
Wyoming Department of
Environmental Quality, Solid
Waste Permitting and Corrective
Action, Solid and Hazardous
Waste Division
250 Lincoln St.
Lander, WY 82520
Phone: (307) 332-6924
smille@missc.state.wv.us
Johnnie Moore
University of Montana
Department of Geology
Missoula, MT 59812-1019
Phone: (406) 243-6807
Fax: (406) 243-4028
gl. inm@selway.umt.edu
159
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Mike Montoya
Ute and Ouray Indian Reservation
P.O. Box 190
FortDuchesne, UT 84066
Phone: (435) 722-0885
Fax: (435) 722-0885
utefish @ubtanet.com
Rich Muza
U.S. EPA, Region 8
999 18th St., Suite 500
Mail Code: 8EPR-EP
Denver, CO 80202-2466
Phone: (303) 312-6595
Fax:(303)312-6065
Tnuza.richard@epa.gov
Paul S. Osborne
U.S. EPA, Region 8
999 18th St., Suite 500
Mail Code: 8P-W-GW
Denver, CO 80202-2466
Phone: (303) 312-6125
osborne.paul @ epa. go v
Ronald Paulsen
Cornell University/Suffolk
County Health Services
3059 Sound Ave.
Riverhead, NY 11901
Phone: (516) 727-3910
Fax: (516) 369-5944
Dave Petrovski
U.S. EPA, Region 5
77 W. Jackson Blvd.
Chicago, JJL 60604
Phone: (312) 886-0997
Fax: (312) 353-9176
petrovskS .david @ epa. go v
Alan Polonsky
Denver Department of
Environmental Health
_
1391 Speer Blvd., Suite 700
Denver, CO 80204
Phone: (303) 285-4060
Fax: (303) 285-5621
polonskva@ci.denver.co.us
Lisa Rosman
National Oceanic and
Atmospheric Administration
290 Broadway, Rm. 1831
New York, NY 10007
Phone: (212) 637-3259
Fax: (212) 637-3253
Lisa.Rosman@noaa.gov &
rosman.lisa@epa.gov
Stephen SchmeUing
U.S. EPA, Robert S. Kerr
Environmental Research Center
Ada, OK 74821-1198
Phone: (580) 436-8540
schmelling.steve@epa.gov
Henry Schuver
U.S. EPA
1200 Pennsylvania Avenue, NW
Mail Code: 5303W
Washington, DC 20460
Phone: (703) 308-8656
Fax: (703) 308-8638
schuver.henry @ epa. go v
Debbie Sherer
U.S. EPA, Region 8
999 18th St., Suite 500 - <•
Mail Code: 8P-HW
Denver, CO 80202-2466
Phone: (303) 312-6429
sherer.deborah @ epa. gov
Christopher Smith
Cornell Cooperative Extension
3059 Sound Ave.
Riverhead, NY 11901
160
Phone: (516) 727-3910
Fax: (516) 369-5944
csmith@cce.cornell.edu
Pete Swarzenski
U.S. Geological Survey
Center for Coastal Geology
600 Fourth St. South
St. Petersburg, FL 33701
Phone: (727) 803-8747, x3072
Fax: (727) 803-2032
pswarzen @cfcg.er.usgs.gov
Jim Schwartz
Wyoming Department of
Agriculture
2219 Carey Ave.
Cheyenne, WY 82002
Phone: (307) 777-6591
Fax: (307) 777-6593
jschwa@missc.state.wy
Guy Tomassoni
U.S. EPA
Office of Solid Waste
1200 Pennsylvania Avenue, NW
Mail Code: 5303W
Washington, DC 20460
Phone: (703) 308-8622
Fax: (703) 308-8638
tomassoni.guv@epa.gov
Patti Tyler
U.S. EPA, Region 1
New England Regional
Laboratory
60 Westview St.
Mail Code: EGA
Lexington, MA 02421
Phone: (781) 860-4342
Fax: (781) 860-4397
tvler.patti @ epa. gov
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Luanne Vanderpool
U.S. EPA, Region 5
77 W. Jackson Blvd.
Mail Code: SR-6J
Chicago, IL 60604
Phone: (312) 353-9296
Fax: (312) 886-4071
vanderpooLluanne @ epa. gov
Don Vroblesky
U.S. Geological Survey
720 Gracern Rd.
Columbia, SC 29210-7651
Phone:(803)750-6115
vroblesk@usgs.gov
Ernie Waterman
U.S. EPA, Region 1
1 Congress St., Suite 1100
Mail Code: HBT
Boston, MA 02114-2023
Phone: (617) 918-1369
Fax:(617) 918-1291
waterman.ernest@epa.gov
Lynn Wellman
U.S. EPA, Region 4
61 Forsyth St.
Atlanta, GA 30303-3415
Phone: (404) 562-8647
Fax: (404) 562-8628
wellman. lynn @ epa. gov
Steve Wharton
U.S. EPA, Region 7
726 Minnesota Ave.
Mail Code: SUPR/FFS
Kansas City, KS 66101
Phone: (913) 551-7819
wharton.steve@epa.gov
Richard Willey
U.S. EPA, Region 1
1 Congress St., Suite 1100
Mail Code: HBS
Boston, MA 02114-2023
Phone: (617) 918-1266
Fax: (617) 918-1291
willey.dick@epa.gov
Dudley Williams
University of Toronto
Surface and Groundwater
Ecology Research Group,
Division of Life Sciences
1265 Military Trail
Scarborough, Ontario, Canada
Phone: (416) 287-7423
Fax: (416) 287-7423
williamsdd @ scar.utoronto.ca
Tom Winter
U.S. Geological Survey
Denver Federal Center
Box 25046
Mail Code: MS 413
Denver, CO 80225-0046
Phone: (303) 236-4987
tcwinter@usgs.gov
Kay Wischkaemper
U.S. EPA, Region 4
61 Forsyth St.
Mail Code: OTS
Atlanta, GA 30303-3415
Phone: (404) 562-8641 - -
Fax: (404) 562-8566
wischkaemper.kav@epa.gov
Carol Witt-Smith
U.S. EPA, Region 5
77 W. Jackson Blvd
Mail Code: DW-8J
Chicago, IL 60604
Phone (312) 886-6146
witt-smith. carol @ epa. gov
William Woessner
University of Montana
Department of Geology
Missoula, MT 59812
Phone: (406) 243-2341
Fax: (406) 243-4028
gl www@selway.umt.edu
161
-------
Proceedings of the Groimd-Water/Suiface- Water Interactions Workshop
July 2000
Appendix B: Discussion Group Focus Issues
A. Hvdrogeologic Data Collection
1. What data are needed to estimate or document temporal changes in ground-water discharges to
various surface water bodies? When and at what frequency should the data be collected?
2. How do the methods of measuring ground-water discharge to surface water depend on
hydrogeologic setting and surface water regime?
3. What are the best methods of measuring ground-water discharges to various surface water bodies?
a. How should measurements be made?
b. Where should measurement be made?
c. Over what area should measurements be made?
d. When should measurements be made?
4. How do we determine the relative proportion of contaminated ground-water flux as a proportion of
the total ground-water flux and/or mass balance for a given area?
B. Chemical Data Collection
1. What are the relevant chemical processes?
2. How should chemical concentrations be measured when determining the flux of contaminated
ground water to a surface water body?
a. Where should the measurements be taken?
b. How should samples be obtained?
c. Over what area should measurements be taken?
d. Over what time period and at what frequency should samples be taken?
3. What are the data quality objectives needed to support an ecological impacts assessment? What are
the proper methods of collecting water and sediment samples to determine ecological impacts?
What is the role of moisture and organic carbon data?
4. How should samples be collected to determine contaminant retention in the biologically active
zone?
5. How should contaminant retention be evaluated in the hyporheic zone and bottom sediments?
162
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
C. Biological Data Collection
1. Is the hyporheic zone considered an ecological habitat to be protected or a "treatment opportunity"
zone for restoration of contaminated ground water discharging into surface water?
a. How can the hyporheic zone be defined biologically? *
b. What are the ecological endpoints in the discharge or hyporheic zones? What ecosystem
functions occur in these zones? *
c. What are the appropriate scales to measure adverse effects to ecological endpoints in riverine,
estuarine, and lacustrine hyporheic systems? *
d. What modifications to existing Guidance or creation of new Guidance are needed to account
for the unique ecological and hydrological aspects (receptors, functions, and routes of
exposure) of the hyporheic zone? *
2. What is the appropriate biological information (data) needed to assess ecological impacts?
a. What should be the structure for evaluating adverse impacts to key ecological endpoints?
(1) What biological monitoring should be performed? Which ecological structures and
functions should be evaluated and why? *
(2) Should it be phased and if so how should priorities be set for the data gathering?
b. How can screening numbers be developed for the hyporheic zone that are protective of
ecological endpoints of concern? Are AWQC and NOAA Sediment Effects Criteria (ER-L and
ER-M data) sufficient as screening numbers for protection of ecological endpoints; or, should
other levels be used or developed for hyporheic zone screening for protection of ecological
endpoints?
3. How should physical biological data be collected?
a. What are the best sampling methods to characterize the biological endpoints and then measure
these for unacceptable impacts? Under what circumstances should filtered or unfiltered water
samples (groundwater and surface water) be taken for environmental purposes? *
b. What sampling locations are appropriate for biological data collection?
c. When should interstitial water samples from sediment (using semi-permeable membrane
devices or other techniques) or whole sediment samples be collected for environmental
purposes? Is there a role for sediment elutriate to be safrnpled?
* Asterisks represent priority issues for the biology discussion group.
163
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
4. Which ecological endpoints are at risk within the ground water/surface water mixing zone for
different surface water regimes?
D. Monitorine Goals and Objectives
1. Identify and characterize zones of interest associated with surface water bodies susceptible to
impact by contaminated groundwater.
2. Determine if discharge of contaminated groundwater is impacting surface water quality or biota in
the zones of interest.
a. Characterize existing impacts.
b. Evaluate the effect of contaminant loading (including seasonal and temporal variations) on
water quality and ecology.
c. Absent current impact on water quality, determine if long-term contaminant loading within a
discharge zone poses a threat to future surface water quality and/or biota.
3. Determine the impact that different hydrogeologic settings and surface water regimes have on the
selection of monitoring methods.
4. Identify prescriptive standards that must be attained.
a. Evaluate the applicability of applicability of a 'mixing zone' to the surface water body
b. Establish regulatory-based (chemical) and/or biota-based compliance standards.
5. Determine the sources of impacts.
164
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Appendix C: Case Study Summaries
Workshop participants submitted 14 case study summaries of ground water/surface water
investigations for inclusion in this report. The purpose of providing these summaries is two-fold. First,
to provide a resource for further information on the various monitoring methods. The case study
summaries represent a range of contaminated media and contaminants within different hydrogeologic
landscapes. Contact names are provided for further information on the use of such monitoring methods
and their utility in obtaining the desired site data. The case studies are also provided as part of an
informal assessment of what techniques are and are not commonly used. The following four tables
provide this assessment.
Table 1 lists the case study sites and the main contaminant types present as well as the type of
monitoring done at each site (physical, geochemical, or biological). Tables 2, 3, and 4 expand on the
type of physical, geochemical, and biological monitoring being done, to summarize in some detail the
type and number of sites using a given monitoring procedure. The tables include a total for the types of
methods used and the number of total different places all the methods have been used at. They show
that physical and geochemical methods are about equally distributed in use, but bioassays (and related
biological monitoring) are much less widely used at the sites.
This appendix is not meant to be a comprehensive list of sites having ground water/surface water
interaction and contamination problems, but simply a tabulation of the types of sites which were
represented through those attending the Workshop, and also a listing of the types of methods which
have been used in the field when dealing with this type of complex ground water/surface water
interaction and contamination sites. It is interesting to note that while the most used monitoring
methods are wells and piezometers,' that there are many other monitoring options that have been used
at sites where there is a ground water to surface water transition zone.
165
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Table 1
GW/SW Case Studies Summary:
Contaminants Studied and Monitoring Methods Used
Site Name
Alcoa/Lavaca'Bay
Angus, Ontario
Everglades National
Park
Hertel Landfill
1-85 Manufacturing
BMI Complex
Ledbetter Cr.
Peconic Estuary
Final Creek
St. Joseph
Union Pacific
Wyckoff Eagle Harbor
Contaminants Monitored
1
V
V
7
V
V
V
V
V
/
/
V
8
/
/
V
?
V
/
V
/
i
'
/
/
•i
i
V
V
V
/
CO
V
V
V
CO
«
°? u
3^
V
V
V
/
Monitoring Methods
•a
o
I
PH
V
•y
V
V
V
V
V
/
V
V
v
V
Geochemical
/
/
7
/
V
/
/
V
/
/
/
1
_0
V
V
/
/
/
/
166
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Table 2
Case Studies Summary of Data Collection Techniques Used
Data Collection Techniques Total Site Names
Current meters
Diffusion sampler
Direct Push Samples
Geophysical Measurements
Geoprobe
GW Water level surveys
GW Mini-piezometer
(Continued)...
Ground water monitoring wells
Ground water multilevel
sampling device
Ground water Waterloo Profiler
Ground water piezometers
In-stream solute tracer
In-stream auto sampler
NAPL studies
Potentiomanometer
SCAPS survey
1
2
1
2
1
3
1
1
3
1
1
Everglades National Park
Pinal Creek
I-85 Manufacturing
I-85 Manufacturing
Wyckoff Eagle Harbor
Peconic Estuary
Wyckoff Eagle Harbor
Wyckoff Eagle Harbor
Angus, Ontario
Wyckoff Eagle Harbor
Angus, Ontario
Ledbetter Cr.
Peconic Estuary
Pinal Creek
Wyckoff Eagle Harbor
Alcoa/Lavaca Bay
Hertel Landfill
BMI Complex
Ledbetter Cr.
Peconic Estuary
Pinal Creek
Union Pacific
Wyckoff Eagle Harbor
Angus, Ontario
St. Joseph
Angus, Ontario
Alcoa/Lavaca Bay
Everglades National Park
Hertel Landfill
Pinal Creek
Pinal Creek
Alcoa/Lavaca Bay
Union Pacific
Wyckoff Eagle Harbor
Angus, Ontario
Wyckoff Eagle Harbor
167
-------
Proceedings of the Gronnd-Water/Suiface- Water Interactions Workshop
July 2000
Table2
Case Studies Summary of Data Collection Techniques Used
Data Collection Techniques Total
Site Names
Sediment sampling
Sediment probe
Seepage meters
Slug testing
Soil cores onshore
(Continued)...
Soil cores offshore
Streambed temperature survey
Surface water monitoring
Time Domain Reflectrometiy
Tracer
Velocity and tracer-dilution
gaging
27
2
5
1
2
Angus, Ontario
I-85 Manufacturing,
Union Pacific
Wyckoff Eagle Harbor
Angus, Ontario
LedbetterCr.
Everglades National Park
Ledbetter Cr.
Peconic Estuary
Pinal Creek
Wyckoff Eagle Harbor
Angus, Ontario
Angus, Ontario
Wyckoff Eagle Harbor
Angus, Ontario
St. Joseph
Wyckoff Eagle Harbor
Angus, Ontario
Wyckoff Eagle Harbor
BMI Complex
Wyckoff Eagle Harbor
Angus, Ontario
Pinal Creek
Pinal Creek
168
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Table 3
Case Studies Summary of Geochemical Techniques Used
Geochemistry
Total
Site Names
Age-dating of GW
Alkalinity
Ammonia
Biochemical Oxygen Demand
(BOD-5)
Cation/Anion
Chemical Oxygen Demand
(COD)
Chloride
Chlorophyll
CO2
Ethene, ethane, methane
Field chemistry tests
(Continued)...
Field Parameters (pH, Temp.,
EH, DO, Elec. Cond.)
Hydrogen Gas—Dissolved
Isotopes
Major ions
NAPL studies
Nitrogen—Dissolved
Nitrogen-Total
1
2
1
1
1
2
1
1
3
1
•9
1
1
2
2
3
Final Creek
Hertel Landfill
Final Creek
Hertel Landfill
Hertel Landfill
Angus, Ontario
Hertel Landfill
Wyckoff Eagle Harbor
Hertel Landfill
Wyckoff Eagle Harbor
Peconic Estuary
Ledbetter Cr.
Everglades National Park
Ledbetter Cr.
Angus, Ontario
St. Joseph
Alcoa/Lavaca Bay
Angus, Ontario
BMI Complex
Ledbetter Cr.
Peconic Estuary
Final Creek
St. Joseph
Union Pacific
Wyckoff Eagle Harbor
St. Joseph
Final Creek
Final Creek
St. Joseph
Wyckoff Eagle Harbor
Union Pacific
Hertel Landfill
Ledbetter Cr.
St. Joseph
Ledbetter Cr.
169
-------
Proceedings of the Ground- Water/Surface- Water Interactions Workshop
July 2000
Geochemistry
Table 3
Case Studies Summary of Geochemical Techniques Used
Total
Site Names
Nutrients
Organic Carbon-Dissolved
Phosphate
Radium isotopes
Radon-222
Redox-sensitive metals
Salinity
Sediment chemistry
Sulfate
(Continued)...
Sulfide
Total Dissolved Solids
Total Suspended Solids
4
1
1
1
1
2
2
1
Everglades National Park
Ledbetter Cr.
Peconic Estuary
Pinal Creek
Angus, Ontario
Hertel Landfill
Hertel Landfill
Ledbetter Cr.
Everglades National Park
Everglades National Park
Everglades National Park
Wyckoff Eagle Harbor
Hertel Landfill
Wyckoff Eagle Harbor
Hertel Landfill
Ledbetter Cr.
Angus, Ontario
Hertel Landfill
St. Joseph
Hertel Landfill
Wyckoff Eagle Harbor
Hertel Landfill
Wyckoff Eagle Harbor
31
57
Table 4
Case Studies Summary of Biological Techniques Used
Biological Data Total Site Names
Bacteriophages 1
Benthic macroinvertebrate 1
Benthic community analysis 2
Biofilm colonization chambers 1
Biomonitoring of plant effluent 1
Everglades National Park
Ledbetter Creek
Wyckoff Eagle Harbor
Ledbetter Creek
Ledbetter Creek
Wyckoff Eagle Harbor
170
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Table 4
Case Studies Summary of Biological Techniques Used
Bioloqical Data
Total
Site Names
Brown tide cell counts 1
Diver surveys 1
Fish pathology 1
Laboratory cultures 1
Laboratory bioassays 1
Sediment chemistry 1
Sediment vertical profiler 1
Trawls " 1
WET testing of plant effluent 1
Peconic Estuary
Wyckoff Eagle Harbor
Wyckoff Eagle Harbor
Wyckoff Eagle Harbor
St. Joseph
Wyckoff Eagle Harbor
Wyckoff Eagle Harbor
Wyckoff Eagle Harbor
Wyckoff Eagle Harbor
Union Pacific
14
15
171
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Case Studies
1) SITE NAME: Alcoa (Point Comfort)/ Lavaca Bay
2) City/State:
Point Comfort, Texas
3) Regulatory Authority:
CERCLA
4) Contact:
Ron Gouguet
Coastal Resource Coordinator
U.S. EPA, Region 6
1445 Ross Avenue
Suite #1200
Dallas, TX 75202-2733
Phone: 214-665-2232
Gouguet.Ron@noaa.gov
Gary Baumgarten
Remedial Project Manager
U.S. EPA, Region 6
1445 Ross Avenue
Suite # 1200
Dallas, TX 75202-2733
Phone: 214-665-6749
Saumgarten.Gary@epa.gov
5) Surface Water Body:
Lavaca and Matagorda Bays
6) Range of Tidal Variation:
0.5-1.5 ft
7) Risk:
Human Health
Fish consumption
Ecological
Fish
Benthos
Shell fish
8)Contaminants:
Ground Water
Hg, PAHs, DNAPL (Hg and tar)
Soil
Creosote compounds, PAHs, Hg
Surface Water
Rarely detected
Pore Water
Hg, MeHg, PAHs
Sediment
PAHs, Hg
3) Monitoring Methods:
Physical Measurements
Geochemical Parameters
Monitoring wells, piezometers, Field parameters, DNAPL
water level surveys, DNAPL studies, salinity
rudies
Bioassays
Unknown at this time
10) COMMENTS:
Contributions of contaminated groundwater appear to be responsible for maintaining Hg and PAH
concentrations in surficial bay sediment above risk based levels of concern. Also, this appears to be the
case for maintaining tissue concentration at levels of concern. The remedy (CERCLA) is expected to
curtail the GW release, remove some sediment and stabilize sources.
172
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
1) SITE NAME: Angus Ontario
2) City/State:
Angus, Ontario, Canada
) Surface Water Body:
Pine River
3) Regulatory Authority:
Ontario Ministry of
Environment and Energy
6) Range of Tidal Variation:
-------
Proceedings oftheGround-Water/Surface-Water Interactions Workshop
July 2000
10) COMMENTS:
Data collected primarily as part of Mr. Conant's PhD research. Pine River typically flows at 1.5 to 2.9
cubic meters per second.
174
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
1) SITE NAME: Everglades National Park/ Florida Bay
2) City/State:
South Florida
5) Surface Water Body:
Wetland, estuary, bay
3) Regulatory Authority:
4) Contact:
Dr. Peter W. Swarzenski
USGS-GD
500 4th Street South
Petersburg, FL 33701
Phone: 727-803-8747 x3072
Dr. Judson W. Harvey
QSGS-WRD (NRP)
12201 Sunrise Valley Drive
MS 430
Reston, VA 20192
Phone: 703-648-5876
6) Range of Tidal Variation:
<10cm
7) Risk:
Human Health
Injection wells?
Ecological
Eutrophication-related issues
8)Contaminants:
Ground Water
Nutrients
Metals?
Soil
Surface Water
Nutrients
Pore Water
Nutrients
Metals?
Sediment
Nutrients
Metals?
9) Monitoring Methods:
Physical Measurements
Current meters, piezometers,
seepage meters
Geochemical Parameter
Radium isotopes, radon-222,
CH4, nutrients, redox-sensitive
metals
Bioassays
Bacteriophages
10) COMMENTS:
A great overview of USGS projects related to South Florida can be found at http://sflwww.er.usgs.gov/
175
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
I) SITE NAME: Exxon Refinery
2) City/State:
Billings, Montana
3) Regulatory Authority:
1) Contact:
HnaDiebold
Region 8-Montana Office
Phone: 406-441-1130 x227
5) Surface Water Body:
Yellowstone River
6) Range of Tidal Variation:
Not applicable
7) Risk:
Human Health
Ecological
Do not know yet
8)Contaminants:
Ground Water
Hydrocarbons
BTEX, SVOC, VOC
Soil
Hydrocarbons
BTEX, SVOC, VOC
Surface Water
Benzene
Pore Water
Benzene
Sediment
Benzene
Monitoring Methods:
Physical Measurements
Wells, laser induced
luorescence, grab samples
Geochemical Parameters Bioassays
None None
10) COMMENTS:
176
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
1) SITE NAME: Hertel Landfill Superfund Site
2) City/State:
Plattekill, New York
3) Regulatory Authority:
3ERCLA
1) Contact:
Dean Maraldo
Hydrogeologist
U.S. EPA, Region 2
ERRD/PSB/TST
290 Broadway
Nlew York, NY 10007-1866
phone: 212-637-3260
maraldo.dean@epa.gov
Sharon Trocher
RPM
U.S. EPA, Region 2
BPA/ERRD/NYRB
290 Broadway
Nfew York, NY 10007-1866
Phone: 212-637-3965
rocher.sharon@epa.gov
5) Surface Water Body:
Wetlands
6) Range of Tidal Variation:
Not applicable
7) Risk:
Human Health
Touching or drinking
contaminated well water or
accidentally ingesting
contaminated soil
Ecological
Pollutants have seeped into on-
site wetlands, posing a threat to
ecologically sensitive
resources, wildlife, or aquatic
biota.
8)Contaminants:
Ground Water
Primarily arsenic, chromium, iron,
manganese
VOCs and CVOCs
Pesticides
Soil
Arsenic, chromium, VOCs
Surface Water
Iron, manganese, pesticides
Pore Water
Sediment
Pesticides, metals
3) Monitoring Methods:
Physical Measurements
Monitoring wells, piezometers
Geochemical Parameter
Bioassays
Surface and ground water: None
phosphate, COD, nitrate-nitrite,
TOC, ammonia, alkalinity,
BOD-5, TKN, sulfide, sulfate, .
chloride, TDS, TSS
10) COMMENTS:
Capping of this 13-acre municipal landfill was completed in the fall of 1998. At this time the primary
COCs are metals in the groundwater and surface. The 1991 ROD remedy included a pump-and-treat
component for groundwater which has been put on hold pending post-cap data evaluation.
177
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
1) SITE NAME: 1-85 Manufacturing and Distribution Center
2) City/State:
Spartanburg, South Carolina
3) Regulatory Authority:
State Superfund
4) Contact:
Fudy Canova
Project Manager
SCDHEC
2600 Bull St.
Columbia, SC 29201
Phone: 803- 896-4046
canovajl®
columb34.dhec.state.sc.us
5) Surface Water Body:
Tributary to Fairforest Creek
6) Range of Tidal Variation:
Not applicable
7) Risk:
Human Health
Contact
Inhalation
Ingestion
Ecological
Fish
Invertebrates
8)Contaminants:
Ground Water
Tetrachloroethylene
Soil
Tetrachloroethylene
Surface Water
Tetrachloroethylene
Pore Water
Unknown
Sediment
Pending
9) Monitoring Methods:
Physical Measurements
Geochemical Parameters
Diffusion samplers, direct push None
samplers, grab samples
Bioassays
None
10) COMMENTS:
The unusual characteristic of this site is the high concentration of tetrachloroethylene observed in
surficial samples from the tributary - up to 10 ppm. It is suspected that NAPL is discharging to the base
of the stream based on groundwatef quality data. At the location of highest contamination within the
stream, there is no visible aquatic life, vertebrate or invertebrate. Contamination persists above ambient
water quality criteria for over half a mile. The length of the discharge coupled with extreme
topographic variation reduces possible remedial options for the stream.
178
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
1) SITE NAME: Kerr-McGee Chemical/ BMI Complex
2) City/State:
Henderson, Nevada
3) Regulatory Authority:
State
1) Contact:
Vlitch Kaplan
Environmental Scientist
U.S. EPA, Region 9
75 Hawthorne Street
San Francisco, CA 94105
Phone: 415- 744-2063
Kaplan.Mitch @ epa.gov
Doug Zimmerman
Chief, Bureau of Corrective
Auction
Nevada Dept. of Environmental
Protection
Phone: 775- 687-4670 x3127
5) Surface Water Body:
Lake Mead, Colorado River
6) Range of Tidal Variation:
Not applicable
7) Risk:
Human Health
Ingestion
Ecological
Unknown (under investigation)
8)Contaminants:
Ground Water
Ammonium perchlorate
Soil
Not analyzed
Surface Water
Ammonium perchlorate
Pore Water
Not analyzed
Sediment
Not analyzed
3) Monitoring Methods:
Physical Measurements
Geochemical Parameters
Monitoring wells, surface water Field parameters
monitoring
Bioassays
None
10) COMMENTS:
179
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
1) SITE NAME: Ledbetter Creek
2) City/State:
Murray, Kentucky
3) Regulatory Authority:
State of Kentucky
4) Contact:
Susan P. Hendricks
rlancock Biological Station
561 Emma Drive
Murray, KY 42071
Phone: 502-474-2272
susan.hendricks @murraystate.e
du
David S. White
Hancock Biological Station
561 Emma Drive
Murray, KY 42071
Phone: 502-474-2272
david.white® murraystate.edu
5) Surface Water Body:
Kentucky Lake Reservoir
6) Range of Tidal Variation:
Hydroelectric/Flood Control
Dam operations result in 2-6 ft
change in water depth at stream
site.
7) Risk:
Human Health
Contact
Ecological
Surface-subsurface microbial
communities
Surface-subsurface macroin-
vertebrate communities
Fish community
Habitat degradation from high
sedimentation/siltation, reduced
surface-subsurface exchange
8)Contaminants:
Ground Water
Nitrates, herbicides, pesticides,
fecal coliforms
Soil
Nitrates, herbicides, pesticides
Surface Water
Nitrates, herbicides, pesticides,
fecal coliforms
Pore Water
Nitrates, herbicides, pesticides,
fecal coliforms
Sediment
Nitrates, herbicides, pesticides,
fecal coliforms
9) Monitoring Methods:
Physical Measurements
Monitoring wells, water table
leights, mini-piezometers,
sediment temperature probes,
seepage meters
Geochemical Parameters
Dissolved oxygen, turbidity,
pH,
ORP, specific conductance,
NO3+NO2,NH4,SRP, Total N,
Total P,SO4,CO2, CH4
Bioassays
Biofilm colonization chambers
for bacterial productivity, activity,
and diversity; benthic and
hyporheic macroinvertebrate
- community structure.
10) COMMENTS:
180
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
1)SITE NAME: Peconic Estuary System
2) City/State:
Suffolk County, New York
3) Regulatory Authority:
National Estuary Program-
Peconic Bay Estuary, Suffolk
County, New York
4) Contact:
Ion Paulsen
Bfydrogeologist
Suffolk County Health
Services-Bureau of Water
Resources
Phone: 516-853-2220
Ronald.paulsen@
co.suffolk.ny.us
!hristopher Smith
Cornell University
^operative Extension Marine
Program Leader
Phone: 516-727-3910
smith@cce.cornell.edu
5) Surface Water Body:
EPA National Estuary Program-
Peconic Estuary System
6) Range of Tidal Variation:
Approximately 2.5-3.5 ft
7) Risk:
Human Health
Estuary is receiving water body
for groundwater discharges that
contains pesticides, VOCs and
elevated nitrates
Ecological
The Peconic Estuary System
has been subjected to the
harmful alga blooms. The HAB
known as brown tide
(Aureococcus onophaefferens)
has plagued the estuary since
1985. Excessive nutrients,
metals, and possibly pesticides
from groundwater seepage are
thought to contribute to the
onset and proliferation of
T-f ARs in thp. Svstfvtn
8)Contaminants:
Ground Water
VOCs, nitrates, pesticides
Soil
Surface Water
VOCs, nitrates, pesticides
.Pore Water
Nitrates, VOCs
Sediment
Monitoring Methods:
Physical Measurements
Installation of monitoring well
and mini-piezometers with
jercussion drill and hollow
augers; geophysical
measurements using logging
techniques including natural
gamma, induction and
resistivity logging; direct
contact resistivity
measurements of bay bottom to
map out groundwater seepage
"aces; groundwater seepage
measurements using time
Geochemical Parameters
Field parameters (conductivity,
temperature, dissolved oxygen,
chlorophyll, pH); nutrient
species including inorganic and"
organic forms of nitrogen;
metals; volatile organic ,
compounds; pesticides
Bioassays
Brown tide (Aureococcus
cmophaefferens) cell counts
181
-------
Proceedings of the Ground-Water/Siuface-Water Interactions Workshop
July 2000
10) COMMENTS:
The Peconic Estuary System is a large estuary system on Long Island, New York that received
National Estuary Status in 1994. Associated with the estuary program are numerous ongoing
investigations and studies. These investigations include studies on the ecological, chemical and
physical properties of the Peconic Bay Estuary. One property being studied is the effect of
groundwater seepage on the chemical and biological conditions in the bay. Direct measurements of
groundwater seepage along with the chemical analysis of coastal groundwater and bay bottom pore
water in the estuary are being made. This information is being used to develop a surface water model
and a groundwater model for the estuary system. The modelling results are being used to developed
guidelines for nutrient loading to the bay especially as they pertain to chlorophyll and dissolved oxygen
levels in the bay.
182
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
) SITE NAME: Pinal Creek Basin, Arizona
2) City/State:
alobe, Arizona
5) Surface Water Body:
Final Creek, Salt River,
Roosevelt Lake (reservoir for
Phoenix)
3) Regulatory Authority:
State- Arizona Dept. of
environmental Quality
WQARF)
Federal-CERCLA
1) Contact:
udson Harvey
LJSGS
130 National Center
Reston, VA 20192
Phone: 703-648-5876
wharvey@usgs.gov
Martha Conklin
Dept. of Hydrology
University of Arizona
Harshbarger Bldg
P.O. Box 210011
Tucson, AZ, 85721
Phone: 520-621-5829
martha@hwr.arizona.edu
Christopher C. Fuller
USGS
345 Middlefield Road, MS465
Menlo Park, CA 94025
Phone: 650-329-4479
ccfuller@usgs.gov
James Brown
USGS
520 N. Park Avenue
Tucson, AZ 85719
Phone: 520-670-6671x280
jgbrown@usgs.gov
6) Range of Tidal Variation:
Not applicable
8) Contaminants:
Ground Water
Dissolved iron, aluminum, copper,
manganese, cobalt, nickel, zinc
pH<4 in some portions of ground
water contamination plume
Soil
7) Risk:
Human Health
Probably minimal. The major
concern is for the small number
of families living in the northern
part of the basin that withdraw
their water from wells emplaced
in the aquifer. For the most part
the affected wells were moved
away from contaminated areas
years ago. There continues to be
concern about downstream
effects of metal pollution in the
basin on water quality in the
Salt River and Roosevelt Lake,
although studies to date suggest
that metals are not reaching the
Lake in appreciable quantities.
Remedial actions are being
undertaken to intercept the
groundwater plume.
Ecological
Largely unstudied at this
location and therefore unknown.-
However, the perennial is
within the Tonto National
Forest with abundant wildlife.
Poor in-stream water quality
and manganese oxide deposits
on the stream bed doubtless are
affecting aquatic and terrestrial
organisms that use the stream
and riparian zone.
Surface Water
Manganese, nickel, cobalt, zinc,
aluminum
pH generally > 6 in surface water.
Pore Water
Sediment
183
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
9) Monitoring Methods:
Physical Measurements
Velocity and tracer-dilution
gaging of stream discharge; in-
stream solute-tracer
iXperiments to determine
surface and hyporheic-zone
water exchange; in-stream
auto-samplers; USGS mini
drivepoint sampler; seepage
meters; stainless-steel
drivepoints; conventional
wells; identification of ground
water source areas using water
stable isotopes; age-dating of
ground water using CFCs.
Geochemical Parameters Bioassays
pH, DO, temperature, alkalinity,
major ions, dissolved metals,
particulate and colloidal metals,
dissolved organic carbon,
nutrients
10) COMMENTS:
USGS and the University of Arizona have identified natural attenuation processes that remove metal
contaminants due to interactions between surface water and ground water. Hydrologic exchange
between the stream that receives the contaminated ground water and the hyporheic zone beneath the
stream delays the downstream movement of contaminants, and also exposes the contaminants to
unique microbial processes that enhance removal of contaminants in the hyporheic zone. USGS and
the University of Arizona have published more than fifteen journal papers and reports on this topic.
Interested readers are encouraged to contact the lead scientists listed above for reprints and more
information.
184
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
371) SITE NAME: St. Joseph, Michigan
>) City/State:
Stevensville, Michigan
J) Regulatory Authority:
"ERCLA/State
Contact:
bhn M. Lendvay
Research Fellow
Jniversity of Michigan
217 EWRE Building
(351 Beal Avenue
Ann Arbor, MI 48109-2125
Phone: 734-764-6350
endvay @ engin.umich.edu
'eter Adriaens
Associate Professor
Jniversity of Michigan
181 EWRE Building
[351 Beal Avenue
Ann Arbor, MI 48109-2125
Phone: 734-763-1464
adriaens @engin.umich.ed
5) Surface Water Body:
Lake Michigan
6) Range of Tidal Variation:
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
1) SITE NAME: Union Pacific Railroad Laramie Tie Plant Site
2) City/State:
Laramie, Wyoming
3) Regulatory Authority:
RCRA, CERCLA, State
4) Contact:
Marisa Latady
Wyoming Department of
Environmental Quality/Soil.
Hazardous Waste
122 West 25th Street
Cheyenne, WY 82002
Phone: 307-777-7752
mlatad @ missc.state. wy.us
Felix Flechas
US EPA
Region VTJDE
999 18th Street
Denver, CO 80202
Phone: 303-312-6014
"elix.flechas @ epa.gov
5) Surface Water Body:
None- The Contaminant
Isolation System prevents
releases to the Laramie River
6) Range of Tidal Variation:
Not applicable
7) Risk:
Human Health
Dermal contact
Incidental ingestion
Inhalation of particulates
Ecological
Direct exposures via soil
ingestion
Direct exposures via dermal
contact with soil
Indirect exposures via ingestion
of contaminated food items
Inhalation of particulate dust
(considered less significant the
others described above)
8)Contaminants:
Ground Water
Residuum oil, PAHs,
pentachlorophenol (PCP),
benzene, ethylbenzene, toluene,
xylene, DNAPL
Soil
PAHs, PCP, dioxin, furans
Surface Water
Not applicable
Pore Water
Residuum oil, PAHs,
pentachlorophenol
benzene, ethylbenzene, toluene,
xylene
Sediment
Not applicable
9) Monitoring Methods:
Physical Measurements
Geochemical Parameters
Monitoring wells; piezometers; Field parameters
sediment sampling; monitoring
of the containment systems for
hydraulic control; DNAPL
thickness
Bioassays
. WET testing of the water treatment
plant effluent under an NPDES
permit
186
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
10) COMMENTS:
UPRR operated the Laramie Tie Plant Site for the treatment of railroad ties and other wood preserving
operations on an intermittent basis from 1886 to 1983.The site borders the Laramie River just south of
the city of Laramie, Wyoming. Waste management practices, such as allowing treated ties to drip dry
onto the ground and discharging wastewater generated in the treating process to an unlined surface
impoundment, are believed to be the causes of contaminated soils and ground water at the site. The
primary contaminants identified at the site include creosote, pentachlorophenol and other residuum
oils.
Contamination at the site was discovered in 1981, and in 1983 the Environmental Protection Agency
(EPA) and UPRR signed a CERCLA Administrative Order on Consent (AOC) to continue the
remedial investigation already in progress and to conduct site cleanup. The investigation identified
contamination in surface soils and ground water contamination, including the presence of oil in the
subsurface [i.e., Dense Non-Aqueous Phase Liquid (DNAPL)].Some of the early activities conducted
by UPRR to address the contamination identified include:
l.In 1983, UPRR decommissioned the facility, including demolition of on-site buildings and
shipment of unused wood treatment materials to another facility.
2.1n 1984, UPRR partially closed the unlined Surface Impoundment that received wastewater. The
Surface Impoundment is a regulated unit as defined by the WDEQ/HWRR Chapter 10, Section
6(a).
3.In 1987, UPRR installed the Contaminant Isolation System (CIS) to prevent migration of
contaminants to the Laramie River. The CIS consists ofrelocation of the Laramie River to an
uncontaminated channel; construction of a cutoff wall; installation of a water management system
consisting of horizontal drain lines along the exterior and interior of the cutoff wall to maintain an
inward hydraulic gradient; construction of a water treatment plant to remove dissolved
contaminants and implementation of a monitoring program to ensure the effectiveness of the CIS.
4.1n 1988, UPRR installed ground water extraction wells, referred to as the Morrison Contaminant
Withdrawal System (MCWS), outside the western site boundary to address a small area of
contaminated ground water in Morrison bedrock.
In 1991, EPA and UPRR entered into an AOC under RCRA that required UPRR to conduct a
Corrective Measure Study (CMS) to identify long-term remedies for implementation at the site,
including pilot tests of various techniques to remove DNAPL from the subsurface.
In 1994, EPA selected the remedy to address contamination at the site. The remedy included continued
operation of the CIS and MCWS systems, removal of DNAPL using the waterflood oil recovery
method, covering a portion of the site with topsoil to address contaminated surface soils, installing a
RCRA cap over the former Surface Impoundment area, and maintaining restricted access to the site.
Nine criteria were selected to evaluate the performance of the final remedy. Detailed descriptions of
these criteria can be found in EPA's September, 1994, "Final Decision and Response to Comments.
In 1995, the RCRA AOC was amended to require UPRR to submit an application for a RCRA Permit
for post-closure care and corrective action by September 1,1995. UPRR submitted an application for a
post-closure care and corrective action permit on September 1, 1995, and revised that application in
May 1996, August 1997 and March 1998.The amendment to the AOC also required UPRR to
implement the final remedy selected by EPA in 1994. The final remedy was amended in 1995 to
include the use of a Corrective Action Management Unit (CAMU) to consolidate contaminated
187
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
fitly 2000
concrete debris and soils in the partially closed unlined Surface Impoundment. The CAMU currently
has an interim soil cover of six inches.
That portion of the final remedy that requires closure (i.e., installation of a RCRA cap) and post-
closure care of the Surface Impoundment, as described in Section A of this Fact Sheet, is deferred to
allow implementation and evaluation of phytoremediation, an innovative technology, designed for in-
situ remediation of waste, contaminated soils and contaminated ground water. Phytoremediation test
plots will be established over a portion of the Surface Impoundment and the western portion of the
facility to determine the effectiveness of this technology. Review of this corrective action program will
be conducted every five (5) years as part of the technical impracticability (TI) determination. The TI
determination is made when ground water restoration to applicable cleanup standards is unattainable
from an engineering perspective. If WDEQ determines, based on the five (5) year review process, that
phytoremediation does not meet the remediation criteria specified in the Permit, UPRR will be
required to implement the closure and post-closure care requirements established in the Permit. Those
portions of the final remedy that are not deferred include continued oil recovery operations in the
Surface Impoundment area until all recovery units have achieved the endpoint criteria, and
implementation of the ground water corrective action program.
As of December 1998 UPRR has recovered approximately 1,500,000 gallons of oil from the
subsurface through the waterflood oil recovery method.
188
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
) SITE NAME: West Branch Canal Creek, Aberdeen Proving Ground
) City/State:
Edgewood, Maryland
5) Surface Water Body:
Wetland and stream
) Regulatory Authority:
CERCLA
4) Contact:
Vlichelle Lorah
U.S. Geological Survey
5987 Yellow Brick Road
Baltimore, MD 21237
'hone: 410-238-4301
'ax: 410-238-4210
mmlorah@usgs.gov
6) Range of Tidal Variation:
About 2 ft change in stage in
creek; affects ground-water
flow direction and plume
distribution
7) Risk:
Human Health
Air transport of VOCs
Ecological
Air transport of VOCs
Possible exposure of benthic
organisms to VOCs in water
and sediment
8)Contaminants:
Ground Water
Chlorinated VOCs
Possible DNAPL
Soil
None
Surface Water
Infrequently detected, low
concentrations of chlorinated
VOCs
Pore Water
Chlorinated VOCs
Sediment
Chlorinated VOCs in wetland
sediment
monitoring Methods:
'hysical Measurements
Geochemical Parameters
Nested piezometers, diffusion VOCs; ethane; ethene;
samplers, cores, field chemistry dissolved organic carbon; total
tests, salinity, pressure
transducers and tide gage
organic carbon redox species-
methane, sulfide, Fe(H)/Fe(HI),
manganese, dissolved oxygen,
nitrate, ammonia; field
parameters (pH, alkalinity,
temperature, conductance,
salinity, turbidity); major
cations and anions; selected
trace metals
Bioassays
Microcosms to measure
biodegradation rates and daughter
products; DNA/RNA analysis of
microbial communities in wetland
sediment
10) COMMENTS:
USGS WRIR 97-4171: Report on project results through 1997 available online:
http://md.usgs.gov/publications/online.hrml
189
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
SITS NAME: Wyckoff Eagle Harbor
2) City/State:
Bainbridge Island, Washington
3) Regulatory Authority:
CERCLA
4) Contact:
Ren6 Fuentes
Hydrogeologist
U.S. EPA, Region 10
1200 Sixth Avenue
Seattle, WA 98101
Phone: 206-553-1599
fuentes.rene@epa.gov
Hahn Gold
Remedial Project Manager
U.S. EPA, Region 10
1200 Sixth Avenue
Seattle, WA 98101
Phone: 206-553-0171
gold.hahn@epa.gov
5) Surface Water Body:
Eagle Harbor Puget Sound
6) Range of Tidal Variation:
14ft
7) Risk:
Human Health
Contact
Inhalation
Fish consumption
Ecological
Fish
Shell fish
DNAPL contact
8)Contaminants:
Ground Water
Creosote compounds, PAHs,
pentachlorophenol, fuel oil,
LNAPL, DNAPL
Soil
Creosote compounds, PAHs,
pentachlorophenol
Surface Water
Rarely detected
Pore Water
Creosote compounds, PAHs,
Sediment
Creosote compounds, PAHs,
DNAPL
9) Monitoring Methods:
Physical Measurements
Monitoring wells, mini-
piezometers, sediment probe
^temperature and electrical
conductivity), seepage meter,
off-shore cores, diver surveys
jNAPL), water level surveys,
LNAPL and DNAPL studies
Geochemical Parameters
Bioassays
Field parameters, LNAPL and Biomonitoring of treatment plant
DNAPL studies, salinity effluent
10) COMMENTS:
190
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
Appendix D: MS-IE Push Point Sampling Tools
by Mark A. Henry*
A new tool and sampling methodology have been devised for collecting pore water samples from
beneath beaches and surface water bodies. The use of this technology enables a single investigator or
small team to rapidly gather pore water samples at or near the interface between groundwater and
receiving bodies of water. From a research perspective, the information gained in analyzing these
samples may be very helpful in understanding the geochemical nature of this transition zone and the
biological processes at work.
This methodology has been used very successfully to locate the expression of contaminated
groundwater venting into several lakes in Michigan. The technique involves the use of an MHE 27-
inch push-point sampling device (PP27), V4-inch outer diameter by Vfc-inch inner diameter Tygon
tubing, and 50 m£, 100% polyethylene syringes or a peristaltic pump. The PP27 is a rigid Vs-inch
diameter stainless steel probe that is screened at one end and ported at the other to allow the collection
of pore water with a syringe or peristaltic pump. In this method's simplest form, the investigator would
walk along a beach or in shallow water paralleling the beach, and at periodic intervals push (by hand) a
decontaminated PP27 into the sand or sediments with a twisting motion until refusal (usually 6-18-
inches). Then the screened zone is exposed and pore water samples are withdrawn at "low-flow
sampling" collection rates using a disposable syringe connected by a length of Tygon tubing. Usually,
only 30-50 mC of water withdrawal is necessary to develop this miniature well; this equates to
approximately 20-35 volume exchanges through the PP27 . Subsequently drawn water is usually non-
turbid and suitable for dispensing directly into sample containers or instruments. A 3-dimensional
sampling array is possible within the sediments and the water column. The PP27 is easily
decontaminated in the field but if the investigator has several of the inexpensive sampling devices on-
hand, sample collection along a transect can be very rapid. When 100% polyethylene syringes are
employed, samples may be collected and stored temporarily within the syringe by placing the full,
sealed syringe in a cooler. Once the sample collection has been completed, the investigator can process
the samples in a controlled environment. As an added benefit, it is possible to use the sample-filled
syringes for on-site headspace analysis of VOC's using a field GC—information that be used to direct
an investigation in real time. If the syringe is half-dispensed and refilled with air, resealed, and
agitated, the headspace in the syringe above a known volume of water can be quickly analyzed.
The Michigan Department of Environmental Quality (MDEQ) uses an enhanced variation of this
method. As samples are being collected, some of the pore water is immediately dispensed into field
analytical equipment for measurement of "stabilization parameters" such as dissolved oxygen, pH,
conductivity, redox, and temperature, or analytes such as dissolved iron, sulfide, etc. The MDEQ
investigators were able identify and map the expression of a groundwater plume venting into Lake
Michigan and several inland lakes using this methodology and/or these techniques and SCUBA gear.
Furthermore, the MDEQ couples its sampling with location information obtained using sub-meter
accuracy global positioning system (GPS) equipment. Plotting the geochemical data onto an accurate
GPS representation of the sampling locations and predominant local features produces a precise plume
expression map. GPS technology allows investigators to reliably relocate previous sampling locations
for additional study and accurately combine and compare data from multiple sampling events.
* Editor's Note: Mark Henry presented this material at a meeting of the Ground Water Forum in April 2000.
191
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
What follows is found in the MHE PP-27 sampler instruction manual. It is presented as additional
information about the sampler and to provide a few practical hints.
MHE PP27" Push-Point Sampling device (Patent Pending)
Operators Manual, Ver. 1.02, May 13, 2000
INTRODUCTION
The groundwater/surface water interface has been a research interest of mine for the past decade.
This transitional zone is usually rich in biomass and may play a predominant role in the bioattenuation
of contaminated groundwater entering surface water bodies. Usually these biologic processes have
limited effectiveness in attenuating highly contaminated groundwater, leaving a plume of parent
contamination and metabolic byproducts that eventually expresses itself in receiving waters—usually
classified as non-point sources of pollution because of the uncertainty of the discharge area. Part of the
problem in the detection and study of these plumes is that there were no devices on the market for the
rapid, discrete collection of pore water samples. Reliance on conventional technology and techniques
to perform a detailed investigation required extensive effort and burdensome equipment.
Through several iterations, I have evolved a simple device for collecting pore water samples from
beneath surface water bodies or the beach areas surrounding them. Pore water sampling using the PP27
becomes a simple and efficient process, generating a wealth of information and very little waste. The
effective working depth is up to 26 inches below the land or sediment surface. If one collects
groundwater samples in a transect perpendicular to groundwater flow in the suspected area of plume
discharge to an open water body, their analysis yields information about the areal extent of
contaminant discharge to the water body. At this point, additional sampling can complement the initial
data and provide the information necessary to map the plume expression in both magnitude and areal
distribution. This is becoming increasingly important to regulators as they decide the ecological
impacts of discharging contaminant plumes.
Sampling at each location usually takes five minutes, allowing a small crew to collect dozens of
samples in an afternoon. These samples can be analyzed in the field for real-time information useful in
directing field investigations and research. The work that I have conducted at several contamination
sites indicates that many groundwater plumes discharge in surface water bodies in 2-3 feet of water
depth—accessible to investigators wearing hip boots or waders. Many plumes, especially light non-
aqueous phase liquid (LNAPL) plumes can be delineated by collection of samples in very shallow
water or from under beaches. My initial experience has shown that dense non-aqueous phase liquid
(DNAPL) contaminant plumes express themselves in the shallbw,"near-shore water as well, even
though the onshore depth of the contaminant mass was deep in the aquifer.
DIRECTIONS
As shown in Figure 1, the PP27 device is a very simple, precisely machined tool consisting of a
tubular body fashioned with a screened zone at one end and a sampling port at the other. The bore of
the PP27 body is fitted with a guard rod that gives structural support to the PP27 and prevents plugging
and deformation of the screened zone during insertion into sediments. The PP27 is made of 316
stainless steel assuring compatibility with most sampling environments. The screened zone consists of
a series of interlaced machined slots which form a short screened zone with approximately 20% open
area.
192
-------
Proceedings of the Ground-"Water/Surface- Water Interactions Workshop
July 2000
Operation of the PP27 is not difficult. Simply hold the device in a manner that squeezes the two
handles towards each other to maintain the guard rod fully inserted in the PP27 body during the
insertion process (as shown in Figure 2). Holding the device in this manner, push the PP27 into the
sediments or beach to the desired depth using a gentle twisting motion. When the desired depth or
refusal is reached remove the guard rod from the PP27 body without disturbing the position of the
deployed sampler. Once the guard rod has been removed from the PP27, it SHOULD NOT be
reinserted into the device until the bore of the PP27 has been thoroughly cleansed of all sand, silt, etc.
Attach a syringe or (peristaltic) pump to the PP27 sample port (see Figure 3) and withdraw water at
a low-flow sampling rate (50-200 ml/min.). Once non-turbid aliquots have been withdrawn,
representative samples can be collected for on-site and off-site analysis.
HELPFUL HINTS, INFORMATION, AND CAUTIONS
• Multiple depths can be sampled in one hole if samples are collected from deepest to shallowest.
Insert the sampler using a twisting motion until you reach refusal. Remove the guard rod. Do not
push the sampler further into the sediments once the guard rod has been removed as this
may damage the screened zone and plug the PP27 with sediment. Once sampling has been
completed at this deepest depth, the PP27 can be partially pulled from the hole to a new sampling
elevation. Remember not to insert the PP27 into the sediments without the guard rod inserted to
prevent screened zone damage. Alternately, multiple holes can be used to collect samples from
multiple depths at a particular sampling location. It is recommended that some type be device be
used to prevent lateral movement and slippage of the PP27 as sampling is conducted near the top
of the hole (see Figure 3). This offsets the leverage of the instrument and reduces hole
degeneration. MHE offers an 8-inch diameter., heavy-duty steel sampling platform engineered for
precise sampling depth requirements of field research. A plate of steel with a 3/16-inch diameter.
hole through its center would serve the fundamental purpose of maintaining a rigid hole opening .
If repeated shallow sampling is to be conducted, it may be more convenient to use a shorter
sampler (MHE - PP15").
• If you wish to reuse the PP27 sampler at a particular sampling location and want to clean the bore
quickly while you're there so that the guard rod may be safely reinserted, you can use a syringe
filled with surface water or deionized water to backflush the bore several times before reinserting
the guard rod. Use at least 100 mfi of water. If you have too much trouble reinserting the guard rod
(e.g., due to grit), it will be necessary to use the standard cleaning procedures with cleaning rod
and soap solution.
• If the screened zone of the PP27 becomes plugged while inserted in the sediments, it is frequently
possible to hydraulically/pneumatically shock the screened zone free of adhering material while it
is inserted into the sediments. Attach a large-volume syringe to the sampling port. In a quick
motion, pull the syringe plunger most of the way back (creating a vacuum) and then immediately
release the plunger—the plunger will slam to a neutral position, sending a shock wave through the
bore of the PP27 and may alleviate the problem.
• The PP27 can be used as a piezometer to determine the static head of the groundwater and hence,
the potential direction of groundwater movement. To do this, a tube is connected to the sample port
as shown in Figure 5. A continuous stream of water is established from the syringe (or pump) to
the screened zone by pumping out any air remaining in the PP27and tubing. When the tube is
193
-------
Proceedings of the Ground-Water/Suiface-Water Interactions Workshop
July 2000
disconnected from the syringe, the static water level in the tube will represent the static water level
at the depth that the screened zone occupies.
It is frequently possible to push the PP27 through thin lenses of low-permeability material and
collect samples from below them and gather valuable geochemical samples. At many of the sites
where the PP27 has been used, sampling from just below a layer of fine sand, silt, or clay, one
occasionally encounters seemingly large pockets of gas that seem to have coalesced and collected
under this less permeable stratum. Analysis of these pockets may provide additional insight to
predominant biological processes. It may true that the concentration of volatile organic compounds
(VOCs) in the groundwater has equilibrated with these bubbles (i.e., steady state), which means
that their presence in a sampling stream or syringe would not significantly affect the concentration
of dissolved VOCs. In fact, if one assumes that steady-state conditions exist, the concentration of
VOCs in the bubbles is directly related the concentration in the surrounding groundwater. An
alternate condition may exist if the groundwater is supersaturated with bacterial metabolic waste
gases and the negative pressure exerted by the pump (or syringe) is initiating a degassing of
dissolved gases from the groundwater. In this instance, VOCs would partition from the
groundwater to the bubbles as they are formed in the sampling tubing (this is fairly evident if
occurring). The consequence is that part of the dissolved contaminant mass has partitioned into the
gas phase; unless the gas-phase is captured, quantified, and accounted for, the native VOC
concentration of the groundwater is not reflected by analysis of the groundwater alone. If this
condition exists, the degassing effect can be minimized by decreasing the sampling rate to a rate
more easily yielded by the sampled formation. With experience, it is easy to distinguish which of
these conditions (or combination of conditions) exist and to what extent they affect sample quality.
The internal volume of a PP27 is approximately 1.5 mi A 50 m« syringe full of distilled water,
decontamination water, methanol, etc. will push about 33 volumes through the bore.
When straightening the screened zone it is sometimes helpful to wash out the bore of the device
and then insert the guard rod or the cleaning rod to the area of the bend in the screened zone.
Gently unbend the portion of the screened zone nearest the rod and carefully advance the rod to the
next bend. After the rod has been fully inserted into the screened zone perform the final screened
zone straightening fine-tuning until the guard rod slides freely through it.
If the sampling port of the PP27 is above the static level of the water body, each time you remove
the syringe or pump from the PP27 sampling port, air will fill the bore of the PP27 allowing the
water level in the bore to reach its static head. To avoid this plug of air from entering the
subsequent syringe, attach a clamp adapter and or a three-way^valve between the sampling port and
the syringe or pump inlet as shown in Figure 7.
I have conducted dye tests by injecting concentrated uradine dye under a perforated 1.5-foot
diameter disk through which the PP27 was inserted 3-12-inches into sediments. The goal of these
tests was to determine whether or not surface water and dye is drawn into samples collected in near
surface sediments (i.e., whether a cone of depression is formed). The results indicated that no
surface water is drawn into samples even though sampling was conducted with a peristaltic pump
at a rate of 600 m5/min.
I usually couple my field investigations with GPS location of the sampling point. If conditions
permit, a pin flag can be placed at the sampling point for later location by GPS. I usually use sub-
194
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
meter grade GPS for this surveying; GPS can then be used to relocate previously sampled point
even if certain site physical characteristics have changed (eroding shorelines, etc.).
Sampling by syringe has many advantages. This is my preferred field method due to its simplicity
and versatility. It is useful to be able to collect several 50 m{ syringes full of groundwater, store
them on ice, and perform the sample transfer to a VOA vial, etc. under more controlled conditions.
To transfer sample to a VOA vial, place the end of the transfer tube (Figure 8) to the bottom of the
VOA vial. Dispense sample into the VOA vial and slowly withdraw the transfer tube from the vial
maintaining the mouth of the transfer tube just below the sample surface. When the transfer tube is
almost out of the vial, continue to dispense sample and leave an "anti-meniscus" of sample above
the rim of the vial. Add several drops of HC1 (which will displace a few drops of sample) and cap.
If VOC samples are to be collected and stored temporarily in a syringe, I recommend using 100%
polyethylene ("two piece") syringes such as those made by Henke Sass Wolf GMBH (NormJect®,
50 m{)) configured as shown in Figure 8. From personal experience I have found that small
amounts of aromatic compounds (BTEX) can leach from the rubber parts of the rubber-tipped
plunger found in common medical syringes. Rubber-tipped plunger syringes have less side-wall
resistance and work much smoother than the 100% polyethylene syringes so I use medical syringes
for "development" of the PP27. Standard medical syringes also work well for collecting samples
for non-VOC analysis. I utilize handheld meters for pH, conductivity, redox, dissolved oxygen, etc.
One can dispense sample from the syringe into these types of instruments for field measurements.
The 50 mi 100% polyethylene syringes mentioned above can be purchased directly from MHE,
configured with tubing as was the example syringe included with your order, or customized to suit
your individual needs. If you would to make your own, the syringes that I am currently using are
purchased from National Scientific Company. The tubing is Tygon 14-inch outer diameter and VB-
inch inner diameter. Be sure to use some type of clamp at the tubing mouth to ensure a good seal at
the sampler port.
Headspace GC analysis of VOCs can be easily accomplished using 100% polyethylene syringes.
Dispense all but 20 m<> of the sampled groundwater from the syringe. Refill the syringe to the 40
m
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
• The PP27 may be used to inject nutrients or dyes into the sediments for field trials of biologic or
geochemical testing or tracing groundwater paths. Simply insert the PP27 to the desired depth and
after the guard rod has been removed, connect a syringe or pump and slowly inject the desired
fluid into the sediments, perhaps followed by a small amount of native groundwater to flush the
instrument.
• These devices can be dedicated as semi-permanent underwater monitoring devices. If a PP-27 is
inserted to the desired depth through a plate (such as the sampling platform mentioned earlier) that
can lock the sampler at the correct insertion depth, a vinyl cap can be placed over the mouth of the
sampler, and the sampler can be dedicated to that location so that future samples can be withdrawn
when desired.
• It has been useful to carry several samplers in "quivers" made of 2-inch PVC tubing: one tube for
10-15 clean and assembled samplers, and one tube for used samplers and their separated guard
rods. This arrangement protects both the investigators and the instruments.
I hope that users will find many useful and innovative uses for this device. If you have other helpful
information, uses, and advice concerning these samplers, please write or e-mail suggestions to me for
inclusion in future manual revisions. I will be forming a website soon, and posting much of my GSI
research with links to as much GSI field research and related topics as I can find.
Thanks.
Mark Henry
MHE Products
3371 Sherman Rd.
East Tawas, MI 48730
Phone: 517-362-5179 or 517-393-0948
e-mail: markhen@engin.umich.edu
196
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
F igure 1
. body of aenvler
Fig, 1a
disassembled sampler
maintoln this __ f
Jdtcnco during——*n
InsertIon J^
Fig. 1b
assembIed samp Ier
F i gure 2
cjrasp Instrument firmly
and squeeze two handles
together to maintain
this distance while
Inserting into sediments
guard-rod handle
• PP 27 handle
197
-------
Proceedings of the Ground-Water/Surface- Water Interactions Workshop
July 2000
FIgure 4
attachment of decon adaptor
27
liquids In
end of wand
of pressurized
garden sprayer
with nozzle
removed
•• Iiqu i ds out
decon adaptor
Tygon or
vinyl tubing"
water level
if positive
hydrostatic head
water body surface
water Ieve I
if negative
hydrostatic head
sediment surface
pore fluids
enter here
or ore
Injected here
pore fluids
enter here
or are
injected here
F i gure 5
198
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
D'ace clorrp here
or cut tubing and
install 2 or- 3-woy valve
Tygon o^ vinyl tubing.
sediment surface •
•tubing coupler
small vlre ties or
very sroolI hose clomps
Figure 6
0! spans Ing somple
into VOA viol
sample tn VOA viol
owotiing preservation
and copping
r>gcn er vinyl tubing
—40 ml VOA vial
'ontt-ralnttctn* of
IWOt* AOv* i-l*
of VOA vtal
FIgure 7
filling VOA vial "headspace free"
199
-------
Proceedings of the Ground-Water/Surface-Water Interactions Workshop
July 2000
GC syringe with
needI« Inserted
through tubing Into''"'
cyring* to collect sonplo
of neodsooca
olr
Tygon or vinyl tubing
-plug In «nd of tubing
-Clomp
50 ml syringo
naif fiiiea
with sonpltt
200
------- |