A
United States
Environmental Protection
Agency
Hydrogeologic Framework,
Ground-Water
Geochemistry, and
Assessment of Nitrogen
Yield from Base Flow in Two
Agricultural Watersheds,
Kent County, Maryland
-------
-------
EPA/600/R-02/008
January 2002
Hydrogeologic Framework, Ground-Water
Geochemistry, and Assessment of Nitrogen
Yield from Base Flow in Two Agricultural
Watersheds, Kent County, Maryland
by
L. Joseph Bachman
U.S. Geological Survey
Baltimore, MD 21237
David E. Krantz
U.S. Geological Survey
Dover, DE 19901
Johnkarl Bohlke
U.S. Geological Survey
U.S. Department of the Interior
Reston, VA20192
IAG #DW14937941
Project Officer
Mohamed M. Hantush
Subsurface Protection and Remediation Division
National Risk Management Research Laboratory
Ada, OK 74820
National Risk Management Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
/T"V Recycled/Recyclable
Printed with vegetable-based ink on
paper that contains a minimum of
60% posVeonswnei «bej content
processed chlorine free.
-------
Notice
The information in this document has been funded by the U.S. Environmental Protection Agency-(EPA)
under the auspices of the Ecosystem Restoration Research Program of the Office of Research and
Development's National Risk Management Research Laboratory through an Interagency Agreement with the
Department of Interior's U.S. Geological Survey under Agreement No. DW14937941. It has been subjected
to the Agency's peer and administrative review, and has been approved for publication as an EPA document.
Mention of trade names or commercial products does not constitute endorsement or recommendation for use.
All research projects making conclusions or recommendations based on environmental data and funded by
the Environmental Protection Agency are required to participate in the Agency Quality Assurance Program.
This project was conducted under an approved Quality Assurance Project Plan. The procedures specified in
this plan were used without exception. Information on the plan and documentation of the quality assurance
activities and results are available from the Principal Investigator. ;
-------
Foreword
The U.S. Environmental Protection Agency is charged by Congress with protecting the Nation's land, air,
and water resources. Under a mandate of national environmental laws, the Agency strives to formulate and
implement actions leading to a compatible balance between human activities and the ability of natural
systems to support and nurture life. To meet this mandate, EPA's research program is providing data and
technical support for solving environmental problems today and building a science knowledge base
necessary to manage our ecological resources wisely, understand how pollutants affect our health, and
prevent or reduce environmental risks in the future.
The National Risk Management Research Laboratory (NRMRL) is the Agency's center for investigation of
technological and management approaches for preventing and reducing risks from pollution that threatens
human health and the environment. The focus of the Laboratory's research program is on methods and their
cost-effectiveness for prevention and control of pollution to air, land, water, and subsurface resources;
protection of water quality in public water systems; remediation of contaminated sites, sediments and ground
water; prevention and control of indoor air pollution; and restoration of ecosystems. NRMRL collaborates
with both public and private sector partners to foster technologies that reduce the cost of compliance and to
anticipate emerging problems. NRMRL's research provides solutions to environmental problems by:
developing and promoting technologies that protect and improve the environment; advancing scientific and
engineering information to support regulatory and policy decisions; and providing the technical support and
information transfer to ensure implementation of environmental regulations and strategies at the national,
state, and community levels.
Nutrient inputs from human activities have resulted in the "cultural eutrophication" of lakes and coastal
waters. This publication documents a study of the role of the shallow subsurface stratigraphy in determining
the potential for nutrient impacts from nitrogen species on water quality in streams as a result of ground-
water discharges to the streams. The study focused on two adjacent watersheds in Kent County, Maryland,
that had similar topography, land use and soils. Although the watersheds were similar in landscape features
and land use, it was found that the hydrostratigraphy of the aquifer underlying the watershed had a dominant
role in producing the observed differences in nitrogen yields from the two watersheds. The study illustrates
the importance of understanding hydrogeology for management of nitrogen yields from watersheds to
coastal waters.
This publication has been produced as part of the Laboratory's strategic long-term research plan. It is
published and made available by EPA's Office of Research and Development to assist the user community
and to link researchers with their clients.
Stephen G. Schmelling, Acting
Subsurface Protection and Remedialion Division
National Risk Management Research Laboratory
-------
-------
Abstract
Hydrostratigraphic and geochemical data collected in two adjacent watersheds on the Delmarva Peninsula, in Kent County,
Maryland, indicate that shallow subsurface stratigraphy is an important factor that affects the concentrations of nitrogen in
ground water discharging as stream base flow. The flux of nitrogen from shallow aquifers can contribute substantially to the
eutrophication of streams and estuaries, degrading water quality and aquatic habitats. The information presented in this report
includes a hydrostratigraphic framework for the Locust Grove study area, analyses and interpretation of ground-water
chemistry, and an analysis of nutrient yields from stream base flow. An understanding of the processes by which ground-water
nitrogen discharges to streams is important for optimal management of nutrients in watersheds in which ground-water
discharge is an appreciable percentage of total streamflow. The U.S. Geological Survey, in cooperation with the U.S.
Environmental Protection Agency (USEPA), collected and analyzed hydrostratigraphic and geochemical data in support of
ground-water flow modeling by the USEPA.
The adjacent watersheds of Morgan Creek and Chesterville Branch have similar topography and land use; however, reported
nitrogen concentrations are generally 6 to 10 milligrams per liter in Chesterville Branch but only 2 to 4 milligrams per liter in
Morgan Creek. Ground water in the surficial aquifer in the recharge areas of both streams has high concentrations of nitrate
(greater than 10 milligrams per liter as N) and dissolved oxygen. One component of the ground water discharging to Morgan
Creek typically is anoxic and contains virtually no dissolved nitrate; most of the ground water discharging to Chesterville
Branch is oxygenated and contains moderately high concentrations of nitrate.
The surficial aquifer in the study area is composed of the deeply weathered sands and gravels of the Pensauken Formation (the
Columbia aquifer) and the underlying glauconitic sands of the upper Aquia Formation (the Aquia aquifer). The lower 6 to 9
meters of the Aquia Formation is a low-permeability silt-clay with abundant glauconite. The Aquia confining layer underlies
the Columbia-Aquia surficial aquifer throughout the study area. The sediment redox transition, identified in cores, that occurs
in the upper 0.5 to 1 meter of the Aquia confining layer is thought to be a site for subsurface denitrification of ground water.
The first confined aquifer is composed of the glauconitic sands in the upper 9 to 11 meters of the Hornerstown Formation. The
Hornerstown aquifer is underlain by 10 to 15 meters of glauconitic silt-clay at the base of the Hornerstown Formation (the
Hornerstown confining layer), and 5 meters of low-permeability clay in the underlying Severn Formation.
The Aquia and Hornerstown Formations dip and thicken to the southeast, and the Aquia confining layer subcrops shallowly
(within 5 meters of the land surface) in a band that strikes southwest to northeast across the northern edge of the study area. The
surficial aquifer is very thin (generally less than 5 meters) north of Morgan Creek, and the alluvial valley of Morgan Creek has
incised into the top of the Aquia confining layer. In contrast, the Aquia confining layer lies 22 meters below Chesterville
Branch, and the surficial aquifer approaches 30 meters in thickness (away from the creek).
Chemically reduced iron sulfides and glauconite in the Aquia confining layer are likely substrates for denitrification of nitrate
in ground water. Evidence from the dissolved concentrations of nitrate, sulfate, iron, argon, and nitrogen gas, and stable
nitrogen isotopes support the interpretation that ground water flowing near the top of the Aquia confining layer, or through the
confined Hornerstown aquifer, has undergone denitrification. This process appears to have the greatest effect on ground-water
chemistry north of Morgan Creek, where the surficial aquifer is thin and a greater percentage of the ground water contacts the
Aquia confining layer.
The base-flow discharges of total nitrogen from the two watersheds are of similar magnitude, although Chesterville Branch has
somewhat higher loads (29,000 kilograms of nitrogen per year) than Morgan Creek (20,000 kilograms of nitrogen per year),
although Morgan Creek has a larger drainage area and a greater discharge of water. The base-flow yield of nitrogen (load per
unit area) in Chesterville Branch (median of 0.058 grams per second per square kilometer at the outlet) is more than twice that
of Morgan Creek (median of 0.022 grams per second per square kilometer at the outlet), reflecting the higher concentration of
nitrate in ground water discharging to Chesterville Branch. Total nitrogen concentrations tend to decrease downstream in
Chesterville Branch and increase downstream in Morgan Creek. The downstream trend in Chesterville Branch may be affected
by instream nitrogen uptake and denitrification, and an increasing proportion of older, denitrified ground water in downstream
discharge. The downstream trends in Morgan Creek may be affected by inflow from tributaries, downstream changes in the
source of discharge water, and downstream changes in the riparian zone, which could affect the processes and degree of
denitrification.
-------
Although these two watersheds appear to have landscape features (such as topography, land use, and soils) that would produce
similar nitrogen discharges, a more detailed examination of landscape features indicates that Chesterville Branch has soils that
are slightly better drained, tributary stream outlets at higher altitudes, and a slightly higher percentage of agricultural land. All
of these factors have been related to higher nitrogen yields. Nonetheless, most of the data support the interpretation that
hydrostratigraphy has the greatest effect in producing the difference in nitrogen yields between the two watersheds.
vi
-------
Contents
Foreword m
Abstract , v
Figures v"'
Tables x
SI Conversion Factors x*
Acknowledgments > xn
Introduction 1
Description of the Study Area 2
Previous Investigations 2
Data-Collection Methods 9
Compilation of Existing Data 9
Corehole Drilling 9
Well Network and Water-Level Measurement 1°
Well Sampling n
Base-Flow Sampling H
Chemical Analysis of Water Samples 13
Data Analysis .' 13
Hydrogeologic Framework 1*
Lithostratigraphic Units 16
Holocene Alluvial Sediments 16
Pensauken Formation, Upper Miocene (?) / Lower Pliocene (?) 17
Old Church (?) Formation, Upper Oligocene, and Calvert Formation,
Lower to Middle Miocene 17
Aquia Formation, Upper Paleocene 17
Hornerstown Formation, Lower Paleocene 18
Composite Hydrostratigraphic Sequence 18
Surficial Aquifer (Columbia and Aquia Aquifers) 18
Aquia Confining Layer 18
Hornerstown Aquifer and Hornerstown Confining Layer 19
Geometry of the Hydrostratigraphic Units 19
Ground-Water Flow Paths and Traveltimes , 23
Ground-Water Geochemistry 27
Dissolved Nitrate and Denitrification - 28
Nutrient Yields from Stream Base Flow 31
Sampling at Fixed Sites - 31
Base-Flow Synoptic Surveys 32
Relation Between Nutrients and Landscape Factors 43
Summary and Conclusions 49
References Cited : • —• 51
Appendices
Appendix A • • • ^5
Appendix B 1 -• 58
Appendix C 60
Appendix D 64
Appendix E •- 67
Appendix F 7/7
vii
-------
Figures
la. Index map showing the location of Chesapeake Bay, Delmarva Peninsula,
and the Locust Grove study area in Kent County, Maryland 3
Ib. Index map showing the location of the Locust Grove study area in Kent County, Maryland „. 4
2. Map of the Locust Grove study area, Kent County, Maryland, showing the location of stream-gaging
stations, surface-water sampling sites, wells, coreholes, and transects of the hydrostratigraphic sections 5
3. Map showing land use in 1994 for the Morgan Creek and Chesterville Branch watersheds,
Kent County, Maryland ; g
4. Diagram showing the relative distribution of the age of discharging ground water in stream
headwaters and downstream , 14
5. Composite hydrostratigraphic sequence of the Locust Grove study area, Kent County, Maryland 15
6. Hydrostratigraphic section A-A' along Route 213 at the northern boundary of the Locust Grove
study area, Kent County, Maryland 20
7. Hydrostratigraphic section B-B' along Blacks Station Road and VanSants Corner Road running
diagonally through the middle of the Locust Grove study area, Kent County, Maryland 20
8. Hydrostratigraphic section C-C" along Route 298, the western boundary of the Locust Grove
study area, Kent County, Maryland .' " 21
9. Hydrostratigraphic section D-D', the primary transect along Locust Grove Road from Locust Grove
to Chesterville, in the Locust Grove study area, Kent County, Maryland ;. 21
10. Hydrostratigraphic section E-E', the eastern boundary of the Locust Grove study area through Angelica
Nurseries, Kent County, Maryland \ 22
11. Map showing the extent of the subcropping Aquia confining layer and redox zone in the Locust Grove
study area, Kent County, Maryland, with subbasins for surface-water sampling sites ; 23
12. Map showing the configuration of the water table and generalized ground-water flow, Locust Grove !
study area, Kent County, Maryland, May 1998 24
13a Hydrogeologic section D-D' (along Locust Grove Road) showing ground-water flow paths in the . j
shallow aquifer system, May 1998, Locust Grove study area, Kent County, Maryland „ 25
13b. Hydrogeologic section C-C' (along Route 298) showing ground-water flow paths in the shallow
aquifer system, May 1998, Locust Grove study area, Kent County, Maryland , 26
14. Piper diagram showing chemistry of water from ground-water samples collected in July 1998 at the
Locust Grove study area, Kent County, Maryland 27
VIII
-------
15. Plot showing the distribution of dissolved argon and nitrogen gas with relation to the solubility in air
and enrichment of nitrogen by denitrification, Locust Grove study area, Kent County, Maryland 28
16. Time-series plots of total dissolved nitrogen concentrations at the two fixed stream-sampling sites
(station numbers 01493112 and 01493500) in the Locust Grove study area, Kent County, Maryland 31
17. Map showing the areal distribution of total dissolved nitrogen concentrations in stream base flow for
the Locust Grove study area, Kent County, Maryland, April 1998 synoptic survey 33
18. Plots of observed downstream trends of stream discharge, specific discharge, total dissolved nitrogen
concentration, instantaneous nitrogen load, and instantaneous nitrogen yield from base-flow synoptic
surveys in April and September 1998, Locust Grove study area, Kent County, Maryland 34
19. Diagram showing expected downstream trends of nitrogen and related geochemical constituents from
a downstream increase in the discharge of old ground water 35
20. Plots of observed downstream trends of sulfate, magnesium, bicarbonate, and silica concentrations
from base-flow synoptic surveys in April and September 1998, Locust Grove study area, Kent County,
Maryland 36
21. Diagram showing relations among discharge and concentrations of dissolved constituents in tributaries
and downstream sites under non-reactive mixing 36
22. Map showing location of surface-water sampling stations and stream reaches on Morgan Creek and
Chesterville Branch in the Locust Grove study area, Kent County, Maryland 37
23. Scatterplot matrix showing relations among stream-outlet altitude, percentage of type C (moderately
poorly drained) soils, and percentage of agriculture in the Morgan Creek and Chesterville Branch basins,
Locust Grove study area, Kent County, Maryland 46
24. Diagram showing relations among basin index, dissolved nitrogen concentration, and location
within the stream network, Locust Grove study area, Kent County, Maryland 47
25. Correlation between regional and local basin index scores for sites in the Locust Grove study area,
Kent County, Maryland 47
ix
-------
Tables
i.
Physical Characteristics of the Two Watersheds in the Locust Grove Study Area,
Kent County, Maryland
2. Summary Information for the Coreholes Drilled in the Locust Grove Study Area, Kent County, Maryland 10
3. Chlorofluorocarbon (CFC) Recharge Dates and Ground-water Ages from Wells in the Locust Grove
Study Area, Kent County, Maryland, sampled in 1998 12
4. Summary of Base-flow Nitrogen Loads from the Locust Grove Study Area, Kent County, Maryland 32
5A-D. Estimated Concentrations of Dissolved Constituents in Ground-water Discharge Calculated from Measured
Streamflow and Concentrations of Constituents at Upstream, Downstream, and Tributary Sampling Sites,
Locust Grove Study Area, Kent County, Maryland 39
6. Correlation Coefficients for Landscape Variables Used in the Analysis of the Morgan Creek and '
Chesterville Branch Watersheds, Locust Grove Study Area, Kent County, Maryland 45
7. Results of Principal-components Analysis of Landscape Variables for the Morgan Creek and
Chesterville Branch Watersheds, Locust Grove Study Area, Kent County, Maryland 46
-------
SI Conversion Factors
Multiply
To obtain
inch (in)
mile (mi)
square mile (mi2)
cubic feet (ft3)
pounds (Ib)
2.54
1.609
2.590
28.32
0.4536
centimeter (cm)
kilometer (km)
square kilometer (km2)
liters
kilograms
Temperature given in degrees Celsius (°C) can be converted to degrees Fahrenheit (°F) by use of the following equation:
°F=1.8(°C) + 32
Measurements in the interpretive sections of this report are given in metric units (modified SI system) to conform with U.S.
Government policy and generally accepted scientific practice. However, many of the field measurements were made using
U.S. Customary ("foot-pound") units, and are so reported in the appendices. The conversion tables above will aid the reader
in converting from one system of measurement to the other.
In this report, "sea level" refers to the National Geodetic Vertical Datum of 1929 - a geodetic datum derived from a general
adjustment of the first-order level nets of the United States and Canada, formerly called the Sea Level Datum of 1929.
Chemical concentration is given in milligrams per liter (mg/L), which expresses the dissolved concentration of a constituent as
weight (milligrams) of solute per unit volume (liter) of water. For concentrations less than 7,000 mg/L, the numerical value is
the same as for concentrations in parts per million. Concentrations of nitrogen compounds, such as nitrate (NO^, are given
in units of milligrams per liter as nitrogen (N).
Stable isotopic composition of nitrogen compounds, such as nitrate, are reported in standard delta notation (815N), which relates
the ratio of 15N to I4N in the sample to the ratio in a reference standard; values are reported in units of per mil (%o, or parts per
thousand).
XI
-------
Acknowledgments
Mohamed Hantush of the U.S. EPA National Risk Management Research Laboratory (NRMRL) in Ada,
Oklahoma, was the U.S. EPA Project Officer and principal investigator for the overall project. Jerome Cruz
of ManTech Environmental Research Services Corporation in Ada, Oklahoma, provided database and
modeling support. Thanks are offered to David Hudson, Deborah Bringman, Anthony Tallman, Joseph
Beman, and Lisa Donohoe of the USGS in Dover, Delaware, and Elizabeth Marchand, James Dine, and Amy
Derosier of the USGS in Baltimore, Maryland, who collected and organized the field data. Sarah Kelley
(USGS in Baltimore) provided GIS support, and Tim Auer (USGS in Baltimore) produced the final graphics
for the report. Roger Starsoneck of the USGS in Baltimore, Maryland, located drilling sites, assisted with
drilling operations, supervised the surveying activities to determine well altitudes, and created and main-
tained data records for the wells. Coreholes were drilled by drillers from the Geologic Division of the USGS,
and monitoring wells were installed by drillers from the New Hampshire District of the USGS Water
Resources Division. Niel Plummer and Eurybiades Busenberg of the USGS National Research Program in
Reston, Virginia, analyzed chlorofluorocarbon age-dating tracers, dissolved nitrogen and argon, and major
ions for some supplemental samples. Chemical analyses of surface-water and ground-water samples were
performed in the USGS National Water Quality Laboratory (NWQL) in Denver, Colorado.
Thanks are offered to the landowners who gave permission to enter then- property to access a number of the
stream-sampling sites, and to the Maryland State Highway Administration and Kent County Government for
permission to drill coreholes and install monitoring wells on highway rights-of-way. The management of
Angelica Nurseries was especially cooperative in allowing access to their land, providing information about
their production wells, and granting permission to install wells on their property and to collect water levels,
samples, and geophysical logs from their monitoring wells. ;
XII
-------
Introduction
Eutrophication of lakes and coastal waters has long been recognized as a major environmental problem in many parts of the world.
Increased inputs of mineral nutrients to lakes and coastal waters has led to increased biomass and organic carbon, which degrade the
aquatic environment. Although some eutrophication is due to natural causes, in the context of environmental management this term
usually represents the effects of "cultural eutrophication," caused by inputs of nutrients from human activities (Richardson and
J0rgensen, 1996).
Eutrophication may be controlled by the availability of mineral nutrients, such as nitrogen, phosphorus, or silica, or it may be
affected by the availability of organic matter. For many years, research on the problem was focused on the input of nutrients from
surface-water discharge, especially nutrients associated with suspended sediment This focus on runoff is reasonable for water
bodies where primary production is controlled by the availability of phosphorus ("phosphorus limited"), because phosphorus is
insoluble in oxidized environments, adsorbs strongly to sediment particles, and is transported with suspended sediment in surface-
water discharge. In many other ecosystems, however, primary productivity generally is nitrogen limited (Fisher and Butt, 1994;
Borum, 1996). The discharge of nitrogen compounds dissolved in ground water is likely to play a major role in the nutrient
dynamics of coastal waters in temperate regions because of the following factors: (1) nitrogen, in the form of nitrate, is much more
soluble than phosphorus in oxygenated environments; (2) nitrogen concentrations elevated above background levels have been
documented in shallow aquifers in many coastal areas (Bachman, 1984a; 1984b; Hamilton and others, 1993; Mueller and others,
1995; and Mueller and Helsel, 1996); and (3) ground-water discharge comprises an appreciable percentage of total streamflow in
many temperate areas. These factors have implications for environmental management, because the beneficial effects of nutrient
control may not be apparent immediately after the control practices have been implemented. Ground-water traveltimes are much
longer than surface-water traveltimes, on the order of years or decades rather than days or weeks, respectively. Consequently, the
effects of reducing the recharge of nutrients to an aquifer may be delayed for many years after the implementation of nutrient-
control practices (Focazio and others, 1998).
The management of nutrient inputs to control eutrophication is relatively advanced in the Chesapeake Bay watershed. The
Chesapeake Bay is the Nation's largest and most productive estuary, but the presence of toxic dinoflagellates and anoxic zones in
its deeper waters stress the estuarine ecosystem. Both of these stresses have been linked to nutrient enrichment (Fisher and Butt,
1994; Lewitus and others, 1995). Stream base-flow surveys indicate that ground-water nitrogen accounts for approximately 40 to
50 percent of the total nitrogen discharge to the bay from nontidal rivers and streams (Bachman and Phillips, 1996; Bachman and
others, 1998). Other investigations have provided estimates of ground-water flow rates from shallow aquifers (Simmons and
others, 1990; McFarland, 1995), descriptions of the relation between ground-water traveltimes and nitrogen concentrations (Dunkle
and others, 1993; Bohlke and Denver, 1995; and Speiran, 1996), and the relation between landscape features and water quality
(Bachman and Phillips, 1996; Phillips and Bachman, 1996).
Many aspects of the relation between the details of ground-water flow and geochemical processes that influence nitrogen discharge
from shallow aquifers have not been resolved. The results presented in this report further document the processes associated with
the flux of nitrogen to Chesapeake Bay by ground-water discharge. In particular, it has been suggested that geochemical reactions,
such as denitrification, may have the capacity to substantially reduce nitrogen loads in a watershed (Bohlke and Denver, 1995).
Denitrification is a process by which bacteria obtain energy by the chemical reduction of nitrate and the oxidation of organic matter
or other reduced compounds in the absence of oxygen. If carried to completion, denitrification transforms dissolved nitrate to
nitrogen gas. When denitrification occurs in a ground-water system, the nitrogen gas remains dissolved in the ground water until
it can escape to the atmosphere when the ground water discharges to a surface-water body.
The objectives of this project were to extend the results of the previous investigation by Bohlke and Denver (1995) that documented
the occurrence of denitrification in deep parts of the surficial aquifer, to relate the occurrence of denitrification to shallow
subsurface geologic formations, and to assess the effect of the discharge of denitrified water on the chemical composition of stream
base flow. This project focused on the relation between the hydrogeologic framework and nitrogen discharge from a shallow
Coastal Plain aquifer system.
This report contains a summary of the results of activities conducted by the U.S. Geological Survey (USGS) from 1997 through
2000 in support of a collaborative research effort between the USGS and the National Risk Management Research Laboratory
-------
(NRMRL) of the U.S. Environmental Protection Agency (USEPA). The research project, "Geohydrologic Foundations for
Ecosystem Restoration," is an examination of the movement and fate of nutrients, particularly nitrogen, in the subsurface at the
Locust Grove research site in the Chester River watershed on the Delmarva Peninsula (Figures la and Ib). Field data-collection
efforts by the USGS were based on an extensive network of stream-gaging stations, wells, surface-water-sampling sites, and
coreholes (Figure 2).
USEPA activities, as described by Hantush and Cruz (1999), included geological modeling, development of analytical and
numerical models of ground-water flow and nitrate transport, assessment of the watershed capacity for nitrogen reduction, and
management of a site database, which has been shared among project collaborators and is available for other investigators. The
major objective of the USGS activities were: (1) to describe the geometry of the shallow subsurface sediments and provide
stratigraphic interpretations for hydrogeologic models; (2) to describe the configuration of the water table and ground-water flow
paths in the surflcial aquifer and the uppermost confined aquifer; (3) to provide evidence of hydrogeochemical processes that affect
nitrogen discharge to streams from ground water; and (4) to provide information needed to estimate the volume of ground-water
discharge and ground-water nitrogen loads.
Description of the Study Area
The Locust Grove study area is located in central Kent County, Maryland, (Figure 1), between the tidal Chester and Sassafras
Rivers, about 30 km south of Elkton, Maryland, and about 15 km northeast of Chestertown, Maryland. The study area consists of
the Morgan Creek and Chesterville Branch watersheds (Figure 2) above their respective stream-gaging stations. Although
topographic maps show no impoundments in these two streams, field technicians observed a small pond in the main channel of
Chesterville Branch downstream of site 01493110, and other field technicians have reported beaver dams and other obstructions to
streamflow on sections of both Chesterville Branch and Morgan Creek.
The gaging station at Morgan Creek near Kennedyville (site 01493500) is part of the USGS basic-data network for Maryland, and
discharge has been measured continuously at this site since 1951. The gaging station at Chesterville Branch near Crumpton
(site 01493112) was established in 1996, and is maintained as part of an ongoing USGS project to estimate river input of nutrients
to Chesapeake Bay. The combined area of both drainage basins is about 47 km2. Morgan Creek is a larger watershed (31 km2) than
Chesterville Branch (16 km2) (Figure 3; Table 1). ;
The soils in the two watersheds are generally well drained, with over half the land area in both watersheds classified as soil
hydrologic classes of "well-drained" (class A) and "moderately well-drained" (class B) (Maryland Department of State Planning,
1973). However, the Morgan Creek watershed does have a somewhat higher percentage of areas with "moderately poorly drained"
(class C) and "poorly drained" (class D) soils (Table 1). !
Because of the fertile, well-drained soils, agriculture is the dominant land use in the study area, accounting for about 90 percent of
the watershed area (Figure 3; Table 1); Most of the cropland is in a rotation of corn, soybeans, and small grains. A significant part
of the Chesterville Branch watershed is under cultivation for nursery stock. Fields near the headwaters of Morgan Creek and along
some tributaries west and northwest of Morgan Creek near Kennedyville, Maryland (Figure 3), are used as pasture for dairy
operations. A few confined animal-feeding operations and poultry houses are scattered throughout the study area, but there are
fewer in this area than farther south on the Delmarva Peninsula. The primary area classified as urban is the village of Kennedyville,
located along the northwest edge of the Morgan Creek watershed (Figure 3). Some other areas shown as urban by the remote-
sensing land-cover classification system of Bara (1994) are commercial establishments or farm buildings. Woodlands and wetlands
are located almost exclusively in the riparian areas along Morgan Creek and Chesterville Branch.
The hydrologic setting of the study area has the potential to discharge large volumes of ground water to the streams. The study area
is located in the Atlantic Coastal Plain Physiographic Province, and is underlain by unconsolidated clastic sediments that can store
and transmit large volumes of ground water. The climate is humid temperate, with an excess of precipitation over evapotranspira-
tion. The average rainfall is about 110 centimeters (42 inches) per year, and the average temperature is 13°C or 55°F ([Yokes and
Edwards, 1974). Although rainfall is distributed relatively evenly throughout the year, evapotranspiration is substantially lower in
the winter because of the lower temperatures and dormant vegetation. !
Previous Investigations • -
The geologic framework of Kent County, Maryland,-was first mapped by Clark (1915). The stratigraphic nomenclature underwent
various revisions over the years, culminating in work by Owens and Denny (1979), Owens and Minard (1979), and Hansen (1992).
Regional descriptions of the configuration of aquifers and confining layers in central Kent County are provided by Overbeck and
Slaughter (1958), Gushing and others (1973), Bachman (1984b), and Drummond (1998). Local-scale work at the Locust Grove site
began in 1988 as part of the Delmarva pilot project of the USGS National Water-Quality Assessment (NAWQA) program
(Shedlock and others, 1999). A general description of the hydrogeologic and water-quality conditions in the arei was presented by
Hamilton and others (1993). Dunkle and others (1993) and Reilly and others (1994) applied age-dating tracers to describe ground-
water flow in parts of the Chesterville Branch watershed. Bohlke and Denver (1995) used the estimated recharge dates and a variety
of geochemical data to describe the history of nitrogen recharge to the surficial aquifer, and discuss possible mechanisms for the
transport and fate of nitrogen in the Chesterville Branch and Morgan Creek watersheds.
-------
EXPLANATION
DELMARVA PENINSULA
LOCUST GROVE STUDY AREA
Location of Locust Grove study area
Figure la. Location of Chesapeake Bay, Delmarva Peninsula, and the Locust Grove study area hi Kent County, Maryland.
-------
7d>05'00*
3 KILOMETERS
Figure Ib. Location of the Locust Grove study area in Kent County, Maryland.
-------
7^55'UU"
\
Nau Farm well
IARMONY CORNER
Well Be 198 «
Multipart 3496
sampler
2 KILOMETERS
EXPLANATION
3500
A SURFACE-WATER SAMPLING STATION AND PARTIAL
IDENTIFICATION NUMBER [The complete identification
number listed in the National Water Information System
(NWIS) data base and in the Appendixes contains the
prefix '0149'.]
51® WELL AND PARTIAL IDENTIFICATION NUMBER, USED
FOR WATER-LEVEL MEASUREMENTS OR SAMPLED
FOR CHEMICAL ANALYSIS (Installed during previous
, studies, 1987 to 1996, or drilled and developed by
private owners prior to 1997.)
[The complete well number listed in the NWIS data base
and in the Appendixes contains the prefix 'KE Be,' with
the following exceptions:
• Wells of the' form 9x-yyyy contain the prefix 'KE' and
are not in the NWIS data base;
• Wells of the form BLK-x-yy have no prefix and are
not in the NWIS data base.]
188
183,
B-
'® WELL AND PARTIAL IDENTIFICATION NUMBER, USED FOR
WATER-LEVEL MEASUREMENTS OR SAMPLED FOR
CHEMICAL ANALYSIS (Drilled and installed during this study,
1997-98. The complete well number contains the prefix 'KE Be.')
COREHOLE (DRILLED IN 1997) OR PREVIOUSLY DRILLED
WELL WITH CORES OR OTHER GEOPHYSICAL DATA,
AND NAME OR PARTIAL IDENTIFICATION NUMBER
(The complete well number contains the prefix 'KE Be,'
except for sites labeled "Nau Farm" and "Nursery View,"
in which case the name shown is the complete well identifier.)
• B1 LINE OF HYDROSTRATIGRAPHIC SECTION
Figure 2. The Locust Grove study area, Kent County, Maryland, showing the locations of stream-gaging stations, surface-
water sampling sites, wells, coreholes, and transects of the hydrostratigraphic sections.
-------
S&O'OO'
2 KILOMETERS
EXPLANATION
SURFACE-WATER DRAINAGE DIVIDE (BASIN BOUNDARY)
3500 A SURFACE-WATER SAMPLING STATION AND PARTIAL
IDENTIFICATION NUMBER (Refer to Figure 2.)
LAND USE
H URBAN
AGRICULTURAL
1~* NURSERY/ORCHARD
ANIMAL FEEDING OPERATIONS
FARM BUILDINGS
FOREST
WETLANDS
WATER
Figure 3. Land use in 1994 for the Morgan Creek and Chesterville Branch watersheds, Kent County, Maryland.
-------
Table 1. Physical Characteristics of the Two Watersheds in the Locust Grove Study Area, Kent County, Maryland
Chesterville Morgan
Branch Creek
Hydrologic
soil class,
percentage of
watershed
Land use,
percentage of
watershed
Drainage area, in km2
Stream length, in km
Class A, well-drained
Class B, moderately well-drained
Class C, moderately poorly drained
Class D, poorly drained
Agricultural
Forested
15.9
5.8
32
49
8
1
93
6
31.1
8.6
15
59
15
4
88
8
[Values are for the watershed areas above gaging station 01493112 on Chesterville Branch and station 01493500 on Morgan Creek; locations
of gaging stations shown in Figure 2. Compiled from data of Maryland Department of State Planning (1973). More complete data for these
watersheds and selected subwatersheds are presented in Appendix F. km, kilometers; knf, square kilometers.]
-------
-------
Data-Collection Methods
Much of the geochemical interpretation presented in this report is based oh numerous chemical analyses of water samples from the
Delmarva NAWQA and related projects; these data have been reported previously by Hamilton and others (1993) and Bohlke and
Denver (1995). New and refined hydrostratigraphic interpretations are based on drillers* and gamma logs from 23 wells originally
described in previous studies, 10 private monitoring wells, and 33 new wells and 4 new coreholes drilled for this project. A
corehole is a drilled hole from which a sediment core has been recovered; a borehole is a more generic term for a drilled hole that
may include a well (a drilled hole with casing) or a rotary-drilled hole that was used only to collect cuttings and/or a geophysical
log. Monthly water-level measurements were made on all of the 63 wells in the network during the project. Additional ground-
water samples were collected from 11 of the newly drilled wells for chemical analyses. Seventy-nine base-flow samples were
collected from 14 stream-sampling sites.
Compilation of Existing Data
Five categories of data were compiled from published and unpublished sources, and from data collected as part of this project.
These categories include: (1) geologic or geophysical logs from drill holes in the area, which provide information on the
hydrogeologic framework of the area; (2) measurements of ground-water levels and stream discharge, which provide information
on flow directions and flow quantities; (3) chemical analyses of water samples from wells, which provide information on the
geochemical processes controlling ground-water quality; (4) chemical analyses of stream base flow, which provide information
about surface- and ground-water geochemistry and nutrient yields from base flow; and (5) digital geographic data used for relating
base-flow nitrogen loads to landscape features. AH existing data compiled for this project have been previously published (except
for some of the digital geographic data), although some original field records of published data also wpre examined. The data
include material originally reported by Overbeck and Slaughter (1958), Hansen (1992), Hamilton and others (1993), Tompkins and
others (1994), and Bohlke and Denver (1995).
The digital geographic data were compiled, stored, and analyzed using the ARC/INFO and ArcView geographic information
systems (GIS). Basin boundaries for all surface-water sampling sites were hand-digitized from paper copies of the USGS Betterton
and Galena 7 Vi-minute topographic quadrangle maps, produced at a scale of 1:24,000. Land use was derived from the Multi-
Resolution Land Characteristics (MRLC) data set, which is based on land cover at 30-m resolution compiled in the early 1990s
(Bara, 1994). Soil characteristics were compiled from U.S. Department of Agriculture sources by Dr. Margaret Mayers-Norton,
formerly of the University of Maryland Horn Point Environmental Laboratory.
Corehole Drilling
Four coreholes drilled at the beginning of the project in October and November 1997, provide ground truth fpr interpreting
geophysical logs from other boreholes in the area, and details of the lithology relevant to ground-water-flow modeling and ground-
water geochemistry. The four coreholes varied in depth from 25.9 to 39.6 m below land surface (Table 2), and all penetrated the
confining layer at the base of the surficial aquifer. Coreholes KE Be 185 (Locust Grove) and KE Be 186 (Chesterville Branch
North) align with the existing corehole KE Bf 180 (Hansen, 1992) along a transect that runs north-northwest to south-southeast, and
is very close to the regional dip of the subsurface formations (Figure 2). Coreholes KE Be 183 (Morgan Creek North) and
KE Be 184 (Morgan Creek South) are located on the upland approximately 0.3 km north and south, respectively, of Morgan Creek.
These two coreholes provide control along strike and away from the primary hydrostratigraphic transect along Maryland Road 444,
and detailed stratigraphy for the wells near Morgan Creek valley (Figure 2).
The four coreholes were drilled with a truck-mounted drill rig using mud-rotary methods and a 9-cm (centimeter)-diameter core
barrel with a hollow cutting head.' A wire-line retrieval system was used to recover a 3-m (10-foot) inner core barrel that protects
a nominal 5-cm (2-inch)-diameter core. The core was removed from the inner core barrel in 0.6-m (2-foot) lengths, washed with
a fine spray of water to remove the drilling mud from the exterior, and placed in a partitioned storage box. The on-site geologist
measured, labeled, photographed, and described each section of the core while it was fresh. The hand-written notes from the
drilling operation and core descriptions, and the photographs of the cores are on file in the USGS Water Resources Division
Maryland-Delaware-D.C. District Office, in Baltimore, Maryland.
After each corehole was drilled to the maximum depth and the last section of core retrieved, the hole was filled with drilling mud
and the outer core barrel was removed. Natural-gamma and spontaneous-potential and electrical-resistivity logs were run in the
mudded hole. These logs are archived in the Maryland-Delaware-D.C. District geophysical-log repository in Baltimore, Maryland.
-------
Table 2. Summary Information for the Coreholes Drilled in the Locust Grove Study Area, Kent County, Maryland
Well ID CoreholeName Latitude Longitude Altitude (m) Total Depth (m)
KE Be 183 Morgan Creek North
KE Be 1 84 Morgan Creek South
KE Be 1 85 Locust Grove
KE Be 186 Chesterville Branch North
[Corehole sites are shown in Picture 2. m. meters:
39° 18' 30"
39° 18' 14"
39° 19' 40"
39° 18' 09"
", degrees; ',
75° 58' 11"
75° 57' 54"
75° 57' 01"
75° 55' 53"
minutes: ", seconds. J
21
20.3
23.2
22.7
25.9
27.9
28.9 '
39.6 ,
Well Network and Water-Level Measurement
A network of 66 wells was used for the project. This network includes 23 existing wells from previous USGS NAWQA projects,
33 new wells installed in April 1998, and 10 private monitoring wells on the property of Angelica Nurseries (Figure 2). Information
on the characteristics of the existing wells is summarized in Appendix A, and information on the characteristics of the newly drilled
wells is summarized in Appendix B. During the course of the project, water levels in all of these wells were measured monthly,
from May 1998 through October 1999. A subset of the new wells was sampled for chemical analyses as described below.
The new wells installed for this project supplemented the existing wells in order to:
(1) Improve the spatial distribution of wells to better represent the water table and ground-water flow system across the
entire study area.
(2) Sample more points immediately below the water table (within 3 m of the seasonal water-table low) to document the
spatial distribution of nitrate-input concentrations and ground-water recharge rates.
(3) Sample ground water from the confined aquifer (Homerstown Formation), which was not sampled previously, to
characterize its geochemistry.
(4) Sample discrete horizons near the base of the surficial aquifer above, at, and below the transition zone between
reduced and oxidized sediments that has been postulated as a source of electron donors for denitrification of ground
water (Bohlke and Denver, 1995).
The new wells were installed using hollow-stem auger methods according to the recommended procedures for the USGS NAWQA
Program (Koterba and others, 1996). The only deviation from these procedures was that the casing was not steam-cleaned prior to
installation, because the wells were not used to sample pesticides or volatile organic compounds. Only freshwater was used to flush
the hole. No bentonite drilling mud was circulated during drilling, which can fill pore spaces in the aquifer materials adjacent to the
well.
Wells were placed singly or in clusters of two or more to allow sampling of shallow and deep parts of the surficial aquifeir. The hole
for the deeper well in a cluster was drilled first, then a gamma log was run through the auger flights before the well casing was put
in place to obtain lithostratigraphic information, and optimize the placement of the screen on the shallow well(s)? and to avoid clay
lenses. In most cases, split-spoon samples were taken 0.5 m above and below the proposed depth for the screen. Because the upper
part of the surficial aquifer generally is very permeable throughout the study area, some of the shallow wells that are 9 m (30 ft)
deep or less were installed directly without running a gamma log first. Any wells that were not logged during installation had
gamma logs run the following week.
All wells were constructed with 5-cm-diameter PVC casing with internal threads. Screens have approximately 0.25-mm slots; most
screens installed were 0.45 or 0.6 m in length. In some cases, 0.3-m screens were installed to sample ground Water from a very
narrow interval; for example,- near the sediment redox transition at the base of the surficial aquifer associated with the first confining
layer.
The annular space between the casing and the walls of the 15-cm hole drilled with the auger was backfilled using materials dug from
the hole. The upper part of each hole was sealed with at least 0.6 m of bentonite at 1.5 m below land surface. Because most of the
wells were placed along the right-of-way of State or County roads, or along the edges of fields or service roads on private property,
all casings were cut off below land surface and a metal well cover was mounted flush with the land surface to avoid damage to the
wells by mowers and other equipment, and to avoid damage to the mowers and equipment from the wells.
AH wells were surveyed for elevation of the measuring point of the well casing and the adjacent land surface using a Zeiss optical
surveying instrument. Survey transects were referenced to USGS benchmarks, or to a previously surveyed well if no benchmark
10
-------
was near. Survey transects were closed by backsighting to temporary benchmarks; all elevations are accurate to 0.3 cm and are
reported relative to sea level.
Latitude and longitude positions for all new USGS wells and private wells used during the project were measured with a Trimble
Pathfinder differential global positioning system (GPS) receiver. Coordinates of discrete reference locations were measured
repeatedly on different days; reproducibility was 0.1 second or better (approximately 3 m). All new wells were marked with the
well identification number on the cap and with an attached aluminum permit tag.
The new wells were developed by purging either manually with a bailer or with a downhole pump until the water was clear of
suspended solids. Some wells with high-turbidity water were revisited over a 2-week period for additional purging.
Two multiport ground-water samplers were installed to obtain water samples near the sediment redox transition at closely spaced
vertical intervals. The locations of these samplers are shown in Figure 2. They were installed at sites in which clusters of wells with
single, short (1- to 1.5-ft) screened intervals were completed at depths bracketing the redox transition zone. Split-spoon sediment
samples were collected during the installation of both multiport ground-water samplers to confirm the depth of the sediment redox
transition and to position the ports accordingly.
The multiport ground-water samplers were constructed with 5-cm-diameter schedule-80 PVC threaded casing. Each sampling port
is constructed of 1-cm-diameter Teflon tubing, which runs down the inside of the casing. The downhole end of each tube exits the
casing through a hole (port) drilled through the side of the casing. The end of the tube is cut diagonally, covered with a fine nylon
mesh, and secured to the casing with nylon pull-ties. Ports are spaced 30 cm (1 ft) apart on alternating sides in the lower 3 m (10 ft)
of the casing in both wells. In sampler KE Be 209, ports in an additional 3-m section from the bottom of the well are spaced 09m
(3 ft) apart. F
To finish the well, the annular space between the casing of the multiport well and the wall of the augered hole was filled with the
native material drilled from the hole. The top of each tube was covered with a tight-fitting removable cap, and the bundle of tubing
was coiled inside a plastic bag and secured in an alcove beneath the flush-mounted well cover. The tubes were color-coded and cut
to different lengths to identify the sampling depth.
The static water level in observation wells was measured by an electrical water-level depth sensor from a marked measurement
point on the well casing, and recorded to the nearest 0101 ft. This follows the procedure described in the National Handbook of
Recommended Methods for Water-Data Acquisition (U.S. Geological Survey, 1977) and by Hardy and others (1989). Depth below
land surface and altitude relative to sea level are reported in feet (the units of the field measurement) and meters. All wells in the
network were measured monthly from May 1998 through October 1999; existing wells from the previous NAWQA project also
were measured in October 1997 and March 1998. The water-level measurements are listed in Appendix C.
Wefl Sampling
Water samples for chemical analysis were collected from 11 wells during July 1998, and from the multiport sampler KE Be 190
(Figure 2) in March 2000. Three Wells were selected to examine the confined Hornerstown aquifer, two wells were selected to
obtain data from a previously unsampled recharge area in the surficial aquifer south of Morgan Creek, and the rest of the wells and
the multiport samplers were selected to obtain samples from the redox-transition zone. No ground-water samples were collected
below the redox transition from multiport sampler KE Be 209 because no water could be pumped from the low-permeability
sediments of the Aquia confining layer. Results of the chemical analyses are in Appendix D.
Water samples for analyses of major ions, nutrients, dissolved gases, and stable isotopes were collected from wells with a positive-
displacement, battery-operated submersible pump with a Teflon discharge tube. Samples for analyses of major ions, nutrients, and
dissolved gases were collected from the multiport sampler using a peristaltic pump and a silicone discharge tube.
Water samples were collected for the chlorofluorocarbon (CFC) age-dating tracers (Busenberg and Plummer, 1992) with a
peristaltic pump and a copper discharge tube. The date of ground-water recharge (and thereby the age of the water) was calculated
by relating historical atmospheric concentrations of CFC-11, CFC-12, and CFC-113 to concentrations measured in the ground
water. The ages were estimated by modeling expected recharge concentrations of the three CFC compounds based on Henry's Law
(Busenberg and Plummer, 1992). Three replicate samples from each well were analyzed, and the reported ground-water age
represents the average of the three replicates (Table 3). Because age estimates from the different CFC compounds varied, the final
ground-water age reported in Appendix D is the average of the ages determined by each individual CFC compound.
Base-Flow Sampling
The seasonal and spatial variability of nitrogen discharge from stream base flow in the Morgan Creek and Chesterville Branch
watersheds were measured in three synoptic surveys from a network of 14 sites (Figure 2). Of these sites, eight are at outlets of first-
order drainage basins; the other six sites are on the mainstems of Morgan Creek and Chesterville Branch. The two sites farthest
downstream (sites 01493500 and 01493112) are located close to the transition between nontidal and tidal conditions. The base-
flow sampling network was designed to describe the areal distribution of base-flow nitrogen and how nitrogen discharge varies
downstream within the watersheds. Samples collected at the two outlets were used to estimate the total discharge of nitrogen from
11
-------
Table 3. Chlorofluorocarbon (CFC) Recharge Dates and Ground-water Ages from Wells Sampled in 1998
Well
Number
ICE Be 50
KB Be 158
KB Be 189
Homerstown
aquifer
KEBcl92
KEBel93
KEBel94
KB Be 195
deepinAquia
aquifer
KB Be 199
deepinAquia
aquifer
KB Be 200
Homerstown
aquifer
KB Be 206
ICE Be 207
deepinAquia
aquifer
KB Be 207
deepinAquia
aquifer
KB Be 208
deepinAquia
aquifer
KB Be 208
deepinAquia
aquifer
KEBe210
Homerstmvn
aquifer
KB Be 211
Homerstown
aquifer
ICE Be 212
KB Be 216
Sampling
Date
07/24/98
07/21/98
07/21/98
07/21/98
07/24/98
07/20/98
07/20/98
07/22/98
07/22/98
07/22/98
07/23/98
12/10/98
07/23/98
12/10/98
07/22/98
12/11/98
07/23/98
07/24/98
CFC Recharge Dates
CFC-11
1989.5
1989.5
1988.0
1985.5
1987.0
1986.5
1949.0
1948.5
<1945
Modern
1993.5/1995.5
Modem
Contaminated
Contaminated
Contaminated
1985.5
1985.5
1985.5
1977.5
1977.5
1977.5
1978.0
1978.0
1976.5
1954.5
1955.0
1953.5
1982.0
1981.5
1981.5
1978.0
1977.5
1977.5
1977.0
1977.
1977.0
1967.
1967.
1967.
1968.
1968.
1968.
1954.
1949.
<194
<194
<194
1986.
1987.
1986.
1987.
1987.
1987.
CFC-12
Modem
Contaminated
1991.5
1986.5
1991.5
1991.5
1954.0
1951.5
1952.5
1997.0
1996.0
1997.0
Contaminated
Modem
Contaminated
1987.5
1988.0
1987.5
1976.5
1976.0
1976.0
1976.5
1978.0
1975.5
1955.0
1964.5
1949.5
1982.0
1982.0
1982.5
1978.0
1978.0
1978.0
1978.0
1978.0
J977.5
1967.0
1967.0
1967.
1968.
1968.
1968.
1959.
1949.
1949.
<194
<194
.1991.
1992.
1992.
1993.
1994.
1991.
CFC-113
1990.5
Modem
1988.5
1988.0
1990.0
1993.0/1995.0
<1955
<1955
<1955
Contaminated
Modern
Modern
Contaminated
Contaminated
Contaminated
1986.5
1991.5/1998.0
1987.5
•1978.0
1980.5
1979.0
1976.5
1976.5
1976.0
1965.5
OSS
<1955
1981.5
1980.5
1980.0
1980.5
1979.5
1977.0
1979.5
1980.0
1979.5
1967.
1967.0
1966.
1968.
1966.
1967.
<195
<195
<195
<195
<195
1991.
1989.
Modem
1990.
1988.
1990.
CFC Recharge Ages
Average
CFC-11
9.6
12.2
-50
I
n/d
n/d
13.1
21.1
21.1
44.2
16.9
20.9
21.
31.
30.
48.
>5
11.
11.
Average
CFC-12
7.1
8.7
45.9
1.9
n/d
10.9
22.4
21.9
42.2
16.4
20.6
21.
31.4
30.
46.
>5
6.
5.
Average
CFC-113
9.1
9.6
>43
n/d
n/d
11.6
19.4
22.2
40.0
17.9
19.6
19.3
31.
31.
>4
>44
8.
8.
Average
Recharge
Age (years)
And Date'
8.6
1990
10.2 :
1988.5
>45
<1953
1.9 .
1996.5 ,
n/d
11.8 !
1987
1
20.9
1978 :
21.7 ,
1977
42.1 ,
1956 i
17.1
1981.5
20.3
1978.5;
20.7 ,
1978 ,
31.5
1967
31.1
1967.51
46 i
<1952
>50
<1950|
8.9
1989.5
8.7 !
1989.5
rCFC Recharge Dates' are the calculated years In which each sampled parcel of ground water recharged the water table,,based on historical
atmospheric CFC concentrations; "Modem" indicates a CFC concentration similar to that expected tor water in equilibrium with 1998-99
atmospheric concentration; "Contaminated" indicates a CFC concentration higher than that expected from equilibrium with current or historical
atmospheric concentrations; n/d - not determined; ~, approximately; <, less than; >, greater than.]
12
-------
the two watersheds during the study period. Samples were analyzed for nitrite-plus-nitrate nitrogen and ammonia-plus-organic
(Kjeldahl) nitrogen. These two analytes were added to obtain the total dissolved nitrogen concentration. The instantaneous
nitrogen load was calculated by multiplying the measured stream discharge (in units of volume per time) by the total dissolved
nitrogen concentration (in units of mass per volume) and adjusting the units to obtain values in gm sec'1 (units of mass per time).
Instantaneous yields were calculated by dividing the load by the drainage area to obtain units of gm sec'1 km'2.
Stream base-flow samples were collected from the entire network during synoptic surveys in April 1998, September 1998, and
February 2000. Samples were collected during both winter base flow and summer base flow to evaluate the effect of seasonal
differences in base-flow discharge and biological activity in the stream. Bachman and Phillips (1996) noted that base-flow
discharge of water in the central upland of the Delmarva Peninsula was approximately an order of magnitude lower in summer than
in winter, when evapotranspiration is minimal. They also noted that over the entire Delmarva Peninsula, base-flow nitrogen loads
and yields tended to be higher in whiter. Base-flow samples also were collected monthly at the two downstream gaging stations
from April 1998 through March 1999. These monthly samples were supplemented by additional base-flow samples collected at
Chesterville Branch (site 01493112) before and after this period for other USGS projects. All base-flow samples were accompanied
by stream-discharge measurements made by use of standard USGS methods (U.S. Geological Survey, 1977). Results of the
chemical analyses of the base-flow samples are listed in Appendix E.
Chemical Analysis of Water Samples
Chemical analyses were made on samples filtered through a membrane filter with a 0.45-um pore diameter to provide an
approximation of the "dissolved" concentration of solutes. In addition, unfiltered base-flow samples were analyzed for nitrogen
species and phosphate to determine the "total" (suspended and dissolved) concentrations of these constituents. Water samples
collected for CFC age-dating tracers and other dissolved gases were not filtered.
Many physical and chemical properties are unstable, and, therefore, were measured in the field. These include pH, temperature,
specific conductance, dissolved oxygen (measured on unfiltered samples), and carbonate alkalinity (measured on filtered samples).
The analysis methods follow those recommended by Wilde and Radtke (1998).
Samples for the analysis of major ions, silica, iron, nutrients, and total organic carbon were sent to the USGS NWQL in Arvada,
Colorado. Samples for stable isotopes, dissolved gases, and chlorofluorocarbon age-dating tracers, and a subset of samples for
major ions and nutrients were analyzed at the laboratories of the USGS National Research Program in Reston, Virginia.
Cations and silica were analyzed by plasma emission spectroscopy, sulfate and chloride by ion chromatography, and nitrogen
species and phosphate by various standard colorimetric methods (Fishman and Friedman, 1985). Quality-assurance methods used
at NWQL are described by Pritt and Raese (1995). In addition, field replicates were collected and analyzed. From these replicates,
variation (using the standard deviation of replicates as described by Taylor, 1988) of dissolved nitrogen concentrations due to
sampling and analytical methods is estimated to be 0.09 mg/L; variation of calcium, magnesium, and silica concentrations are less
than the precision of the analysis as described by Fishman and Friedman (1985).
Data Analysis
Processes affecting ground-water and surface-water geochemistry were evaluated by examining the relations between selected
constituents and their location in the watershed or along a ground-water flow path. Previous studies indicated that, in general, the
surficial aquifer has multiple ground-water flow paths with traveltimes (from recharge to discharge) that range from a year to
decades (Dunkle and others, 1993; Bohlke and Denver, 1995). The older ground water (from flow paths with long traveltimes)
tends to have lower nitrate concentrations than younger ground water (from flow paths with short traveltimes). The relative
proportion of old and young ground water discharging to a stream upstream of a sampling point will affect the concentration of
solutes in base flow at that sampling point. In a detailed investigation, Modica and others (1998) evaluated the flow systems,
ground-water traveltimes, and nitrate concentrations in a similar watershed in the Coastal Plain of New Jersey. Their model
simulation showed that ground-water discharge in the headwaters of a stream is dominated by relatively young water, whereas
farther downstream, ground-water discharge contains an appreciable component of older water mixed with the young water
(Figure 4). In order to evaluate this process in the Locust Grove study area, concentrations of nitrogen and other constituents that
are geochemical markers for old and young ground water were examined at seven sites along the mainstems of Morgan Creek and
Chesterville Branch. Observed trends in concentrations were compared to trends expected from various mixtures of old and young
ground water.
Although samples were analyzed for both nitrogen and phosphorus, this report focuses on the nitrogen data. Because phosphorus
adsorbs to soil particles under oxic conditions, it is likely that little phosphorus enters the oxygenated environment of the surficial
aquifer with recharging ground water. Further, the near-surface sediments in this area are not rich in phosphate, and even if there
were chemically reducing environments in the surficial aquifer, there is little potential for the mobilization of phosphorus. There is
some question as to whether the ground-water sampling methods used in this project actually measure dissolved phosphorus in the
aquifer, or colloidal particulate phosphorus that was stirred up by the sampling pumps and passed through the 0.45-um filters used
to separate the "dissolved" from "particulate" fractions (M. T. Koterba, U.S. Geological Survey, oral commun., 1999).
13
-------
0.6
jlj
2
O
CO
1
s
HI
CD 0.3
HI
u.
O
UPSTREAM
DOWNSTREAM
YOUNG
OLD
RELATIVE GROUND-WATER AGE
Figure 4. Relative distribution of the age of discharging ground water in stream headwaters and downstream (adapted from
Modica and others, 1998). '
14
-------
Hydrogeologic Framework
The study area is located in the updip region of the Delmarva Peninsula, approximately 30 km from the Fall Line (Figure la). A
broad band of lower Tertiary and upper Cretaceous marine sedimentary sequences crops out or subcrops shallowly between the
Sassafras and Chester Rivers (Owens, 1967; Drammond, 1998). These Cretaceous and lower Tertiary units dip and thicken to the
southeast with the regional dip of crystalline rocks beneath the Coastal Plain.
Formations exposed above sea level in the area near Locust Grove (Figure 5) include the upper Cretaceous Mt. Laurel and Severn
Formations of the Monmouth Group, the lower Paleocene Hornerstown Formation, and the upper Paleocene Aquia Formation
(Minard, 1974; Hansen, 1992; Drummond, 1998). In the southeast corner of Kent County, including the southeast part of the study
area, younger marine sediments of the upper Oligocene Old Church (?) Formation and the Miocene Calvert Formation may overlie
the Aquia Formation (Hansen, 1992; Drummond, 1998). However, the Oligocene and Miocene sediments generally are very thin
(less than 1-2 m thick) north of the Chester River, and probably are not important as hydrostratigraphic units.
CO
DC
in
\a
1
I
til
m
in
Q
NATURAL GAMMA RAY LOG
Well KE Be 186 Well KE-94-0146
HORIZONS
ON GAMMA LOG
LITHOSTRATIGRAPHIC
UNITS
HYDROSTRATIGRAPHIC
UNITS
0 -
10
20
30
40
50
60
70
80
_ HORNERSTOWN
CONFINING
c LAYER
SEVERE
-CONFINING
LAYER
a land surface
b base of Pensauken formation,
top of Aquia Formation
c base of Aquia aquifer, top of
Aquia confining layer
d top of tight Aquia confining
layer
e peak gamma response in
Aquia confining layer
f base of Aquia confining layer,
top of Hornerstown Formation
g top of sand of the
Hornerstown aquifer
h base of Hornerstown aquifer,
top of Hornerstown confining
layer
I peak gamma response in
Hornerstown confining layer
j base of Hornerstown
confining layer
j-k sandy bed in upper Severn
Formation'
k top of Severn confining layer
I peak gamma response in
Severn confining layer • . .
m base of Severn confining
layer, top of Monmouth
aquifer
£ £
Sz Sz
1 E
HI
DC HI
£8
2
UJ
SQ
§1
UJ
g
5
0.
PENSAUKEN
FORMATION
AQUIA
FORMATION
HORNERSTOWN
FORMATION
SEVERN
FORMATION
MT. LAUREL
FORMATION
DC
B1
•*
~4
s
a
£
g
'
. ,'
COLUMBIA
AQUIFER
AQUIA
AQUIFER
AQUIA "
CONFINING
LAYER
HORNERSTOWN
AQUIFER
t HORNERSTOWN
- ^ CONFINING "*
„
COI
c
LAYER ^ ^
r /f -i *1
SEVERN
'JFINING LAYER
MONMOUTH
AQUIFER
0 20 40 60 80 100
GAMMA, IN COUNTS PER SECOND
Figure 5. Composite hydrostratigraphic sequence of the Locust Grove study area, Kent County, Maryland.
15
-------
The upland in central and eastern Kent County above an altitude of approximately 15m (50 ft) is capped by a gravelly coarse sand.
This unit has been mapped as the Pensauken Formation on the northern and central Delmarva Peninsula by correlation with similar
deposits along the flanks of the Delaware River valley in New Jersey (Owens and Denny, 1979; Owens and Minard, 1979). The age
of the Pensauken Formation has been estimated as latest Miocene to early Pliocene based on its stratigraphic position relative to
other dated units in New Jersey. Lithologically similar deposits that cover two-thirds of Delaware have been mapped as the
Columbia Formation (Jordan, 1962; 1964). Age estimates for the Columbia Formation are early to middle Pleistocene based on
pollen assemblages (Groot and Jordan, 1999). It is not clear whether the sediments mapped as the Pensauken Formation in
Maryland and the Columbia Formation in Delaware were deposited during the same period, or whether they represent two or more
dcpositional episodes with similar sediment provenance and depositional environment. :
Middle to upper Quaternary deposits in Kent County include Pleistocene estuarine-terrace deposits and Holocene alluvial
sediments. Pleistocene terraces are restricted to the margins of the Sassafras and Chester Rivers, the mouths of the larger tributary
creeks, and the Chesapeake Bay shoreline in western Kent County. The highest terraces are at altitudes of approximately 15m
above sea level. Although the lower reaches of Morgan Creek and Chesterville Branch cut across the Pleistocene terraces along the
Chester River, no Pleistocene estuarine deposits are within the immediate study area. However, the incised valleys of Morgan
Creek and Chesterville Branch are filled with Holocene alluvial sediments derived from the erosion of the adjacent upland.
Most of the upland in the study area away from the stream valleys is covered with a surficial deposit of loess (wind-blown silt) that
may be 0.3 to 1 m thick. Previous studies of soils in the northern Delmarva Peninsula have attributed the source of the loess to
estuarine silts that were deposited in the Delaware River estuary and subsequently exposed during late Quaternary lowstands of sea
level to be eroded and transported by winds (Carey and others, 1976).
Lithostratigraphic Units
Four primary lithostratigraphic units compose the upper 30 m of the sedimentary column in the Locust Grove study area. These
units are:
(1) the Holocene alluvial sediments that fill the incised valleys of Morgan Creek and Chesterville Branch;
(2) the coarse fluvial sands and gravels of the Pensauken Formation that cap the upland; ,
(3) the Aquia Formation, which immediately underlies the Pensauken Formation throughout most of the study area; and
(4) the Homerstown Formation, which underlies the Aquia Formation.
In this report, the Severn confining layer (Figure 5) is considered as the base of the active hydrogeologic system that interacts with
the streams in the study area, and, therefore, descriptions of underlying stratigraphic units are omitted.
Developing the hydrostratigraphic framework for the Locust Grove study area was an iterative process of interpreting cores and
geophysical logs, correlating lithostratigraphic units across the study area, and defining hydrostratigraphic units. The first step was
the description and interpretation of the lithology of the four new cores drilled for the project (Table 2). These cores were not
studied in the same detail as core KE Bf 180 (Hansen, 1992), and inference of physical properties and other characteristics is based
on correlation with core KE Bf 180. The new cores were described in sufficient detail to identify changes in grain size that relate
to permeability, and mineral composition and degree of weathering that relate to redox chemistry. The lithologic changes were
matched with excursions in the geophysical logs (natural gamma, electrical resistivity, and spontaneous potential) from the
coreholes. These features in the logs were used to define a set of specific horizons in the geophysical logs wjith consistent and
characteristic responses that could then be used to correlate lithostratigraphic horizons among the geophysical logs from all wells
in the study area. i
An initial stratigraphic framework was developed by correlating the geophysical logs of the deepest wells and coreholes to
determine the general geometry and distribution of units. This correlation also was used to assemble a composite stratigraphic
sequence and define hydrostratigraphic units for the study area (Figure 5). Specific surfaces, such as the top of the clay-rich section
in the lower Aquia Formation, were mapped throughout the study area. This preliminary framework was refined subsequently by
checking the consistency of picks in cases where the interpretation of the geophysical log was uncertain, and by inferring the depths
to specific horizons in shallow wells. . . ! .
Holocene Alluvia! Sediments :
The valleys of Morgan Creek and Chesterville Branch are filled with 4 to 6 m of alluvium in the axis of the valleys, although this
thickness is inferred from relatively few holes. These sediments are dominantly coarse sands with gravels that wejre eroded from the
upland, and overlain by silty overbank deposits. The sands hi the lower part of the alluvium may have interbedded lenses of silt,
poorly sorted silty sand, and peats; some beds also contain trace amounts of glauconite reworked from the Aquia Formation. The
silt overbank deposits that form the modern floodplain typically are 0.5 to 1 m thick and contain organic material accumulated from
the riparian forest Bedding within the alluvial sediments represents transport and deposition during episodic storm events. The
alluvial valley of Chesterville Branch at Road 444 is 50 to 70 m wide; the valley of Morgan Creek at Road 298 is 80 to 100 m wide.
16
-------
Pensauken Formation, Upper Miocene (?)/Lower Pliocene (?)
The Pensauken Formation in Kent County is a thick sequence of medium to coarse sands and gravels, with common cobbles and
less common boulders. Sandy beds typically are poorly sorted and may have interbedded lenses of silt or clay. In exposures, this
unit contains cross-bedded sets in cut-and-fill channels 2 to 5 m deep. The sand grains generally are subrounded to subangular, and
are predominantly quartz with some feldspar. The fine-sand fraction may contain abundant heavy minerals; Hansen (1992)
reported the mineralogy of the heavy mineral sands. X-ray diffraction analyses of samples from core KE Bf 180 showed kaolinite,
illite, and smectite in the clay fraction (Hansen, 1992). Pebbles are rounded and dominantly vein quartz, quartzite, and chert, with
uncommon lithic clasts. The sediments of the Pensauken Formation typically are deeply weathered; the dominant colors are
yellowish-oranges, light to very light browns (tan), and light grays. Much of the feldspar and labile heavy mineral suite has been
leached and hydrated to form clays.
The sediments of the Pensauken Formation were deposited by a very large braided-river system mat covered most of the central
Delmarva Peninsula (Owens and Denny, 1979). Fluvial erosion during the deposition of the Pensauken Formation truncated the
older Tertiary and Cretaceous sediments to create a regional angular unconformity. The base of the Pensauken Formation
underlying the Locust Grove study area generally is between 12 and 15 m asl, but is very irregular because of channelization and
scouring. The unit typically is 6 to 15 m thick, but Minard (1974) reported a maximum observed thickness of 44 m (145 ft) in a
paleochannel. Determining the base of the Pensauken Formation in the Locust Grove study area from geophysical logs without
lithologic control was difficult because the gamma response of the lower part of the Pensauken is very similar to that of the sand
facies in the upper Aquia Formation.
Old Church (?) Formation, Upper Oligocene, and Culvert Formation, lower to Middle Miocene
Hansen (1992) reported 1.5 m (5 ft) of a glauconitic fine to medium sand with a late Oligocene foraminiferal assemblage hi core KE
Bf 180. This unit was tentatively assigned to the Old Church Formation. Drummond (1998) showed the updip limit of the Calvert
Formation just to the north of the Chester River; the Calvert may extend into the southeast comer of the Locust Grove study area,
although it was not identified in core KE Bf 180. The Calvert Formation typically is a compacted marine silt with low permeability,
and is a regional confining or semi-confining layer south of the Chester River. Neither Old Church nor Calvert sediments were
confirmed in cores or geophysical logs collected for this project. However, a thin bed of silly fine sand similar to the description
of the Old Church (?) Formation was observed when augering the hole for well KE Be 216, which is approximately 700 m
northwest (updip) from corehole KE Bf 180.
Aquia Formation, Upper Paleocene
The upper Paleocene Aquia Formation underlies the Pensauken Formation throughout the study area except where the Pensauken
has been removed by erosion. The Aquia Formation is a coarsening-upward marine sequence with two distinct facies. The lower
section of the Aquia is a fine to medium sand with abundant (30 to 50 percent) glauconite grains and a sticky silt-clay matrix. When
seen in fresh cores, the sediments of this facies are dark greenish-gray to dusky blue-green, and show no evidence of leaching or
oxidation except at the upper and lower boundaries. Hansen (1992) and Minard (1974) give detailed descriptions of the lithology
of this facies, including the characteristics of the glauconite grains. Mollusk and brachiopod shells are preserved in this section of
the Aquia in coreholes KE Bf 180 (Hansen, 1992) and KE Be 186. The upper section of the Aquia Formation is a medium quartz
sand with a minor component of coarse sand grams and 5 to 15 percent glauconite. This section is weathered throughout its vertical
extent, and there is little silt-clay matrix between th&sand grains. The sediments of the upper Aquia are light olive to light olive-
brown with dark reddish-brown to dark gray flecks of glauconite, giving a "salt and pepper" appearance. Individual glauconite
grains in the upper section commonly have oxidized rinds or are worn and polished.
The transition between the upper sandy facies and the lower facies with the silt-clay matrix, referred to as the "redox transition
zone," occurs in a 0.5 to 1 m interval at the base of the sandy facies. This textural change coincides with a transition from leached
and oxidized sediments above to relatively unaltered sediments with reduced mineral phases below. Iron oxides were observed
near this boundary in the cores. Joints and fractures that extend below this textural and sediment-redox transition have brown or
reddish-brown oxidized zones 1 to 5 mm wide, indicating localized flow of oxygenated ground water into this layer. A second
sediment-redox transition was observed at .the base of the Aquia silt-clay facies, with oxidation proceeding upward from the
underlying sandy facies of the Hornerstown Formation. This lower redox transition was most prominent in corehole KE Be 183,
but was observed in all four cores collected in this project.
The updip limit of the Aquia Formation appears to be in a zone trending west-southwest to east-northeast immediately north of
Route 213 (Figure 2) at the northern boundary of the study area, but its exact location is poorly delineated by Hie available data. The
updip limit of the Aquia aquifer shown in a subcrop map for Kent County published by Drummond (1998) includes the
Hornerstown Formation as part of the Aquia aquifer, and does not specifically represent the areal extent of the Aquia Formation.
The silt-clay facies at the base of the Aquia Formation is within 14 m of the land surface in the three northernmost wells
(KE Be 187, KE Be 197, and KE Ae 29), and corehole KE Be 185 (Locust Grove) (Figure 2). The base of the Aquia Formation
drops from 10 m asl in corehole KE Be 185 (Locust Grove) at the northern boundary of the study area, to 21 m below sea level in
KE Bf 180 at the southern boundary. The Aquia thickens from 5 m to 28 m in the 6 km between the two coreholes. The
17
-------
stratigraphic boundary between the upper and lower fades of the Aquia Formation appears clearly in gamma logs (Figure 5) and
was used as a marker horizon for correlation throughout the study area. Where the entire sequence of the Aquia Formation is
preserved downdip, the upper and lower facies are 19 and 9 m thick, respectively. :
Homerstown Formation, Lower Paleocene
The lower Paleocene Homerstown Formation is lithologically very similar to the overlying upper Paleocene Aquja Formation. The
Hornerstown Formation also is a coarsening-upward sequence of marine glauconitic sands. In the study area, only corehole KE Bf
180 and wells KE-94-0146 and KE Be 171 penetrate the entire thickness of the Hornerstown Formation. The;coreholes and the
deeper wells drilled for this project were completed in the upper section of the Homerstown Formation. The lower 15 to 16 m of
the Homerstown Formation has a substantial clay content based on the gamma logs; a prominent gamma excursion in the lower 2 m
of the unit probably is a relatively low-permeability clay bed. A larger gamma excursion in the underlying ;Sevem Formation
indicates a clay bed that is a regional confining layer and an excellent marker horizon (Figure 5). The upper 10 to 13 m of the
Hornerstown Formation is predominantly sand. However, the uppermost 2 to 4m of the Hornerstown, immediately below the silt-
clay-rich facies of the Aquia Formation, commonly has a substantial clay content, although not as much as in the lower section of
the Hornerstown.
The contacts between the Aquia and Homerstown Formations in the four cores collected in this project were not as distinct as the
one described in core KE Bf 180 (Hansen, 1992). In this project, the base of the Aquia Formation was interpreted consistently as
the transition between the bottom of the silt-clay-rich facies of the Aquia and the top of the sandy facies of theiHornerstown; this
boundary can be identified readily in most of the geophysical logs. The actual stratigraphic unconformity between the two
formations could be verified only by detailed biostratigraphy across this interval.
Composite Hydrostratigraphic Sequence
Hydrostratigraphic units in the Locust Grove study area were defined from the permeabilities inferred from the sediment textures
in the cores and the character of the geophysical logs. Five hydrostratigraphic units were identified in the surficial and first confined
aquifer systems: the Columbia aquifer and the Aquia aquifer, which together make up the surficial aquifer, the Aquia confining
layer, the Hornerstown aquifer, and the Hornerstown confining layer. A very low-permeability clay bed in the Severn Formation
lies immediately below the Hornerstown confining layer throughout most of the study area to form a regional confining layer.
Surficial Aquifer (Columbia and Aquia Aquifers)
By regional usage for the Delmarva Peninsula, the term "Columbia aquifer" includes the Pensauken Formation in Maryland and the
Columbia Formation in Delaware (Bachman 1984b). Drummond (1998) also included the Pleistocene terrace deposits in his
definition of the Columbia aquifer in Kent County, Maryland. Because no Pleistocene terrace deposits occur within the study area,
the Columbia aquifer as used in this project is restricted to the sands and gravels of the Pensauken Formation. In the Locust Grove
study area, the Columbia aquifer is less than 2.5 m thick updip in well KE Be 187, and greater than 10 m thick downdip. The base
of the Columbia aquifer cannot be readily identified in geophysical logs without lithologic control. !
The upper sandy facies of the Aquia Formation is designated the "Aquia aquifer." In previous usage by other investigators, the
Aquia aquifer in Kent County, Maryland, includes the glauconitic sands of the Aquia and Homerstown Formations (Overbeck and
Slaughter, 1958; Otton and Mandle, 1984; Hansen, 1992; Drummond, 1998). However, the cores obtained as part of mis project
indicate that the silt-clay layer at the base of the Aquia Formation is very impermeable (see "Aquia confining layer," below), and
there is much less hydraulic connection between the Aquia and Hornerstown Formations than between the Aquia and overlying
Pensauken Formation. Thus, the "surficial aquifer" is defined here as the combined Pensauken Formation and|upper sandy facies
of the Aquia Formation, and referred to as the Columbia-Aquia aquifer. As the sandy facies of the Aquia Formation pinches out
updip, toward the north of the study area, the surficial aquifer is composed solely of the Pensauken Formation, or Columbia aquifer.
AquiaConfiningLayer
The silt-clay at the base of the Aquia Formation was recognized in previous studies, but was considered permeable or at least semi-
permeable, and therefore the Aquia and Hornerstown Formations were interpreted as hydraulically connected. The silt-clay facies
at the base of the Aquia Formation typically is from 6 to 9 m thick and can be identified in geophysical logs throughout the study
area. A very low-permeability clay-rich bed from 2 to 3 m thick occurs in the lower third of the unit and is associated with a peak
in the gamma logs. Fresh core material from this clay-rich bed, although plastic, was nearly dry in appearance. Geochemical and
age-dating evidence, discussed in the Ground-Water Geochemistry section, show substantial differences in the properties of ground
water above and below the silt-clay layer at the base of the Aquia Formation. These lines of evidence suggest that the lower Aquia
Formation is an effective confining layer. The possible exceptions to this are in areas of enhanced vertical hydraulic gradients, such
as discharge areas beneath streams and near production wells pumping from the sands in the upper part of the Homerstown
Formation. •
18
-------
Hornerstown Aquifer andHornerstown Confining Layer
The Hornerstown Formation, similar to the Aquia Formation, has an upper sandy facies and a lower silt-clay-rich facies that are
designated the "Hornerstown aquifer" and the "Hornerstown confining layer," respectively (Figure 5). The thickness and clay
content inferred from gamma logs of these two Hornerstown units are more variable than those of the Aquia Formation in the
Locust Grove study area and across Kent County (see the hydrogeologic sections of Drummond, 1998). In the study area, the
Hornerstown aquifer is from 9 to 11 m thick, and the Hornerstown confining layer is from 10 to 15m thick. The Hornerstown
confining layer is underlain by the Severn Formation, which includes at its base approximately 5 m of low-permeability clay; the
two units together form a regional confining layer that separates the Aquia/Hornerstown aquifer system from the Monmouth
aquifer. The subcrop area of the Hornerstown aquifer is to the north and northwest of Route 213 (Figure Ib). Part of this area is
deeply incised by tributaries extending south from the Sassafras River. The most likely recharge area for the Hornerstown aquifer
is the upland plain north of Kennedyville, Maryland, and northwest of the intersection of Routes 213 and 298.
Geometry of the Hydrostratigraphic Units
The five hydrostratigraphic sections presented in Figures 6 through 10 show the geometry of the primary hydrostratigraphic units,
coreholes, and wells in the Locust Grove study area. These sections were prepared by correlating lithostratigraphic and
hydrostratigraphic units away from the five coreholes using the geophysical logs from the deeper wells at each site. Sections A-A'
(Figure 6) and B-B' (Figure 7) run mostly southwest to northeast, close to the regional strike of the Cretaceous and Tertiary units.
Sections C-C" (Figure 8), D-D' (Figure 9), and E-E' (Figure 10) run mostly north to south, close to the regional dip. Section D-D'
was the primary transect for the previous NAWQA project (Hamilton and others, 1993) and ground-water dating and flow-path
modeling studies (Dunkle and others, 1993; Reilly and others, 1994).
Hydrostratigraphic section A-A' (Figure 6) runs west-southwest to east-northeast along Route 213 at the northern boundary of the
study area. This transect runs along the southern edge of the subcrop zone of the Aquia confining layer (Aquia CL). The surficial
aquifer is very thin, less than 3 m thick in places, because the sandy facies of the Aquia Formation (the Aquia aquifer, or Aquia AQ)
was removed by erosion prior to or during the deposition of the Pensauken Formation. The sands of the Pensauken Formation (the
Columbia aquifer) also thin across a subsurface high that probably was created because the silt-clay facies in the lower Aquia
Formation (the Aquia CL) is more resistant to erosion than the sands in the upper Aquia Formation. The updip limit of the
subcropping Aquia CL and the Hornerstown aquifer (Hornerstown AQ) are not known precisely because of a lack of boreholes for
control.
Section A-A' straddles the surface-water drainage divide between the tributaries of the Sassafras River to the north and Morgan
Creek to the south. The saturated zone in the surficial aquifer overlying the Aquia CL is very thin. Wells KE Be 187 and KE Be
197 are screened at the base of the surficial aquifer. For well KE Be 187, the maximum thickness of the saturated zone was 1.5 m
in May 1998 and the minimum was 0.1 m in May 1999; for well KE Be 197, the maximum was 0.8 m in May 1998, and the well
was dry from November 1998 through August 1999, during a prolonged drought. It is inferred that ground water drains quickly off
the subsurface high of the subcropping Aquia CL through the coarse sands and gravels of the Pensauken Formation.
Hydrostratigraphic section B-B' (Figure 7) runs diagonally across the study area, including part of the high ground along Blacks
Station Road mat is the drainage divide between Morgan Creek and Chesterville Branch. The transect is oblique to the regional
strike, so it has a dip component. Only one well on the transect (KE Be 210) penetrates the Aquia CL; consequently,
hydrostratigraphic units could not be identified for most of this section. . . . .: • .
Hydrostratigraphic section C-C" (Figure 8) runs north-south on the western border of the study area along Road 298 and crosses
Morgan Creek. The transect extends from the northern drainage divide for Morgan Creek (at well KE Be 187) to just south of the
southern drainage divide (at the well KE Be 195 cluster). Two coreholes, KE Be 183 (Morgan Creek North) and KE Be 184
(Morgan Creek South), provide lithologic control for the updip part of the section. The Aquia CL subcrops shallowly at the north
(updip) end of the section, beneath wells KE Be 187 and 188. These two wells are screened in the surficial (Columbia) aquifer at
the top of the confining layer. The holes for these two wells were not drilled through the Aquia CL, so the thickness of the confining
layer at these sites is not known. The most complete hydrostratigraphic sequence for the updip part of the section is provided by the
gamma log of the Nau Farm well. This sequence is projected along strike approximately 1.2 km due southwest onto section C-C".
The surficial aquifer thins appreciably north of Morgan Creek as the top of the Aquia CL rises in altitude. The Pensauken
Formation typically is from 7 to 10 m thick south of Morgan Creek and .the Aquia AQ thickens appreciably downdip, although the
actual geometry cannot be represented because of a lack of deep holes south of the KE Be 184 corehole. In Morgan Creek valley,
both the Pensauken Formation and the sandy facies of the Aquia Formation (the Aquia AQ) have been removed by stream incision.
The alluvial sediments filling the incised valley are 3.5 to 4 m thick in the center of the valley and sit directly on the Aquia CL. The
upper 2 or 3 meters of the Aquia CL also may have been eroded in the deepest part of the valley.
One notable feature of the C-C" section is the abrupt offset of the base of the Aquia CL between well KE Be 189 in Morgan Creek
valley and corehole KE Be 183 approximately 350 m to the north. This offset might be related to a fault, although this is a tentative
interpretation. Supporting evidence for a fault is based on the correlation of the base of the Aquia CL and the gamma log peak in
the Aquia CL in three deep holes (coreholes KE Be 183 and 184, and well KE Be 189) near Morgan Creek valley. The updip part
19
-------
HYDROSTRATIGRAPHIC SECTION A-A'
(along Route 213)
METERS
30-1
-10-
(1511)
-20
METERS
-30 "|
EXPLANATION !
? ? HYDROSTRATIGRAPHIC CONTACT (Queried whsre uncertain)
I WELL, WITH SCREENED INTERVAL!
TDl TOTAL DEPTH OF WELL, IN METERS (m) AND FEET (ft)
(;B?ra| CONFINING LAYER
-10 CZH AQUIFER ;
-Sea Level
1 2
DISTANCE, IN KILOMETERS
Figure 6. Hydrostratigraphic section A-A' along Route 213 at the northern boundary of the Locust Grove study area, Kent
County, Maryland.
HYDROSTRATIGRAPHIC SECTION B-B1
(along Blacks Station Road and VanSants Corner Road)
B
(Southwest)
B'
(Northeast)
METERS
30-i
10-
-10-
-20
METERS
-30
DISTANCE, IN KILOMETERS
EXPLANATION
? HYDROSTRATIGRAPHIC CONTACT
(Dashed and queried whero uncertain)
I WEU, WITH SCREENED INTERVAL
TDl TOTAL DEPTH OF WELL,
IN METERS (m) AND FEET (ft)
CONFINING LAYER
I | AQUIFER '.
Figure 7. Hydrostratigraphic section B-B' along Blacks Station Road and VanSants Comer Road through th? middle of the
Locust Grove study area, Kent County, Maryland.
20
-------
HYDROSTRATIGRAPHIC SECTION C-C"
(along Route 298)
Sea Level - -
-20-
T
0 1
Vertical scale greatly
-Sea Level
EXPLANATION
---? HYDROSTRATIGRAPHIC CONTACT
(Dashed and queried where uncertain)
I WELL, WITH SCREENED INTERVAL
TDl TOTAL DEPTH OF WELL,
: IN METERS (m) AND FEET (ft)
CONFINING LAYER
I AQUIFER
DISTANCE, IN KILOMETERS
Figure 8. Hydrostratigraphic section C-C" along Harmony Woods Road (State Highway 298), the western boundary of the
Locust Grove study area, Kent County, Maryland.
Sea Level- •
HYDROSTRATIGRAPHIC SECTION D-D1
(along Locust Grove Road)
COLUMBIA AQUIFER
{PENSAUKEN FORMATION)
- -SeaLaval
EXPLANATION
? ? HYDROSTRATIGRAPHIC CONTACT
(Dashed and queried where uncertain)
I WELL, WITH SCREENED INTERVAL
TDl TOTAL DEPTH OF WELL,
IN METERS (m) AND FEET (ft)
CONFINING LAYER
I | AQUIFER
DISTANCE, IN KILOMETERS
Figure 9. Hydrostratigraphic section D-D', the primary transect along Locust Grove Road from Locust Grove to Chesterville,
in the Locust Grove study area, Kent County, Maryland.
21
-------
HYDROSTRAT1GRAPHIC SECTION E-E1
(along Bolton Road and through Angelica Nurseries)
EXPLANATION
? HYDROSTRATiSRAPHIO CONTACT
(Dashed and queried when) uncertain)
1 WELL, WITH SCREENED INTERVAL
TDI TOTAL DEPTH OF WELL,
IN METERS (ni) AND FEET (ft)
SftLmvl- •
CONFINING LAYER
| | AQUIFER i
VarOcilxfhgniaay
exaggtmltd
DISTANCE, IN KILOMETERS
Figure 10. Hydrostratigraphic section E-E', the eastern boundary of the Locust Grove study area through Angelica Nurseries,
Kent County, Maryland. \
of section C-C" (Figure 8) shows a reversal of the regional dip to the southeast between corehole KE Be 184 (soulh) and well
KE Be 189 in Morgan Creek valley, and a steep rise between KE Be 189 and corehole KE Be 183 (north). The igamma peak in the
Aquia CL and the bottom of the Aquia CL drop in altitude by 2.0 and 0.8 m, respectively, from KE Be 184 to 189, and then rise by
6.75 and 8.0 m between KE Be 189 and KE Be 183. The regional dip estimated from section D-D', which is roughly parallel to
section C-C", is from 5.3 to 5.5 m/km (meters per kilometer). The steep dip between corehole KE Be 183 and well KE Be 189 is
between 20 and 23 m/km; even when using the more conservative correlation between coreholes KE Be 183 (north) and 184
(south), the dip is between 7.0 and 10.5 m/km. If this interpretation is correct, the downdropped side of the fault lies directly below
Morgan Creek, and the upthrown side is immediately to the north of the creek. The displacement across the possible fatult is on the
order of 5 m. With only one transect crossing the potential fault, it is not possible to estimate the strike of the fault, although the
southwest-northeast trend of Morgan Creek may be a geomorphic expression.
If a fault is present beneath Morgan Creek, it may have a large effect on ground-water flow from the Hornerstown aquifer. The
entire confining layer in this area is 9-10 m thick, and the lowest-permeability part of the confining layer is 2-3 m thick. A 5-m
offset caused by a fault could breach the low-permeability part of the confining layer to create a lateral conduit that would allow
discharge of Hornerstown water directly to Morgan Creek.
Hydrostratigraphic section D-D' (Figure 9) is the primary transect running north-south through the study area from Locust Grove
to Chesterville, Maryland. This transect is very close to the regional dip, contains two coreholes (KE Be 185 and 186), and five
deep wells that penetrate the Aquia CL and part or all of the Hornerstown Formation. All of the hydrostratigraphic units of the
surficial and first confined aquifer systems are shown in this section. The updip pinchout of the Aquia AQ is apparent, although the
Pensauken Formation (Columbia aquifer) is not as thin in the Locust Grove corehole (KE Be 185) as at the site of well KE Be 187
(Figure 9). Downdip, the Old Church (?) Formation overlies the Aquia AQ; farther downdip,.south of the Chester River, the. silly
Calvert Formation overlies the Aquia and it becomes a confined aquifer. The base of the Pensauken Formation drops 5 :m in altitude
from the updip to the downdip end of the section.
Hydrostratigraphic section E-E' (Figure 10) runs sub-parallel to section D-D' and intersects D-D' at the southern end. Section E-
E* runs through Angelica Nurseries and is the eastern boundary of the study area. Because of the regional southwest-tp-northeast
strike, the Aquia AQ is thicker in section E-E' than in D-D'. Five deep holes provide excellent control for the position of the
Aquia CL. The transect crosses the three-way drainage divide between Morgan Creek, Chesterville Branch, arid Woodland Creek
(to the northeast) near the site of wells KE Be 154 and KE Be 213. The Pensauken Formation thins considerably to the south
because of the headward erosion of the Chesterville Branch drainage. ;
22
-------
The extent of the shallow subcropping of the Aquia CL (Figure 11) was mapped using the coreholes and geophysical logs from this
project, and the more regional map presented by Drummond (1998). The subcropping zone was developed from the hydrostratigraphic
sections (Figures 6-10), and projected or inferred where core control is lacking. The trend of the subcropping zone roughly parallels
Morgan Creek to the north of the creek, and underlies the northern drainage divide for the Morgan Creek watershed. The tributaries
north of Morgan Creek all drain basins where the sedimentary redox zone at the top of the Aquia CL is within 10 or 15 m of the land
surface.
Ground-Water Flow Paths and Traveltimes
The water-level measurements of wells screened in the Columbia-Aquia surficial aquifer indicate that ground-water flow patterns
are similar to those reported by Hamilton and others (1993). Ground-water levels from May 1998 (Appendix C), when the water
table was at its maximum height during the project, were used to create a water-table map (Figure 12) and to represent flow paths
in vertical sections (Figures 13a and 13b). The basic flow patterns were similar throughout the 1997-99 period of record, including
a period of extreme drought during the summer of 1999.
In general, water is recharged in the uplands, flows through the surficial aquifer, and discharges to the perennial streams (Figures
12 and 13). There is some evidence, particularly on the northern boundary of the Chesterville Branch watershed (Figure 12), that
ground-water divides do not coincide with surface-water divides; however, the relatively poor resolution of topographic mapping
in the study area makes precise delineation of surface-water divides difficult. A water-table high in the northeastern part of the
Morgan Creek watershed is inferred from data presented by Hamilton and others (1993). However, the wells that showed this
mound subsequently were destroyed, and were not available for this investigation.
The recharge areas and flow patterns of ground water in the Hornerstown aquifer are difficult to interpret from the limited number
of available wells. Downdip flow is indicated from well KE Be 210 to KE Be 211 (at Chesterville Branch) (Figure 13), but there
is also the possibility of flow to the north (updip) from a divide south of well KE Be 200 toward the Sassafras River watershed. It
is unclear whether most of the recharge to the Hornerstown aquifer comes from the outcrop/subcrop zone north of the study area,
or whether an appreciable proportion of the recharge is through the Aquia confining layer.
Ground-water traveltimes, as determined from CFC ages (Appendix D; E. Busenberg, U.S. Geological Survey, written commun.,
1998) are similar to those reported in earlier work (Dunkle and others, 1993; Reilly and others, 1994; Bohlke and Denver, 1995),
3&0'00
EXPLANATION
(TIGHT PART) VERY LOW PERMEABILITY ZONE OFTHE
LOWER AQUIA CONFINING LAYER
j HEDOX ZONE OF THE UPPER AQUIA CONFINING LAYER
SURFACE-WATER DRAINAGE DIVIDE (BASIN BOUNDARY)
3500A SURFACE-WATER SAMPLING STATION AND PARTIAL
IDENTIFICATION NUMBER (Refer to figure 2.)
1 2 KILOMETERS
Figure 11. The extent of the subcropping Aquia confining layer and redox zone in the Locust Grove study area, Kent County,
Maryland, with subbasins for surface-water sampling sites.
23
-------
and generally confirm the relation between traveltimes and parts of the ground-water flow system. The youngest ground water is
less than 10 years old and is found in shallow wells in recharge areas (for example, wells KE Be 192 and 194; Appendix D). Ground
water collected deeper in the surficial aquifer, such as from well KE Be 195, is on the order of 20 to 30 years old. Ground water
from the Hornerstown aquifer in the study area had not been sampled for geochemistry or age dating prior to this project.
Hornerstown water is consistently greater than 40 years old in all four wells sampled (Table 3), with progressively older CFC ages
downdip, from 42 years in well KE Be 200 to greater than 50 years in well KE Be 210. In addition, local trends in the
concentrations of 4He (helium-4) indicate that the downdip Hornerstown water may be 75 years old or older (E. Busenberg, U.S.
Geological Survey, written commun., 2000).
The ground water in the surficial aquifer that discharges to Morgan Creek follows flow paths very different from ground water that
discharges to Chesterville Branch because of the aquifer geometry. This geometry also results in generally longer ground-water
traveltimes for the Chesterville Branch watershed. Ground water discharging to Chesterville Branch behaves as if in a simple,
moderately deep, unconfined aquifer with an impermeable horizontal base. The details of flow may be slightly! different from the
idealized conceptualization because of the southward dip (approximately 5 m/km) of the base of the Aquia aquifer. In general, the
flow paths that contact the reducing sediments at the base of the aquifer represent a relatively small fraction of the total discharge
to Chesterville Branch. Discharge to Morgan Creek is more complex. The Columbia-Aquia surficial aquifer is relatively thick only
southeast of Morgan Creek. The incised valley of Morgan Creek has cut down to, and probably into, the reduced sediments of the
Aquia confining layer (Figure 13b). Much of the ground water moving to Morgan Creek probably flows updip and contacts the
reducing sediments. This process was evaluated by Hantush and Cruz (1999) using numerical simulations based on the data
presented here. They found that the data were consistent with the concept that a large percentage of discharge to Morgan Creek
flows through the reducing layer at the base of the aquifer.
In addition to ground-water discharge from the lower part of the Aquia aquifer, Morgan Creek also may receive considerable
discharge from the Hornerstown aquifer. If a fault is present beneath Morgan Creek, it may allow appreciable ground-water flow
from the Hornerstown aquifer through a breached Aquia confining layer. The entire confining layer in this area is 9 to 10 m thick
and the lowest-permeability part of the confining layer is 2 to 3 m thick. A 5-m offset caused by a fault could breach the low-
permeability part of the confining layer to create a lateral conduit that would allow discharge of Hornerstown water directly to
Morgan Creek. However, further study is needed to verify the presence of the fault, and to document ground-water flow through
the confining layer.
3&OW
EXPLANATION
—-M— — WATER-TABLE CONTOUR, IN METERS ABOVE SEA LEVEL
(Dashed where approximately located. Contour Interval 2 meters.)
18.4J WEa USED FOR WATEP-LEVEL MEASUREMENTS
(Number Is altitude of water table, In meters above
sea level.)
SURFACE-WATER DRAINAGE DIVIDE (BASIN BOUNDARY)
GENERALIZED GROUND-WATER-FLOW DIRECTION
OfliOO
2 KILOMETERS
Figure 12. Configuration of the water table and generalized ground-water flow, Locust Grove study area, Kent County,
Maryland, May 1998.
24
-------
HYDROGEOLOGIC SECTION D-D1
(along Locust Grove Road)
— * 18.4*18-5* I 4
--20
Vertical scale greatly
exaggerated
DISTANCE, IN KILOMETERS
EXPLANATION
©
? ? HYDROSTRATIGRAPHIC CONTACT (Dashed and
queried where uncertain)
- -1—>- GENERALIZED GROUND-WATER FLOW PATH
(Dashed where inferred; queried where uncertain)
-• HIGH WATER TABLE, MAY 1998
LOW WATER TABLE, AUGUST 1999
18.5*
GROUND-WATER FLOW OUT OF THE
SECTION TO SOUTHWEST
WELL, WITH SCREENED INTERVAL
(Number is head representing the screened
interval, in meters above sea level.)
CONFINING LAYER
AQUIFER
Figure 13a. Hydrogeologic section D-D' (along Locust Grove Road) showing ground-water flow paths in the shallow aquifer
system, May 1998, Locust Grove study area, Kent County, Maryland.
25
-------
HYDROGEOLOGIC SECTION C-C1
(updip part of section, along Route 298)
METERS
30-i
20-
10-
Sea J_
Level
-10-
Vertical scale greatly
exaggerated
DISTANCE, IN KILOMETERS
EXPLANATION
? ? HYDROSTRATIGRAPHIC CONTACT
(Dashed and queried where uncertain)
>~ GENERALIZED GROUND-WATER FLOW
PATH (Dashed where inferred)
HIGH WATER TABLE, MAY 1998
• LOW WATER TABLE, AUGUST 1999
18.57*
WELL, WITH SCREENED INTERVAL
(Number is head representing the screened
interval, in meters above sea level.)
| | AQUIFER
t'iJeai CONFINING LAYER
LOW PERMEABILITY ZONE OF :
CONFINING LAYER
Figure 13b. Hydrogeologic section C-C' (along Harmony Woods Road, State Highway 298) showing ground--jvater flow paths
in the shallow aquifer system, May 1998, Locust Grove study area, Kent County, Maryland.
26
-------
Ground-Water Geochemistry
The ground-water geochemistry of the study area has been described in some detail by Hamilton and others (1993) and Bohlke and
Denver (1995). The shallow aquifers contain three major water types: "young" (post-1970 recharge) agricultural ground water;
"old" (pre-1970 recharge) agricultural ground water; and "old" calcareous water. The "agricultural" ground water has concentra-
tions of dissolved nitrate that are considerably elevated above natural background levels; "old" agricultural water has nitrate
concentrations of 2-3 mg/L as N and "young" agricultural water has nitrate concentrations greater than 3 mg/L as N. Nitrate
concentrations in young water can exceed 20 mg/L as N. Nitrate concentrations in the "calcareous" waters are near or below the
analytical detection limit (0.05 mg/L). Bohlke and Denver (1995) also reported that the relative concentrations of magnesium
generally covary with nitrate concentrations in ground water. This covariance results from the application of dolomitic limestone
(calcium-magnesium carbonate) as a soil amendment, which is apparently applied in proportion to the amount of nitrogen fertilizer
used.
Agricultural waters have dissolved oxygen concentrations exceeding 5 mg/L, whereas the "calcareous" waters have dissolved
oxygen concentrations close to zero mg/L. Low, but measurable, concentrations of dissolved oxygen were reported in a number of
wells sampled in the study area. These low concentrations were interpreted by Bohlke and Denver (1995) as representing mixtures
of oxygenated agricultural water and anoxic calcareous water. For the hydrogeochemical interpretations, these mixtures were
classified with the anoxic waters as "suboxic" (Bohlke and Denver, 1995).
Young agricultural waters were a calcium-magnesium nitrate-chloride type (Figure 14), reflecting the application of nitrogen and
potash (KC1) fertilizers and of dolomitic limestone for pH adjustment of soil (Hamilton and others, 1993). The older agricultural
CALCIUM SULFATE
PERCENTAGE OF TOTAL MILLIEQUIVALENTS PER LITER
EXPLANATION
O WELL SCREENED IN THE HORNERSTOWN AQUIFER
• WELL SCREENED IN THE SURHCIAL AQUIFER
Figure 14. Chemistry of water from ground-water samples collected in July 1998 at the Locust Grove study area, Kent County,
Maryland.
27
-------
water was relatively enriched in sodium (Figure 14), probably because of water-rock interactions (Bohlke and Denver, 1995). The
calcareous water was found previously in discharge areas at various depths above the confining layer. The water is of a reduced
(suboxic) calcium-bicarbonate type (Figure 14), with bicarbonate concentrations roughly an order of magnitude higher than those
in the oxic agricultural waters because of calcium carbonate dissolution.
Ground-water sampling conducted during this project.generally confirmed the conclusions of Bohlke and Denver (1995), including
the relations between recharge dates and nitrate concentrations (Appendix D). In addition, samples were collected from four new
wells in the Homerstown aquifer, and a series of samples were collected from a multiport well near the bottom of the Aquia
Formation in the sediment redox transition. Ground water sampled from the multiport well had relatively low pH values, with
relatively low concentrations of dissolved oxygen, and relatively high concentrations of nitrate and sulfate (Appendix D).
Dissolved-gas analyses indicate that the multiport samples did not have substantial amounts of excess nitrogen gas from
denitrification (Figure 15). Thus, the multiport samples appear to represent a water type with agricultural contaminants and is partly
reduced, but has not been denitrified and has not encountered calcium carbonate. Waters from the confined Homerstcwn aquifer
were an anoxic, calcium-bicarbonate type similar to the old calcareous waters identified previously above the 'confining layer in
discharge areas in stream valleys. The Homerstown waters were relatively old (some more than 50 years old) and had relatively
high silica and sulfate concentrations and minor amounts of excess nitrogen gas, as did the discharging waters.!
Dissolved Nitrate and Denitrification
In general, the highest concentrations of dissolved nitrate were found in the recharge areas of the surficial aquifer, whereas waters
with low concentrations of nitrate were found in discharge areas and commonly exhibited evidence of denitrification (Bohlke and
Denver, 1995). Stream base flow, which is composed primarily of ground-water discharge, had nitrate concentrations in Morgan
Creek (2.3 - 4.1 mg/L as N) that were appreciably lower than concentrations in Chesterville Branch (4.0 - 8.4 mg/L as N). Morgan
Creek appears to receive a greater proportion of denitrified ground water than Chesterville Branch. ;
All water samples from the Homerstown aquifer contained excess dissolved nitrogen gas above saturated equilibrium, as
normalized to argon concentrations (Figure 15). The presence of excess nitrogen indicates that the water that recharged the
Homerstown aquifer contained measurable dissolved nitrate, probably on the order of 1-2 mg/L as N, which had been denitrified
and transformed to dissolved nitrogen gas. Isotopic analyses of the nitrogen gas indicate that the Homerstown waters may have had
nitrate with nitrogen isotope d15N values between 2 and 5 %> (per mil, or parts per thousand), similar to the values obtained by
Bdhlke and Denver (1995) for ground water discharging in the stream valleys. The amount of excess nitrogen was slightly less in
the southernmost and deepest Homerstown well (KE Be 211, farthest downgradient) than in the three shallower, upgradient
Homerstown wells (KE Be 189, 200, and 210). All of the Homerstown samples had low concentrations of CFCs and tritium,
• JULY 1998 SAMPLES
O DECEMBER 1998 SAMPLES
400
450
500
600
650
700
750
800
850
CONCENTRATION OF DISSOLVED NITROGEN GAS,
IN MICROMOLES PER LITER
Figure 15. Distribution of dissolved argon and dissolved nitrogen gas with relation to solubility in air and enrichment of
nitrogen by denitrification, Locust Grove study area, Kent County, Maryland. !
28
-------
indicating that they were recharged before the mid- 1900s. According to E. Busenberg (U.S. Geological Survey, written commun.,
2000), the ground water in well KE Be 21 1 had a higher concentration of radiogenic helium than the other samples, indicating that
well KE Be 211 probably contained the oldest ground water sampled, with an apparent age of about 75 years (although the
uncertainty of this age estimate is not well known). These data indicate that the ground water in the Hornerstown aquifer was
recharged within the last 100 years, and that the recharge area probably is in the northern part of the study area. Although the
recharge areas are not the same for all samples represented, the overall pattern of increasing nitrate concentrations over the last
75 years, with the most rapid increase between about 1960 and 1980, appears to be a general trend for the region.
Bohlke and Denver (1995) suggested three possible sources of electron acceptors for denitrification of ground water in the study
area:
• Oxidation of glauconite;
• Oxidation of iron-sulfide minerals; and/or
• Oxidation of organic matter.
The cores collected for this project show that the sediments in the upper sandy section of the Aquia Formation are extensively
oxidized, but there is an abrupt transition to sediments with predominately reduced mineral phases at the top of the Aquia confining
layer. Silica concentrations on the order of 20 mg/L in suboxic Aquia waters and 30 mg/L in Hornerstown waters are elevated
relative to concentrations of 1 0 mg/L that are typical of oxic Columbia-Aquia waters. The higher silica concentrations in the deeper
units are consistent with longer residence times and the dissolution of glauconite to release silica. Sulfate concentrations of oxic
Columbia-Aquia waters generally are below reporting levels (0.10 mg/L), whereas concentrations in suboxic Aquia and
Hornerstown waters generally range from 5 to 20 mg/L. Sulfate is a common product of the dissolution and oxidation of iron-
sulfide minerals, such as pyrite or marcasite. The data presented here provide no new insights into the relative importance of
organic matter oxidation. From mass-balance modeling and carbon stable-isotope data, Bohlke and Denver (1995) inferred that the
oxidation of organic matter was not the major mechanism for denitrification of the discharging ground water.
The samples collected from the multiport well (KE Bel 90) and the nearby wells that are screened above and within the Aquia redox
transition zone provided inconclusive evidence on the relative importance of glauconite or iron-sulfide oxidation. Concentrations
of dissolved oxygen decreased in the ports located within the redox transition zone, but none were lower than 2 mg/L, indicating
oxic conditions or mixtures of oxic and anoxic water; this result also was true of the data from the cluster of individual wells. These
samples also contained nitrate concentrations of 5 to 6 mg/L as N from the sampling ports of well KE Be 190 and concentrations
of 1 1 to 12 mg/L as N in the well cluster. Sulfate concentrations from the ports of well KE Be 190 ranged from 15 to 23 mg/L,
similar to those of old anoxic water. The geochemical data for ground water from the lower Aquia aquifer could indicate that
sulfate was produced by iron-sulfide oxidation coupled with reduction of dissolved oxygen, which would not consume nitrate
(because the thermodynamic sequence of electron-accepting processes generally consumes dissolved oxygen before nitrate). If
iron-sulfide oxidation were driven mainly by dissolved oxygen, then the lower nitrate concentrations could represent older, pre-
1970 recharge, when nitrate concentrations were lower than in waters currently recharging at the water table.
Oxidation of pyrite by reduction of dissolved oxygen can be represented by:
4FeS2 + 1502 + 10H2O « 8SO/ + 4FeOOH + 16H+
(1)
This reaction requires about 1.9 moles of oxygen to yield 1 mole of sulfate. In contrast, oxidation of pyrite by denitrification can
be represented by:
2FeS
6NO3- + 2H2O
4SO4'2 + 2FeOOH + 3N2 + 2H+
(2)
In equation 2, 1.5 moles of nitrate are required to yield an additional mole of sulfate. Water recharging the surficial aquifer contains
on the order of 10 mg/L (~0.3 mmol/L, or millimoles per liter) dissolved oxygen, but variable concentrations of nitrate, ranging
from 2 mg/L (0.15 mmol/L) as N to over 1 0 mg/L (0.7 1 mmol/L) as N. In a ground-water chemical system not limited by the supply
of iron sulfide, consumption of dissolved oxygen should result in a sulfate concentration of 15 mg/L (0.16 mmol/L). Denitrification
of ground water with a nitrate concentration of 2 mg/L as N would result in an additional 9.6 mg/L (0.1 mmol/L) of sulfate.
Denitrification of ground water with a nitrate concentration of 10 mg/L as N would result in ah additional sulfate concentration of
45 mg/L (0.47 mmol/L). Thus, sulfate concentrations above 15 mg/L could be interpreted as evidence that denitrification has
occurred, or that the ground water contained some sulfate when recharged. The latter appears to be the case for the multiport
samples, which did not contain excess nitrogen gas.
In summary, these results indicate that the samples collected from the multiport well did not contain denitrified ground water,
although several ports were deep enough to be within what appeared to be reduced glauconitic sediments in nearby cores. If the
samples from these sites did indeed represent ground water from within the reduced zone at the top of the confining layer, then these
observations could indicate that denitrification did not have a large effect on ground-water chemistry in the Aquia aquifer flow
system. It is possible that suboxic ground waters carrying nitrate moved along the redox boundary toward discharge sites without
losing measurable amounts of nitrate. Alternatively, reduced ground waters may be present but could not be sampled because the
construction of the multiport wells and piezometers permitted overlying oxic water to enter the screens (by downward flow in the
29
-------
annular space around the well casing). This flow might have occurred because hydraulic conductivities within and below the
sediment redox transition at the top of the Aquia confining unit are much lower than that of the overlying sand of the Aquia aquifer.
Whether or not the persistence of nitrate in the multiport samples is an artifact of the well design, these results do not provide strong
support for appreciable denitrification above the confining layer.
Many of the chemical and chronological characteristics of the Homerstown waters appear to match those ,of ground waters
discharging beneath the streams. Ground water from the four new wells screened in the Homerstown confined aquifer was reduced,
alkaline, and nitrate-free. Ground water from all four contained minor amounts of excess nitrogen gas with an isotopic composition
(dlsN [of excess NJ = 2-5 %o; J. Bohlke, U.S. Geological Survey, written commun., 2000) slightly different from that of
atmospheric nitrogen, which indicates that the Homerstown waters were recharged with minor amounts of nitrate that was
denitrified subsequently. Bohlke and Denver (1995) identified tritium-poor ground water discharging to both streams, and
Homerstown waters all appear to be older than mid-1900s. The isotopic composition of the Homerstown N2 gas is consistent with
denitrification of nitrate similar to that of older ground water above the confining layer (data presented by Bohlke and Denver
[1995] in their figures 12 and 13); further, the apparent initial nitrate concentrations of the Homerstown ground waters are
consistent with those of the discharging ground waters. The high alkalinities, moderate sulfate concentrations, and other chemical
characteristics of the Homerstown waters match those of ground waters discharging at stream-sampling sites KE Be 172,173, and
178. Water of this type has not yet been identified within the surficial aquifer, except near discharge zones (for example, well KE
Be 159 in the Chesterville Branch valley) with upward flow from deep in the surficial aquifer. These observations could be readily
explained if the Homerstown ground-water flow system were connected with the overlying Columbia-Aquia aquifer, as suggested
by Bohlke and Denver (1995), but they are difficult to explain if the two flow systems are separated completely.
The similarity of nitrate concentrations, dissolved nitrogen and argon gases, ground-water age-dating estimates, and other chemical
signatures such as alkalinity and silica concentrations between discharging ground water in Morgan Creek as reported by Bohlke
and Denver (1995) and the Homerstown ground water reported here, could indicate that discharge from the Homerstown aquifer
may influence much of the chemical composition of ground-water discharge to Morgan Creek.
30
-------
Nutrient Yields from Stream Base Flow
The chemistry of stream base flow relates closely to the chemistry of ground water discharging to the stream. The spatial patterns
of base-flow chemistry within each watershed provide insight into the chemical environment of the surficial aquifer, allowing for
instream processes. The current project included three synoptic surveys of stream base flow at 14 sites in the Morgan Creek and
Chesterville Branch watersheds. These synoptic surveys sampled conditions during periods of low water table (September 1998)
and high water table (April 1998 and February 2000). The synoptic surveys were supplemented by monthly base-flow samples at
the Morgan Creek and Chesterville Branch gaging stations to provide a record of seasonal variability. These data were evaluated
to relate the loads and yields of nutrients from the two watersheds to landscape factors and geologic setting.
Sampling at Fixed Sites
Time-series plots of samples collected from the fixed sites (Figure 16) show seasonal variations in dissolved nitrogen concentra-
tions. Concentrations tend to be lowest during the summer months. Low stream discharge in the summer is common in this area
because evapotranspiration removes water that would otherwise recharge the aquifer. The lower nitrogen concentrations are likely
due to uptake by stream phytoplankton, macrophytes, and submerged aquatic vegetation during the warmer growing season. The
data also show that Morgan Creek has lower nitrogen concentrations than Chesterville Branch in the upstream parts of the
watershed, as previously reported by Bohlke and Denver (1995).
STATION 01493112 (Chesterville Branch near Crumpton, Maryland)
•* STATION 10493500 (Morgan Creek near Kennedyville, Maryland)
Jan 'Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar
1998 I 1999 I 2000
Figure 16. Time-series plots of total dissolved nitrogen concentrations at the two fixed stream-sampling sites (station numbers
01493112 and 01493500) in the Locust Grove study area, Kent County, Maryland.
31
-------
Although there is some overlap in the range of values, the nitrogen loads (mass per year) from Chesterville Branch generally are
higher than those from Morgan Creek (Table 4). Nitrogen yields (load per unit watershed area) are higher from Chesterville Branch
than from Morgan Creek. Chesterville Branch yields of 0.04-0.08 gm sec'1 km'2 are higher than those reported by Bachman and
Phillips (1996) for well-drained landscapes on the Delmarva Peninsula, which ranged from 0.02 to 0.04 gm sec'1 km"2. Morgan
Creek yields (0.01-0.03 gm sec'1 km'2) overlapped those of Chesterville Branch, and were slightly higher than yields reported for
poorly drained landscapes on the Delmarva Peninsula (0.007-0.02 gm sec'1 knr2) (Bachman and Phillips, 1996). Both of these
watersheds were considered to be in the well-drained landscape in the analysis by Bachman and Phillips (1996); the results
presented here further underscore the extent of variation in nitrogen yields within a given landscape class reported in that study.
Base-Flow Synoptic Surveys •
The data from the base-flow synoptic surveys also confirm observations by Bohlke and Denver (1995) that total nitrogen
concentrations in Morgan Creek generally are lower than those in Chesterville Branch (Figure 17). The synoptic surveys revealed
some fundamental differences in the patterns of occurrence of nitrogen and related chemical constituents between the two
watersheds. In particular, total nitrogen concentrations tended to decrease downstream in Chesterville Branch, whereas they
increased downstream in Morgan Creek (Figure 18). In both watersheds, stream discharge (of water) and nitrogen loads increase
monotonically downstream. The increasing trend of stream discharge indicates that both streams are gaining flow from ground-
water discharge. Although loads in the two streams appear similar in the upstream reaches, they are consistently higher in
Chesterville Branch than in Morgan Creek at the downstream gage (Figure 18), even though Morgan Creek has consistently higher
discharge than Chesterville Branch for any given season (Figure 18).
The higher stream discharge in Morgan Creek may be due to the larger watershed area of Morgan Creek (31 km2) as compared to
Chesterville Branch (16 km2). The specific discharge (discharge per unit area of the watershed) for any given season is greater in
Chesterville Branch than in Morgan Creek (Figure 18). This pattern also occurs in the downstream trends of nitrogen yields
(Figure 18). Yields in Morgan Creek increase downstream, similar to nitrogen concentrations (Figure 18), whereas Chesterville
Branch displays a more variable trend similar to that of specific discharge. The higher yields in Chesterville Branch also appear to
be related to the nitrogen concentrations.
Bohlke and Denver (1995) presented evidence that the lower concentrations of nitrate in Morgan Creek were due to denitrification
in the discharge areas where the Aquia confining layer is shallow. However, it also is possible that the lower concentrations could
be due to the mixing of higher proportions of older ground water with low nitrate concentrations in base flow. These two
Table 4. Summary of Base-flow Nitrogen Loads from the Locust Grove Study Area, Kent County, Maryland
Chesterville Branch
near Crumpton, Md.
(station 01493112)
Morgan Creek near
Kennedyville, Md.
(station 01493500)
Instantaneous load,
in gm N sec"1
Instantaneous yield,
in sm N sec" km
Annualized load,
in kg N year"
minimum
25m percentile
median
75th percentile
maximum
number of samples
25m percentile
median
75to percentile
25th percentile
median
75tt percentile
0.27
0.71
0.92
1.2
2.3
33
0.045
0.058
0.076
22,000
29,000
39,000
0.36
0.47 ;
0.62
1.0
1.2 ;
11
0.015
0.022
0.032
. 15,000
20,000
32,000
[Estimates are based on monthly samples collected from April 1998 to March 1999 and one sample collected in February 2000. gm N sec-1,
grams of nitrogen per second; gm N secr'krrr2, grams of nitrogen per second per square kilometer; kg N year1, kilograms of,nitrogen per year.]
32
-------
3^15W
2 KILOMETERS
EXPLANATION
TOML DISSOLVED NITROGEN CONCENTRATION,
in milligrams per liter as nitrogen
• GREATER THAN 8.3
O 5.9-8.3
• 3.25-5.8
• LESS THAN 3.25
SUBBASIN CONTRIBUTING TO EACH
STREAM-BASE-FLOW SITE
Figure 17. Areal distribution of total dissolved nitrogen concentrations in stream base flow for the Locust Grove study area,
Kent County, Maryland, April 1998 synoptic survey.
33
-------
o
£5
o
z
0.30
0.25
0.20
0.15
0.10
0.05
0.00
Flow direction
ilisiilS ...... ,:i»fi.ii ..... I
1 , ,J, i ..... !ii'» ;„„!' ;!'i!," •. •:»:„;;!:!, '.; ...... i - :S^i^fS;s::7~:—~'s:^,s^ --J\ - ~
-
" '
rpTsw.rfjwntTrwrflr :t;:
CC
0£b
I"* rn —J
o cc
E LU
j- Q.
183
g£2!
1°^
12
10
8
6
4
2
0
0.015
Q
nfO UJ
s
O°-
O)
S~2
O
ffi § 0.010
0.005
0.000
DISTANCE UPSTREAM OF GAGE,
IN KILOMETERS
|V|:;: .^^ ,,_',, j.^,.^. ^^.-,
DISTANCE UPSTREAM OF GAGE,
IN KILOMETERS
Si
Oo
O LU
CC W
nco
Si
DISTANCE UPSTREAM OF GAGE,
IN KILOMETERS
"*'"'""'1'''''r' ^'' -
0.0
DISTANCE UPSTREAM OF GAGE,
IN KILOMETERS
EXPLANATION
BASE-FLOW SYNOPTIC SURVEYS
» » MORGAN CREEK, APRIL 1998
*•---» MORGAN CREEK, SEPTEMBER 1998
O O CHESTERVILLE BRANCH, APRIL 1998
O O CHESTERVILLE BRANCH, SEPTEMBER 1998
D
111
0.15
0.10
0.05
0.00 1
DISTANCE UPSTREAM OF GAGE,
IN KILOMETERS
Figure 18. Observed downstream trends of stream discharge, specific discharge, total dissolved nitrogen concentration,
instantaneous nitrogen load, and instantaneous nitrogen yield from base-flow synoptic surveys in April and
September 1998, Locust Grove study area, Kent County, Maryland.
34
-------
hypotheses can be evaluated from the distribution of selected geochemical constituents in base flow from the two basins. As
described above, ground water recharged before 1970 has lower nitrate concentrations than water recharged more recently. In
addition, old agricultural and old calcareous waters have had more contact time with the aquifer materials, and may be more
chemically evolved with higher concentrations of dissolved bicarbonate and silica than the young agricultural water. Further, the
work of Modica and others (1 998) suggests that the proportion of older ground water discharging to a stream increases downstream
(Figure 4). In addition to lowering nitrogen concentrations, a larger proportion of older ground water in discharge to base flow also
would have the cumulative effect of an increasing downstream trend of bicarbonate and silica, and a decreasing trend of magnesium
(Figure 19).
The observed trends of bicarbonate, silica, and magnesium in Morgan Creek and Chesterville Branch were compared to the trends
expected from a systematically increasing proportion of discharge from old ground water (Figure 20). The observed data are
consistent with the expected trend for magnesium in all cases, and for bicarbonate in Chesterville Branch. However, bicarbonate
shows a decreasing trend in Morgan Creek (especially for the April 1998 samples), and both watersheds have decreasing
downstream trends in silica concentrations. Decreasing silica concentrations would be expected in Chesterville Branch, because
discharge of high-silica water from the Hornerstown aquifer is less likely than in the upstream reaches of Morgan Creek. The top
of the Hornerstown Formation dips to the south, and Chesterville Branch has not incised a stream channel through the Aquia
confining layer. This may not be the case in Morgan Creek, however, where the channel appears to follow the outcrop zone of the
Aquia confining layer (Figure 1 1), and there is some evidence of ground-water discharge from the Hornerstown aquifer (Bohlke
and Denver, 1995).
Ifon-sulfide oxidation to produce sulfate is one possible redox process associated with denitrification. If denitrification were related
to iron-sulfide oxidation, then denitrified water would have sulfate concentrations elevated above background. However, the trends
of sulfate concentration shown in Figure 20 indicate that downstream water, which is hypothesized to have a greater proportion of
denitrified ground-water discharge, has lower concentrations of sulfate than upstream water.
The concentrations of dissolved constituents in base flow are not merely the aggregation of the concentrations of ground-water
discharge to the mainstem channel of the stream. Concentrations in base flow also are affected by the loads of constituents entering
from tributaries and processes occurring in the stream channel or alluvial sediments that may be sinks and sources for the dissolved
substances. Although a thorough discussion of riparian and instream processes is beyond the scope of this report, an analysis of the
effects of the mixing of water from tributaries provides some evidence that processes other than ground-water discharge are
affecting base-flow concentrations.
In a typical stream, the flow from two tributaries mixes in a downstream channel (Figure 21). Without the physical or geochemical
removal of a constituent, the downstream load, C3Q3, is equal to the sum of the loads of the two upstream tributaries (CjQ, andC2Q2)
plus the load derived from ground-water discharge CGWQGW:
QGW = C3Q3 (3)
where C is concentration (mass per volume), Q is discharge (volume per time), and Gw refers to ground water.
Headwaters
Gage
CO
HI
Headwaters
Gage
DC
<
O
m
Headwaters Gage Headwaters Gage
Figure 19. Expected downstream trends of nitrogen and related geochemical constituents from a downstream increase in the
discharge of old ground water.
35
-------
0.3
0.2
ao
DISTANCE UPSTREAM OF GAGE,
IN KILOMETERS
EC
u-t
Of
DISTANCE UPSTREAM OF GAGE,
IN KILOMETERS
i 3 0.12
EXPLANATION
BASE-FLOW SYNOPTIC SURVEYS
« » MORGAN CREEK, ^PRIL 1998
*• •* MORGAN CREEK, iSEPTEMBER 1998
o o CHESTERVILLE BRANCH, APRIL 1998
o -O CHESTERVILLE BRANCH, SEPTEMBER 1998
DISTANCE UPSTREAM OF GAGE.
IN KILOMETERS
DISTANCE UPSTREAM OF GAGE,
IN KILOMETERS
Figure 20. Observed downstream trends of sulfate, magnesium, bicarbonate, and silica concentrations from base-flow synoptic
surveys in April and September 1998, Locust Grove study area, Kent County, Maryland. j
Figure 21. Relations among discharge and concentrations of dissolved constituents in tributaries and downstream sites under
non-reactive mixing. !
36
-------
An estimate of ground-water inflow can be made by subtracting the upstream discharges (Q, and Q2) from the downstream
discharge (Q3):
Qow = Q3-(Q, + Q2) (4)
The expected average concentration of ground-water discharge between the tributaries and the downstream site can be estimated by
dividing equation (3) by the estimate of ground-water inflow and rearranging the terms:
= C3(Q3/QGw) - C,(Q/QGw) ~ C2(Q/QOW)
(5)
A negative value of CGW for cases in which Q3 is greater than the sum Q,+Q2 (for example, an effluent stream system with C3
considerably less than C, or C2) is evidence of a physical or geochemical sink for the constituent. In such cases, the concentration
of ground-water inflow cannot be estimated. In other cases, further evidence is needed to determine whether instream or alluvial-
channel processes are acting as a geochemical source or sink for the constituent.
An estimate of the downstream concentration C'3 can be made on the basis of any hypothesized average concentration of ground-
water discharge, and this estimate can be compared to the actual measured downstream concentration. Rearranging equation 5
results in:
C'3 = C,(Q/Q3) + C2(Q/Q3) + CGW(QOW/Q3)
(6)
If the estimated value for C'3 is close to the measured value, it is likely that ground-water discharge with non-reactive mixing
accounts for most of the observed concentrations of dissolved constituents in base flow.
In the study area, these relations can be calculated at four reaches that have sites with upstream measurements, downstream
measurements, and measurements at tributaries:
Chesterville Branch
Reach Cl:
Reach C2:
Upstream
site
01493110
01493111
Downstream
site
01493111
01493112
Tributary
site(s)
01493109
0149311110
Morgan Creek
Reach Ml:
Reach M2:
01493495
01493497
01493497
01493499
01493496
01493498,
0149349810,
0149349820
The locations of these sites and the stream reaches Cl, C2, Ml, and M2 are shown in Figure 22.
TftSW
EXPLANATION
. SURFACE-WATER SAMPLING STATION AND MHTIAL.
IDENTIFICATION NUMBER [TIM comptau ktantiflcfttfon
number Hated In H» Nation] W«t« Intofnwtfcn Syatwn
(NWIS) lingua and In tha AppMvlbcu contains the
prefix -OUT.]
2 KILOMETERS
Figure 22. Location of surface-water sampling stations and stream reaches on Morgan Creek and Chesterville Branch in the
Locust Grove study area, Kent County, Maryland.
37
-------
Estimates from equations (5) and (6) of ground-water discharge, concentrations of selected constituents in ground water, and the
downstream concentration in the reach under the bounding condition of ground-water concentrations equal to zero are presented in
Tables 5a through 5d. A comparison of the estimated and observed constituents shows that they only are partially consistent with
the hypothesis of increased downstream contribution of older ground water. However, a number of interesting'trends are evident
that could justify further evaluation of the data and additional study of the processes of ground-water discharge to streams in the
study area. •
The most striking observation in Table 5a is that the decrease in nitrate (from greater than 10 mg/L down to 7.84 mg/L) observed
in Chesterville Branch reach Cl cannot be accounted for by inflow of lower-nitrate ground water. Even where the ground-water
discharge is nitrogen-free (the case in which Cow in equation 6 equals 0), the estimated downstream total-nitrogen concentration is
8.9 mg/L, higher than what is observed. Reach Cl of Chesterville Branch has a manmade pond that is approximately 10,000 m2
(about 2.5 acres). It is possible that this pond provides a geochemical environment that facilitates denitrification; alternatively, the
dissolved nitrogen may be taken up by plants. ;
^^ ' i
The data in Table 5b provide evidence that inflow of low-nitrogen ground water contributes to the downstream decrease of nitrogen
concentrations in Chesterville Branch. In reach C2, the observed decline in total nitrogen concentrations (from 7.8 to 7.6 mg/L) can
be explained by the dilution of ground-water inflow with an average concentration of 8.1 mg/L (which can be obtained from a
mixture of old and young ground water) and by inflow from the tributary (site 0149311110) with a lower nitrate concentration
(5.8 mg/L). Bicarbonate concentrations in both reaches increase to a degree that can be explained by the input of high (greater than
0.5 meq/L) bicarbonate water typical of old calcareous ground water. ,
In Morgan Creek, the higher total nitrogen concentrations in the tributaries may explain in part why concentrations increase
downstream. In reach Ml (Figure 22), the average ground-water nitrogen concentration was calculated as 2.8 mg/L (Table 5c), this
mixture would have a considerable proportion of old calcareous water mixed with the young, high-nitrogen water. Discharge of
nitrogen-free ground water would result in a downstream concentration of 2.2 mg/L, below the observed value of 3.25 mg/L. In
addition, one tributary (site 01493496) with a higher nitrogen concentration (4.75 mg/L) may contribute to the observed
downstream increase in nitrogen concentration; this tributary watershed is underlain by a thicker sequence of the Columbia-Aquia
surficial aquifer rather than the shallow Aquia confining layer. Estimated ground-water bicarbonate inputs are consistent with a
source of old calcareous water, but estimated silica concentrations appear to be lower than those observed in ground water. •
Estimated sulfate concentrations appear to be more typical of those observed in the Aquia redox transition zone than in the
Hornerstown aquifer. j
In Morgan Creek reach M2 (Figure 22), the calculated ground-water nitrogen input is 6.3 mg/L (Table 5d); apparently the discharge
of the high-nitrogen tributaries (sites 01493498, 0149349810, and 0149349820) is not sufficient to account for the observed
increase in nitrogen concentration (from 3.3 to 3.9 mg/L) in this reach. The bicarbonate decrease observed in this reach may
indicate that the ground water being discharged is young, low-bicarbonate, high-nitrate water; however, the calculated ground-
water concentration is negative, indicating that riparian or instream processes may be removing bicarbonate.' Silica displays a
pattern similar to bicarbonate. The calculated sulfate concentration is not consistent with the calculated nitrogen concentration,
because it is more typical of Hornerstown water than water from the surficial aquifer.
The patterns observed in the base-flow synoptic surveys can be related to the hydrogeologic framework of the two watersheds.
Chesterville Branch reaches Cl and C2 are underlain by the thick Columbia-Aquia surficial aquifer. It is likely that only a small
percentage of the ground water discharging to Chesterville Branch flowed through the redox zone at the base of the Aquia aquifer
or from the Hornerstown aquifer. This watershed functions in a manner similar to those described by Modica and others (1998),
with a greater proportion of older ground water contributing to flow in the downstream reach. In addition, impoundments, both
natural (such as beaver dams) and artificial, may be sinks for nitrogen.
Morgan Creek reach Ml is underlain by a thin cover of the Columbia-Aquia aquifer, which makes it more likely that a greater
proportion of ground-water discharge flowed through the Aquia redox zone and was denitrified. At the downstream end, the stream
channel overlies the subcrop of the Aquia redox zone. The Aquia confining layer may be breached locally, either by faulting or
Stream incision, and water from the Hornerstown aquifer may discharge through the alluvial fill. Alternatively, the tributary in
reach Ml is underlain by a thicker section of the Columbia-Aquia surficial aquifer, and higher-nitrate ground water may discharge
as base flow. |
In Morgan Creek reach M2, the streambed and alluvial-channel are underlain along the entire length of the reach by the Aquia
confining layer and the sedimentary redox zone. Nitrogen concentrations increase downstream, which could be explained by an
increased discharge of relatively high-nitrogen (~6 mg/L) ground water. These observations indicate that either! the incised valley
in this reach does not penetrate the Aquia confining layer, or the alluvial fill is more fine-grained and Hornerstown water cannot
easily discharge to the stream. In addition, riparian and instream processes may be acting as sinks for bicarbonate and silica, or
sources for sulfate and other constituents. !
38
-------
i
§
'
§ £
u 2
-o O
3
o
on •-
li
§ 1
-OCQ
II
U S?
•gfi
22 c
Q §
00
3 d,
8 S
c '
o
s
. o
« .
c S"
O cd
•^ a>
S -b
C "
§ "8|;
o •§ S.
i| §
o o
u o
-c
a
a>
I
!» 13 -5 t»
§-=
O 0)
.§ a a o
353 J- aa
-
i n az
gs-Sa
E?
a
JS
2 M
5 e
O S3
*
e S
o ja
II
"0 O C4
CO OJ i—<
— "* fM
o o o
CJ
e
itimated coi
s
I*
=2
a!
egative valui
c
o
II
c/
g
CO
B
CO
•a
c
1
oo
'•g
1
•a
CO
C
t~^
'1 1
0
§
1
C
o
c
o
o
S£
,^
g
CO
1
^
_c
00
issumin
ra
§
concentrati
g
S
CO
mated down
£Q
I
81
?>
I
39
-------
I
-o a
's-o"
g o
a «
Jog
w -a
&
II
en a
o o o
•^f co •*
o c5 o
>o co co
o o o
O C3 ?3
o o o
CO CO
O\ ON
••
O O O
U
§
w
c/5
1
MATED
E
S
^_
C3
Total streamflow (measured)
i
i
i
1
i
o
o
1
"^
o
e
S
a
S
Cfl
3
'S
"S
S
u
Ground-water discharge (48 p
CO
o
CO
o
oo
9
<
~
0
oo
oo
1
1
^
Concentration in ground wate
>o
o
o
CN
O
s
o
B
o
VO
CO
1
1
O
II
£>
U
"5
Downstream concentration ^
1
i
03
S
lent in the
3
to
§
8
£
o
.2
CLc
S
3
to
C
O
o
o
S
a.
3
S
1
e*
•S
1
8
§
O
A
A negative value for estimatec
S
u
J:
o
GO
1
T3
3
tituent in
M
8
•S
'o
o
•s
S
G
o
C
8
1
*o
to
•o
s
-1-1
1
oo
c
assumi
g
•s
is
n
Estimated downstream concer
I
I
a
I
^>
40
-------
C
'S o
0 'S
•rt
03 ^
8 £
§<
1|
^ >
o g
s i
-b §
OO 0
2 a>
CO "*^
03 C
CD ""
lated from M
k, Reach Ml
3
5 £
0 ^
8) S)
tfl O
•ss
M £
5 o
•i g1
3 "a.
CJ rS
t-
§-2
• — £
c/3 .
ac c .
S3
-o p"
0} m
O -C
CO CO
co C
5 |
s s
.2 *
c -s
o
0 S
Estimated
Constituer
Maryland
.
V)
CU
1
H
•o
cu1' S
»C3 o
— g
OO T3
a"
.3 13
S £
£ -^
&c ®
« 6°
J-.B
Is-
5 t»
^^
^ o
5
CO
§
o
o o
OO ^
o o
vo oo
ex; r~
o o
>n o
0 '—
CD O
t-2 cQ
•^f CO
CO vo
0 ^->
O CD
e
>> §
s 1
J 1
•cl o
H Q
vo t—
5 5
CO CO
§ 2
o o
X
U
5
(^
s
m
fe
Pi
0
B.
Q
1
P
S
VO
T— I
CD
Total streamflow (measured)
i
i
i
i
i
\D
O
O
1
•?
O
S*
1
^M
-2
®
g
u
s«
s
1
"3
b
13
^
!
o
S
o
OS
CO
O
OS
oo
O
CO
o
CO
°9
1
1
c3
Concentration in ground wat
0
o
05
•"3"
o
s
CD"
\O
*~l
1
w*
S,
0
"2
"i
Downstream concentration
o"
II
B6
2
S
CO
v-
s
KJ
£
1
e
eb
_n
1
c
8
o>
4=
4-'
O
C
_o
c
8
"8
>
"3
CO
CO
1>
£
•s
^
OO
C
's
D
CO
CO
C5
_O
C5
• i
g
Estimated downstream conce
I
i
I
03
Q.
1
1
>a
£vi
41
-------
o o
So
1s2-
<1> O
pj ^C
6-s
•o S
C3 ^^
& >
c 2
§2
J3>J
4> O
z5 *j3
M C
S "•*
H
5 ,*^r
o J>
" w
03 "~
»> ^^
•f 60
C* T"
3 O.
S |
i 1
jj2 *M
o c:
,. .„. ^
g* C
^ ^
o£
vj tn
/^ £
2 o
O ti
(O «
S w
Estimated Conct
Constituents at I
Maryland
!:
1
H
«fl
»2 ®
to 'a
j-
5
*^ &l
|s
iS .a
IS -a
«r
i*
o /•?
S S
tn
jj
S*^
E 5
N^ *^
CS
iS
p vJ
^& ffi
ca
^ -
1 !2
•» CO
S tn
*i^ CS
M
es ,_>
•CS jn
.Sme
pj NM'
&o
.S
to
es ~
V3 V)
O a
o ja
1 i
or> B
en c^» •"•« en
— « cs cs
o c> c> o
g- o\ vo —
*^r co */% i""-
0 O O 0
CO Tj- VO CO
o o o o
i
go r~ o\ * vi oo
•^1
o
— O O 0
o o o o
V V
§ fc* & S
g W CS ^
t! 3 B fa
CO .U ^3 c/3
!§•££;§•
O 0
^- CN
r^ oo oo oo
O^ o\ ON O^
CO CO CO CO
ON ON ON ON
o o o o
o
,.,.
•^-
o
o
VO
o
o
o
t-
CO
en
t
cs
o
Downstream
1
o
1
a
0
< 1
§
oo '
O
1 '
^
|
CO
W
T-H CO
CN 0
O O
1
>
o
1
01
-&
GO
•S
3
*S
e
«
iflow (measured)
ter discharge (14 pe
2 f
* •§
1 1
f2 o
0
CO
en
o
• f~^
^
o
0
c>
en
vd
1
1
on in ground water
oncentrati
u
cs
o
CO
en
0
c~-
vq
O
o
o
CTv
CS
1
1
o
II
&
U
€
am concentration w
I
1
O
93
o
cd
a
: in the stream
B
.1
n
0
o
r£!
'o
r consumption
o
o
-S
g.
^
a.
.£
G
o
!
o
a
s
i value for estimated
1
§
«jj
0s
||
*>.
l>
y,
N
.OT
|
•O
_B
'B
.1
B
S
CD
•£
O
§
^ed concentrat
**•
"o
CO
CO
^o
$->
^
-r^l
C
's
g
CA
C3
C
o
1
downstream concent
w
•8
•Q
eo
•8
o
c
t
i
S
s£
CD
Q.
1
a
I
I
1
si
1
I
CO
SI
I
42
-------
The relative importance of ground-water discharge to these two streams compared to the effects of riparian and instream processes
is not well understood. Additional investigations, including coring the alluvial fill, sampling ground water in the riparian zone, and
conducting more intensive biological and hydrochemical sampling in the streams, are needed to improve the understanding of base-
flow nitrogen discharge in this area.
Relation Between Nutrients and Landscape Factors
Bachman (1984b), Hamilton and others (1993), Bachman and Phillips (1996), and Phillips and Bachman (1996) described the
variability in the landscape features on the Delmarva Peninsula and how this variability was related to the quality of ground water
and base flow. Hamilton and others (1993) developed the "hydrogeomorphic region" approach, whereby landscapes were defined
on the basis of the overall occurrence of certain mappable features, most notably the degree of stream incision, the drainage
characteristics of the soil, and the presence of shallow confining layers within the surficial aquifers. These features appear to be
related to the arrangement of ground-water flow paths and the presence of geochemical processes that can affect the chemical
composition of ground water (Bachman and Katz, 1986; Hamilton and others, 1993; and Bachman, 1994). Hydrogeomorphic
regions are independent of stream basins; thus the discharge from a basin could be a combination of several hydrogeomorphic
regions. Phillips and Bachman (1996) and Bachman and Phillips (1996) developed a "basin index" to characterize landscapes
within a drainage basin. The index is essentially the principal-components score using three basin characteristics (stream slope,
percentage of poorly drained soils, and forest cover in the basin) that appear to be related to the hydrogeomorphic regions.
The Locust Grove study area is in the well-drained upland hydrogeomorphic region defined by Hamilton and others (1993). This
area is characterized by a relatively thick (15-30 m) surficial aquifer, deeply incised stream valleys, and well-drained soils. Ground-
water flow paths may extend from drainage divides to discharge in streams and penetrate to the base of the surficial aquifer. The
land use in the study area is mostly agricultural, and there is little interspersion of agricultural and wooded areas. The uplands are
fanned, and the woodlands and wetlands are restricted almost exclusively to the riparian areas (forest land use in Figure 3). The
well-drained upland was reported to have ground water with high concentrations of nitrogen (Hamilton and others, 1993).
Although the study area is entirely within the well-drained upland, the dissolved nitrogen concentrations vary considerably between
Chesterville Branch and Morgan Creek (Bohlke and Denver, 1995). A local-scale landscape analysis was conducted to aid
modelers in locating subregions within the study area that have similar hydrologic conditions. The analysis was based on the "basin
index" approach of Phillips and Bachman (1996), and tested by relating the basin index to constituent concentrations of water
samples collected during the base-flow synoptic surveys.
The original basin index was developed from landscape data for basins located throughout the Delmarva Peninsula. The well-
drained upland in the Locust Grove study area represents one end-member of the landscapes used to classify the Delmarva
Peninsula; consequently, alternative classification variables were examined to determine whether they could provide better
correlation with water quality at the local scale. The following basin characteristics were examined:
(1) Topography
(la) Stream slope and the variables used to calculate stream slope;
(Ib) Stream-reach length;
(Ic) Altitude of stream at the basin outlet; and
(Id) Stream-reach relief (difference in altitude between stream outlet and highest altitude of stream reach).
(2) Soil type
(2a) Percentage of well-drained (type A) soils in basin;
(2b) Percentage of moderately well-drained (type B) soils in basin;
(2c) Percentage of moderately poorly drained (type C) soils in basin; and
(2d) Percentage of poorly drained (type D) soils in basin.
(3) Land use ....
(3a) Percentage of agricultural land in basin;
• (3b) Percentage of land in Animal Feeding Operations in basin;
(3c) Percentage of nursery operations in basin;
(3d) Percentage of urban land in basin; and
(3e) Percentage of forests and wetlands in basin.
Fourteen basins in the Chesterville Branch and Morgan Creek watersheds were delineated and used for the local-scale landscape
analysis (Figures 3 and 11). Nine of the basins are first-order, headwater watersheds; the others were included to examine
43
-------
systematic changes in landscape and nutrient concentrations within a watershed. Basin characteristics for the sites sampled in the
base-flow synoptic surveys are listed in Appendix F. '
The correlation pattern among the basin characteristics is shown in Table 6. In contrast to the findings of Phillips and Bachman
(1996), stream slope, the percentage of type D soils, and the percentage of forested land show no statistically significant correlation.
This result is not entirely unexpected, because the Locust Grove area represents one extreme of the landscapes in the Delmarva
Peninsula, and even the best correlations reported by Phillips and Bachman (1996) accounted for less than 40-percent of the total
variation. Despite this lack of correlation, the coefficients in Table 6 show some correlation among basin characteristics related to
topography, soil type, and land use. It appears that the strongest correlations are among the outlet altitude, percentage of type C
soils, and percentage of the basin in agriculture (Figure 23). Principal-components analysis was used on these three variables to
generate a principal-components score, which is essentially a new, synthetic variable that is based on the shared variance among the
original variables (Phillips and Bachman, 1996). The analysis was done on the rank-scores of the basin characteristics, which
results in a non-parametric test that requires no assumptions about the frequency distributions of the variables (Conover, 1980).
The analysis (Table 7) indicates that 77 percent of the variation was accounted for by the first eigenvalue of the correlation matrix
(the first principal component). All three variables are heavily loaded on the first component, and the communaHties (the percent
of total variance of each variable accounted for by the first component) range from 71 to 86 percent. The ^resulting principal-
components score is called the "local basin index" in this report to distinguish it from the "regional basin index" of Phillips and
Bachman (1996). The local basin index is computed from the altitude of the basin outlet, the percentage ojf type C soils, and
agricultural land use; the regional basin index is computed from the stream slope, the percentage of type D soils, and forested land
use. •
The areal distribution of the local basin index (Figure 24) displays a systematic decrease from the headwaters of each basin to the
gaging station. This result is not unexpected because the outlet altitude also systematically decreases from the headwaters to the
gaging stations (which have the lowest outlet altitudes), and the other variables are correlated with the outlet altitude. This trend is
true for both the first-order headwater basins as well as the nested mainstem basins. The pattern is found in both Morgan Creek and
Chesterville Branch; however, the values of the basin index in the downstream basins of Morgan Creek are much more negative
than those in the downstream part of Chesterville Branch. ;
Although the variables of the regional basin index do not show significant correlations at the local scale, ther£ are some indirect
relations (Table 6). The altitude of the basin outlet is significantly correlated with the length of the stream reach which, in turn, is
correlated with stream slope and the percent forested land in the basin. The outlet altitude also is significantly; correlated with the
percentage of type D soils in the basin. The percentages of type D soils and forested land have a significant inverse con-elation with
percentage of agricultural land within a basin. ;
The relations among basin index, nitrogen concentration, and location of sites within the stream network (Figure 24) show patterns
that could indicate that denitrification at the top of the Aquia confining layer is an important mechanism to explain the differences
between nitrogen concentrations in Morgan Creek and Chesterville Branch. This relation for the April 1998 synoptic data is shown
in Figure 24. The September 1998 synoptic data show a similar pattern, although the nitrogen concentrations generally are lower.
Most of the first-order basins, and many of the nested mainstem basins, show a rough correlation of increasing nitrogen with
increasing value of the local basin index. The nested mainstem basins display downstream patterns (increasing concentrations for
Morgan Creek and decreasing concentrations for Chesterville Branch) similar to the actual downstream plots of concentrations by
distance from the gage (Figures 17 and 24). The three samples that depart significantly from the correlation (sites 01493491,
01493495, and 01493496) are from the headwaters of Morgan Creek, in the area where the Aquia confining layer subcrops
shallowly, and allows a larger percentage of flow paths in the surficial aquifer to flow through the redox transition zone. The
nitrogen concentration at site 01493496 also may be explained in that the tributary sampled there was dammed into a small farm
pond until shortly before sampling, when the dam was breached. Base flow from site 01493496 could represent a certain amount
of ground-water discharge that has passed through former pond-bottom sediments that may still be under reducing conditions.
44
-------
o
o
I
."I
I
ji
cq
U3
.^
0
s
'1
Q.
s
«
c
Q8L
v
O
0
cL w
*"* OT
C/3
2? CQ
O v.
X<
£ «
S 5-
2 3.
£o w
o «•>
•§ M
3 c
2 -|
•g^
s^
s ^
s.^
o
i — i
11
00 —
o
r— 1
M
''O
o
t
B-s
II
o
o
t.
o
^o
0
| a
|j
< o
o
o
0
o
s
9
00
VO
o
tream slope
00
o
o
1— 1
CO
o
f-
o
oo
CO
9
,_ i
o
^» o
••-> w
0
O
VO
O
a
o
CO
o
CO
o
>n
T~~l
9
CQ
O cfl
o
C3
! 1
oo
9
o
o
t
CO
o
VO
o
>n
o
i
o
CN
O
O
<1> V.
O
O
00
o
VD
CN
9
CN
9
CO
9
oo
CO
o
0
o
1
1 — t
o
Q
g
*
VO
9
VO
o
VO
o
s
9
,_
o
s
o
1
p
0
^
*-J
o
1
%
agriculture
o
o
'""'
V)
o
CO
o
CO
o
vo
o
o
0
o
o
t—
VO
o
t-
9
J— 1
CO
o
o
?
o
o
' '
o
o
t— 1
o
o
9
CN
9
o
CN
0
CO
O
v>
o
VO
O
ON
O
1
VO
VO
e
% nursery
o
o
CN
CN
O
O
o
i — r.
O
CO
0
oo
CO
o
s
o
s
o
t-
o
o
CO
0
CO
o
1
.00
CO
0
c
CO
-e
o
0
§
o
ON
o
CO
CO
o
oo
o
CO
CO
0
CN
CO
o
o
9
CN
o
o
CO
9
oo
vo
0
vo
vo
o
VO
o
1
45
-------
O 15
100
l^Speanpanrankconelatiori v
coefHtieat=673
S 88S
- p = 00013
- f -> *.
j
•*^s •
. rfrte
#. % •
• •
^.sf - V
ALTITUDE OF BASIN OUTLET.
IN FEET ABOVE SEA LEVEL
ALTITUDE OF BASIN OUTLET,
IN FEET ABOVE SEA LEVEL
PERCENTAGE OF BASIN WITH
TYPE C SOILS
Figure 23. Relations among stream-outlet altitude, percentage of type C (moderately poorly drained) soils, and percentage of
agriculture in the Morgan Creek and Chesterville Branch basins, Locust Grove study area, Kent County, Maryland.
Table 7. Results of Principal-components Analysis of Landscape Variables for the Morgan Creek and Chesterville Branch
Watersheds, Locust Grove Study Area, Kent County, Maryland |
Number of samples =16
Three variables:
ALT_MIN - altitude of the basin outlet
PCT_C - percentage of the basin with Type C (moderately poorly drained) soil
PCT_AG - percentage of the basin under agriculture
Values of all variables were converted to rank scores and normalized
(means = 0; standard deviation and variance = 1)
Total variance of all three variables = 3
Eigenvalues of the correlation matrix:
Component 1
Eigenvalue 2.31
Proportion of total variance 0.77
2
0.46
0.15
3
0.23
0.08
Component # 1 is the only one retained; 77 percent of the total variance is explained by •
this component.
Component loadings for each variable:
ALT_MIN: 0.86
PCT_C: -0.84
PCT AG: 0.92
Final communality estimates: Total = 2.31 (0.77)
ALT MIN: 0.75 PCT C: 0.71
PCT AG:
0.86
46
-------
or
iii
cc
m
Q.
CO
EC
O
HI
a
§
Q
111
§
yfe*3~y l-Incfraanrf^i' '•>> -/fe^-r:'-:.:-': iv:~:' -
-2
-1.5
-1
-0.5 0 0.5
LOCAL-SCALE BASIN INDEX
1.5
EXPLANATION
Q SITE ON FIRST-ORDER TRIBUTARY OF CHESTERVILLE BRANCH
0 SITE ON FIRST-ORDER TRIBUTARY OF MORGAN CREEK
• SITE IN NESTED MAINSTEM BASIN
* PATH OF STREAMFLOW ALONG MAINSTEM OF CREEK
«-— PATH OF STREAMFLOW FROM TRIBUTARY INTO MAINSTEM
Figure 24. Relations among basin index, dissolved nitrogen concentration, and location within the stream network, Locust
Grove study area, Kent County, Maryland.
-2.6
-2.4 -2.2 -2.0
REGIONAL BASIN INDEX
-1.8
Figure 25. Correlation between regional and local basin index scores for sites in the Locust Grove study area, Kent County,
Maryland.
47
-------
-------
Summary and Conclusions
This report presents the results of data collection and analysis by the U.S. Geological Survey in support of ground-water flow
modeling by the U.S. Environmental Protection Agency for the Locust Grove study area, Kent County, Maryland. This effort has
provided evidence that subsurface hydrostratigraphic units substantially affect the transport and fate of dissolved nitrogen through
the surficial aquifer to discharge as stream base flow. The highly permeable surficial aquifer in the study area is underlain by a low-
permeability confining layer, which dips to the south. Continuous cores and geophysical logs indicate that the confining layer
approaches the surface near the riparian zone of Morgan Creek, but is deeper than 20 to 30 meters below the surface near
Chesterville Branch. The Chesterville Branch watershed is completely underlain by a thick, permeable surficial aquifer that yields
considerable base flow, whereas much of the Morgan Creek watershed is underlain shallowly by low-permeability sediments.
In addition to affecting ground-water flow paths and discharge to streams, the confining layer also has a geochemical environment
that may be conducive to denitrification of dissolved nitrate in the ground water. Reduced-iron minerals are abundant in the
confining layer but generally have been removed by weathering from the overlying aquifer. Chemical analyses indicate that ground
water in the surficial aquifer evolves from a high-nitrate water dominated by calcium to a low-nitrate, bicarbonate water with more
sodium. The concentrations of dissolved nitrogen and argon gases; dissolved oxygen, iron, and sulfate; and nitrogen and carbon
isotopes as tracers, all provide evidence that the low-nitrogen water discharging to the streams originally contained considerable
amounts of nitrate that was subsequently denitrified. Ground water that discharges to Morgan Creek is more likely to interact with
the sedimentary redox zone at the top of the Aquia confining layer, and appears to have a higher proportion of denitrified water than
ground water that discharges to Chesterville Branch. The geochemical data also confirm that denitrification was not active in
ground waters in the surficial aquifer above the sediment redox transition. Nitrate in the ground water has remained stable for at
least 40 to 50 years where it flowed through the deeply weathered sediments and oxic conditions of the Columbia-Aquia surficial
aquifer.
Base-flow nitrogen yields are about twice as high from Chesterville Branch (median of 0.058 grams per second per square
kilometer) than from Morgan Creek (0.022 grams per second per square kilometer), although annual loads are only 45 percent
higher in Chesterville Branch (median of 29,000 kilograms per year) than in Morgan Creek (20,000 kilograms per year). Although
the two watersheds were identified as having similar hydrologic landscape characteristics in previous studies, a more detailed
analysis indicates that the Morgan Creek basin contains soils that are somewhat more poorly drained. Soil drainage, altitude of the
watershed or sub-watershed outlet, and the percentage of agricultural land use are related to each other and to total dissolved
nitrogen concentrations in stream base flow.
The base-flow synoptic data show that nitrate concentrations tend to increase downstream in Morgan Creek and decrease
downstream in Chesterville Branch. The downstream increase of nitrate concentrations in Morgan Creek may be due to flow from
tributaries in which ground-water discharge is less likely to be denitrified, or from a downstream decrease in denitrification of
discharging ground water. The downstream decrease of nitrate concentrations in Chesterville Branch may be due to a downstream
increase of instream denitrification and uptake by biota, or by an increase in the relative proportion of older ground-water
discharge; this older ground water may have relatively low nitrate concentrations either because of lower historic inputs of nitrogen
or because of denitrification. There also is some evidence that the shallow Aquia confining layer may be breached, possibly by a
fault, under the riparian zone of Morgan Creek near site 01493495, and low-nitrate water from the underlying confined
Hornerstown aquifer may be contributing to base flow.
Further study would help to better define the processes affecting ground-water discharge to streams. In particular, an extended
hydrostratigraphic framework and ground-water geochemistry from wells in downstream sections of the watersheds, below sites
01493495 and 01493110, would contribute to an improved understanding of the lower reaches of the two streams. More detailed
data on the shallow subsurface geology of the riparian zone would help determine whether Hornerstown water is discharging to
Morgan Creek, and provide further evidence to evaluate the role of organic-carbon oxidation as a mechanism for denitrification in
both watersheds. Additional wells in the Hornerstown aquifer, particularly to the north of the study area, would provide data on the
source of ground-water recharge and the geochemical evolution of water in the Hornerstown aquifer. Finally, additional chemical
and biological sampling of streamwater would be helpful in characterizing instream processes that may be sources and sinks for
nitrogen and other dissolved constituents.
49
-------
-------
References Cited
Bachman, L.J., 1984a, Nitrate in the Columbia aquifer, central Delmarva Peninsula, Maryland: U.S. Geological Survey Water-
Resources Investigations Report 84-4322, 51 p.
, 1984b, The Columbia aquifer of the Eastern Shore of Maryland, part 1: Hydrogeology: Maryland Geological Survey
Report of Investigations No. 40, 34 p.
Bachman, L.J., 1994, Hydrologic control of dissolved aluminum in the base flow of streams on the Delmarva Peninsula, Delaware
and Maryland, USA: Washington, D.C., George Washington University, Ph.D. dissertation, 107 p.
Bachman, L.J., and Kate, E.G., 1986, Relationship between precipitation quality, shallow ground-water geochemistry, and
dissolved aluminum in eastern Maryland: Baltimore, Maryland Power-Plant Siring Program, Publication PPSP-AD-14, 37 p.
Bachman, L.J., Lindsey, B.D., Brakebill, J.W., and Powars, D.S., 1998, Ground-water discharge and base-flow nitrate loads of
nontidal streams, and their relation to a hydrogeomorphic classification of the Chesapeake Bay watershed, Middle Atlantic
Coast: U.S. Geological Survey Water-Resources Investigations Report 98-4059, 71 p. (Also available online at http://
md.usgs.gov/publications/wrir-98-4059)
Bachman, L.J., and Phillips, P.J., 1996, Hydrologic landscapes on the Delmarva Peninsula, part 2: Estimates of base-flow nitrogen
load to Chesapeake Bay: Journal of the American Water Resources Association, v. 32, p. 779-791.
Bara, T.J., compiler and editor, 1994, Multi-resolution characteristics consortium - Documentation handbook [Environmental
Monitoring and Assessment Program - Landscape Characterization, Contract 68-DO-0106]: Research Triangle Park, N.C.,
ManTech Environmental Technology, Inc. [variously paged].
Bohlke, J.K., and Denver, J.M., 1995, Combined use of groundwater dating, chemical, and isotopic analyses to resolve the history
and fate of nitrate contamination in two agricultural watersheds, Atlantic Coastal Plain, Maryland: Water Resources Research,
v. 31, p. 2,319-2,339.
Borum, Jens, 1996, Shallow waters and land/sea boundaries, wEutrophication in coastal marine ecosystems, J0rgensen, B.B., and
Richards, K., eds., Coastal and Estuarine Studies, v. 52, p. 179-203
Busenberg, Eurybiades and Plummer, L.N., 1992, Use of chlorofluorocarbons as hydrologic tracers and age-dating tools: Water
Resources Research, v. 28, p. 2,275-2,283.
Carey, J.B., Cunningham, R.L., and Williams, E.G., 1976, Loess identification in soils of southeastern Pennsylvania: Soil Science
Society of America Journal, v.40, p. 745-750.
Clark, W.B., 1915, Map of Kent County showing the geological formations: Maryland Geological Survey, 1 sheet, scale 1:62,500.
Conover, W. J., 1980, Practical nonparametric statistics, (2nd ed.): New York, John Wiley and Sons, 493 p.
Gushing, E.M., Kantrowitz, I.H., and Taylor, K.R., 1973, Water resources of the Delmarva Peninsula: U.S. Geological Survey
Professional Paper 822, 58 p.
Drummond, D.D., 1998, Hydrogeology, simulation of ground-water flow, and ground-water quality of the upper Coastal Plain
aquifers in Kent County, Maryland: Maryland Geological Survey Report of Investigations No. 68, 76 p.
Dunkle, S.A., Plummer, L.N., Busenberg, Eurybiades, Phillips, P.J., Denver, J.M., Hamilton, P.A., Michel, R.L., and Coplen, T.B.,
1993, Chlorofluorocarbons (CC13F and CC12F2) as dating tools and hydrologic tracers in shallow groundwater of the Delmarva
Peninsula, Atlantic Coastal Plain, United States: Water Resources Research, v. 29, p. 3,837-3,860.
Fisher, T.R., and Butt, A.J., 1994, The role of nitrogen and phosphorus in Chesapeake Bay anoxia, in Perspectives on Chesapeake
Bay 1994: Advances in estuarine sciences, Nelson, S., and Elliott, P., eds., Chesapeake Research Consortium Publication No.
147, p. 1-43.
Fishman, M.J., and Friedman, L.C., eds., 1985, Methods for the determination of inorganic substances in water and fluvial
sediments: U.S. Geological Survey Open-File Report 85-493, 709 p.
Focazio, M.J., Plummer, L.N., Bohlke, J.K., Busenberg, Eurybiades, Bachman, L.J., and Powars, D.S., 1998, Preliminary estimates
of residence times and apparent ages of ground water in the Chesapeake Bay watershed, and water-quality data from a survey
51
-------
of springs: U.S. Geological Survey Water-Resources Investigations Report 97-4225, 75 p. (Also available online at http://1
md.usgs.gov/publications/wrir-97-4225) ! ',
Groot, J.J. and Jordan, R.R., 1999, The Pliocene and Quaternary deposits of Delaware: Palynology, ages, and paleoenvironments:'
Delaware Geological Survey Report of Investigations No. 58, 36 p.
Hamilton, P.A., Denver, J.M., Phillips, P.J., and Shedlock, R.J., 1993, Water-quality assessment of the Delmarva Peninsula,!
Delaware, Maryland, and Virginia-Effects of agricultural activities on, and distribution of nitrate and other inorganic
constituents in the surficial aquifer: U.S. Geological Survey Open-File Report 93-40, 87 p. ,
Hansen, H. J., 1992, Stratigraphy of Upper Cretaceous and Tertiary sediments in a core-hole drilled near Chesterville, Kent County,
Maryland: Maryland Geological Survey, Open-File Report No. 93-02-7, 38 p. I \
Hantush, M.M., and Cruz, Jerome, 1999, Hydrogeologic foundations in support of ecosystem restoration: Base-flow loadings of
nitrate in Mid-Atlantic agricultural watersheds: U. S. Environmental Protection Agency, EPA/600/R-&9/104, 46 p. (Also
available online at http://www.epa.gov/ada/) •
Hardy, M.A., Leahy, P.O., and Alley, W.M., 1989, Well installation and documentation, and ground-water sampling protocols for
the pilot National Water-Quality Assessment program: U.S. Geological Survey Open-File Report 89-396, 36 p. i
Jordan, R.R., 1962, Stratigraphy of the sedimentary rocks of Delaware: Delaware Geological Survey Bulletin No. 9, 51 p.
Jordan, R.R., 1964, Columbia (Pleistocene) sediments of Delaware: Delaware Geological Survey Bulletin No|. 12,69 p.
Koterba, M.T., Wilde, F. D., and Lapham, W.M. 1996, Ground-water data-collection protocols and procedures for the National
Water-Quality Assessment Program: Collection and documentation of water-quality samples and related data: U.S. Geological
Survey Open-File Report 95-399, 113 p. !
Lewitus, A.J., Jesien, R.V., Kana, T.M., Burkholder, J.M., Glasgow, H.B., Jr., and May, E., 1995, Discovery of the "phantom"
dinoflagellate in Chesapeake Bay: Estuaries, v. 18, 373-378. j
Maryland Department of State Planning, 1973, Natural soils groups of Maryland: Maryland Department of State Planning,
Technical Series, General Land-Use Plan, 153 p.
McFarland, E.R., 1995, Ground-water flow, geochemistry, and the effects of agricultural practices on nitrogen transport at study
sites in the Piedmont and Coastal Plain Physiographic Provinces, Patuxent River Basin, Maryland: U.Sl Geological Survey
Open-File Report 94-507, 78 p. I
Minard, J.P., 1974, Geology of the Betterton quadrangle, Kent County, Maryland, and a discussion of the regional stratigraphy: ;
U.S. Geological Survey Professional Paper 816, 27 p., 1 plate, scale 1:24,000. |
Modica, E., Buxton, H.T., and Plummer, L. N., 1998, Evaluating the source and residence times of groundwater seepage to streams,
New Jersey Coastal Plain: Water Resources Research, v. 34, p. 2,797-2,810. i [
Mueller, D.K., Hamilton, P.A., Helsel, D.R., Hitt, K.J., and Ruddy, B.C., 1995, Nutrients in ground water and [surface water of the ;
United States: An analysis of data through 1992: U.S. Geological Survey Water-Resources Investigations Report 95-4031,
74p. |
Mueller, O.K., and Helsel, D.R., 1996, Nutrients in the Nation's waters: Too much of a good thing?: U.S.j Geological Survey !
Circular 1136,24 p. (Also available online at http://wwwrvares.er.usgs.gov/nawqa/CIRC-l 136.html) | :
Otton, E.G., and Mandle, R.J., 1984, Hydrogeology of the upper Chesapeake Bay area, Maryland, with an emphasis on aquifers in
the Potomac Group: Maryland Geological Survey Report of Investigations No. 39, 62 p. j
Overbeck, R.M., and Slaughter, T.H., 1958, The water resources of Cecil, Kent, and Queen Annes Counties: Maryland Department
of Geology, Mines, and Water Resources (now Maryland Geological Survey) Bulletin No. 21, p. 1-382 J
Owens, J.P., compiler, 1967, Engineering geology of the Northeast Corridor- Washington, D.C. to Boston, Massachusetts - ;
Coastal Plain and surficial deposits: U.S. Geological Survey Miscellaneous Geologic Investigations Map I-514-B. '
Owens, J.P., and Denny, C.S., 1979, Upper Cenozoic deposits of the central Delmarva Peninsula, Maryland and Delaware: U.S.
Geological Survey Professional Paper 1067-A, 28 p. . j i
Owens, J. P., and Minard, J. P., 1979, Upper Cenozoic deposits of the lower Delaware Valley and northern Delmarva Peninsula,
New Jersey, Pennsylvania, Delaware, and Maryland: U.S. Geological Survey Professional Paper 1067-D, 47 p.
Phillips, P.J., and Bachman, L.J., 1996, Hydrologic landscapes on the Delmarva Peninsula, part 1: Drainage basin type and base-
flow chemistry; Journal of the American Water Resources Association, v. 32, p. 767-778. I
Pritt, J.W., and Raese, J.W., eds., 1995, Quality assurance/quality control manual - National Water-Quality Laboratory: U.S. ..
Geological Survey Open-File Report 95-443, 35 p. '
Reilly, T.E., Plummer, L.N., Phillips, P.J., and Busenberg, Eurybiades, 1994, The use of simulation and multiple environmental
tracers to quantify groundwater flow in a shallow aquifer: Water Resources Research, v. 30, p. 421-433. i '•
52
-------
Richardson^ Katherine, and J0rgensen, B.B., 1996, Eutrophication: Definitions, history and effects: in Eutrophication in coastal
marine ecosystems, Jorgensen, B.B., and Richardson, Katherine, eds., Coastal and Estuarine Studies, v. 52, p. 1-19.
Shedlock, R.J., Denver, J.M., Hayes, M.A., Hamilton, P.A., Koterba, M.T., Bachman, L.J., Phillips, P.J., and Banks, W.S.L., 1999,
Water-quality assessment of the Delmarva Peninsula, Delaware, Maryland, and Virginia: Results of investigations, 1987-91:
U.S. Geological Survey Water-Supply Paper 2355-A, 41 p.
Simmons, G.M., Jr., Reay, W., Smedley, S.B., and Williams, T.M., 1990, Assessing submarine ground-water discharge in relation
to land use, in New perspectives in the Chesapeake system: Proceedings of the Chesapeake Bay Research Conference,
December 4-6, 1990, Baltimore, Md., p. 50.
Speiran, O.K., 1996, Geohydrology and geochemistry near coastal ground-water-discharge areas of the Eastern Shore, Virginia:
U.S. Geological Survey Water-Supply Paper 2479, 73 p.
Taylor, J.K., 1988, Quality assurance of chemical measurements: Chelsea, Mich., Lewis Publishers, 328 p.
Tompkins, M.D., Cooper, B.F., and Drummond, D.D., 1994, Ground-water and surface-water data for Kent County, Maryland:
Maryland Geological Survey Basic Data Report No. 20, 155 p.
U.S. Geological Survey, 1977, National handbook of recommended methods for water-data acquisition: U.S. Geological Survey,
Office of Water Data Coordination, 872 p.
Vokes, H. E., and Edwards, Jr., Jonathan, 1974, Geography and geology of Maryland: Maryland Geological Survey Bulletin 19,
242 p.
Wilde, F.D., and Radtke, D.B., 1998, Field measurements; in National field manual for the collection of water-quality data: U.S.
Geological Survey Techniques of Water-Resources Investigations, Book 9, Chapter A6, [variously paged].
53
-------
-------
Appendix A
55
-------
o
1
o
§0
p
£
I
1
O
to
o.
(3
CO
I
if
g
M $
*"* P
E-S
^ s
^£
$
2
"a
ea
•Si §
'
§ i
"3.5.2
>»
t3 /— •*
.3 >•• "a*
._ oo
^ £ a
•O C -O
ill
i|«
***
— ° .—,
2 .= MtS
£ 0..2 ,2
t"« « *— '
i
2 a
j?~
M g
"3 o
o-S
1=3 C-
0 <-, <••,
•3 o gC-
~ H S '— '
E.2-S.
S 2 "S
G JS Q>
e v« Jj-,
PQ o
11^"
« "3 !='
ft** C x— v
o* ^ >— >
CS Q"
O O
S f2
CO OO
•^- co
o £
•* ••*
o P
CO C^
r~* vo
Ol CO
•^- co
Ol CO
CO to
O f-;
t- t-
^ °°
to to
C""~ C~^
— 0
oo oo
ON ON
en en
i 1
to to
JO *O
0 g
oo oo
ON ON
CO CO
— 01
O 4>
CQ CQ
a a
o o
0 0
J -1
to 0
Ol en
O Q
1 f
§ o]
^-i \O
VO OO
oi -^
to ^3-
^-i VO
VO oo
to -^J-
»-4 VO
vq co
^J- Tf
oi to
<"•* \o
OJ
^ to
^* c^
01
OJ CO
;-« vo
vq co
r- r-
oo oo
to vo
to to
O OJ
i— • CO
oo oo
ON ON
CO CO
Ol CO
to VD
to to
to . to
§ §
~^ CO
oo oo
ON ON
CO CO
CO ON
o> o>
CQ CQ
rtl rrl
S S
O O
O O
J -1
0 «0
m vo
•^j-
Q "^
Q
D 3
^ ^
to >o
01 01
Ol OO
VO O
1—4 TJ-
to oi
Ol OO
VO O
CO f^
to OJ
Ol oo
06 ^
to to
S ^
to to
OJ 5
CO VO
01 00
-- IO
oo -3-
r— r—
ON Ol
-3- to
vo >o
to >o
to to
= S
oo oo
ON ON
CO CO
•— t CO
0 0
ON OO
-?J- to
vo to
to to
to to
2 §
OO OO
ON ON
CO CO
O —
vo vo
o
Ol Ol
S S
oo oo
•^J- oo
- _
,— « 10
o to
Ol Ol
OO CO
CO ^fr
ON Ol
o o
oo oo
to to
to to
ON OO
oo oo
ON ON
CO CO
o o
ON ON
o o
oo vo
to to
to >o
o o
ON oo
00 00
ON ON
CO CO
to »o
o o
CQ CQ
a a
<^~.> t_ '
S S
o o
*<3" t*-
3 3
Q" Q"
II
to to
Ol Ol
CO Ol
CO
*"!*
"* CO
CO OJ
s*
to co
Ol I-;
OJ
T^* IO
<^> 00
vo
*-« »o
co ^
vo
§ R
to t^
vo ol
^
to vo
to -^
r- to
to to
to to
*•* o
-- OJ
oo r-
ON ON
CO CO
o o
to vo
to -^j-
r-- to
10 to
o
i
VO VO
to to
>o «o
to >o
o cr>
OJ Ol
r— f-
ON ON
CO CO
i
01 CO
vo vo
to to
to to
to to
o o
o o
01 01
r- t--
ON ON
CO CO
1
S —
vo
di '(o
m m
PQ fT1
kyt *syj
"5"
"5
§'=
CO
o
1
-.s
to
vo
•S
r
o
1
to
-S
.§
£
1
IW.
•p*
f
S
m
S
E
1=
56
-------
'So
t
I
(4-1
o
60
!
t/D
"3
o
I
O
I
o
£!
d> C3
^£
4
|
I
«
£
«
•0
"3
j.
§
Q
"O
1
|
"a!
X)
JS
1
OJ >»
lit
_ Ml
0.5 „
T3 C -O
H-B
S £ 2
J3
S s g> 3!
•0 ^
<~
° 10
Q. "
£~
11
"
"S £
o S
(S
*•
o
Q
D
c^
oo
ON
*
OO
ON
1
OO
oo*
f^
vo
vo
o
o
s
oo
p2
n —
vo oo
0 0
01 ^f
W-l — '
vq oo
^-^ en
00 00
^t" Ol
^ fQ
o »n
\2 00
\o \o
ON 0\
O O
r- r*
OO CO
en en
ON ON
en en
•S.8
g g
VO "O
o o
oo oo
en en
CO OO
ON ON
en en
S S
£ £
W £l3
•"--' ' ' "- •" •" o S S o
o o o o
S >— 3 fcJ 1 I 1 *g ^ ^ ^ 1 ^ j ] ; i
ol ntncnc40vncNON —
°^ ^ %d oo oo oo **? "7 ^ °. °^ ON
>n en ~ — -^ «
cM^Tfooo^ooiOtncooN —
Oencnenenen ' ' ' \p o en »— »
•<3- en ^- 20>nooog
en en en *~* — ol
en en en *~ ' ^^ ol
01 — enOio^iololngggj
vom \ovovo \ovovovo
o>t-r-oootacsot>oooooo
Svii?j>n5«?>n>r5°'ri>o>o
icjc^pio^p^pjcppp
oochOoloot-ooen.-t-ololol
SPPJPSPSSPPPPP
G>i9*C7*O*GNO'\C7\ON
en en en en en en en en en en en en en
S 0
g s
VO Vt
jn i ^ | . | | 1 1 t 1 1 ! t -
o o
oo o
en CM
oo r^
ON ON
en en
kSP-og-^^SS^SS
03 03 CQ CQ CQ CQ "o "o ^ oo ON ON ON
SSS'SSSsa^^a^a
57
-------
Appendix B
58
-------
T3
03
£
O
60
_C
"el
I
oo
"5
o
1
O
S
(D
ii
w ca
CO *^L
r^
-I
•S a
^a
l|
io
^ -4^
$3
*£
3-2
M I
o-g J
OJ fl ^^ ^
1 slS
2 e 1
« « S o
O, OB g «
al'a'25
1,1 Si
** *••* fl /"~v
fill
ill'
.3
o S
Q.s
- S
pvqcM -^ en en O CM
en ON ^ £ t^ co
\0 ~ --
=? I
— '^enen^-cnmcMcM O —
O
'— eoen
ovor^ ooocoooovocooiooco-^vqovoo^-^focjj inr^vq^tcoco^-^r-;^
vo m — o^cococMincMcoininri-—. •n-coco-*cocooJo) o^ — -* -^t -^- »n —*
r- oo oo oo oo oo tr-
s
COCOOOCOCOOOOOCOOOOOOOOOOOOOOOOOOOOOOOOOOOCOCOOOOOOOOOOOOOOOOOOOOO
I
fO PO
g§
^ ^
O O — CM T
s s i s s s
ON ON ON ON ON ON
H3 [Tl [ii fT1 fT1 TT1
E3? ^2 ^ 2 2 2
oo in *n
CMCM— ; CM
^OOO
ONONON
VO vo vo vo VO vo vo
CM CM CM CM CM ~' ~"
o o o o o
ON ON ON ON ON
CMCMCMCMcM
OOOOO
-
t*- r— r~-
s s s
CM
O
ONONONONON
59
-------
Appendix C
60
-------
1
OS
"•5
ro
1
OS
s
i
ON
§
o
OS
)£
s
OS
i
OS
0
1 *
c3 in
|l I
••* fl* fc
5 M
§ 1 8
rf 1 1
^ o £
p% s
5 S
00 "
> °°
o S
2 i
to
3 'S
j i
•1 *
«i J2
'S £
^ 00
O 52
*•£> S
rt o
g
1> os
.2 ^
o §
C *
« 5
O rj
g. |
-S 5
rt oo
u
€
g
11
CM NO
r-; CM
ON O
CM CO
CO 0
CM CO
o ••*•
CM CO
oo •*
XO CM
od ON
CM CM
& ~
CM CO
0 ON
CM r-
o o
xrj ~
NO CM
ON O*
NO CM
OO ON
— * t~.
ON ON
CM CM
t5 Z^
\o •— •
ON oo
o" o
CO CO
\o —
CM r-^
CM CM
CO $
CO CO
CM 00
CM CM
S S
CM CM
CO CO
1 !
%
J3
1 55 S
C — 12
•^j CQ CQ
£s« pj pj
i? 2 2
us
NO in co
o O co
^j~ ^ i—-
•3- Tj- Ttf-
XO NO £
CO CO \C5
rf- -^j- -<3-
co m m
t~- t** oo
CM CM" **
Tf ^" TT
£ E o
CO CO \O
^ r^ o
•<* T£ od
OO t^ CM
ON ON NO
555
CO CO OO
T£ Tf OO
ON OO —
co co od
O\ ON O
co to od
co -— O
O O NO
Tj- -^- NO
^r ^r ^r
ro co *n
It 2 °°
o o\ r-
•— t ^ \O
>n »n od
Tt •* T3-
r- NO vo
>n wS co
NO S o7
•"* ^* v>
CM CM 00
^F ?? v>
NO
•^r
I 1 t
0^ \0 vo
0^2
oo CQ CQ
pj pq pi
S S S
a - s
\O CM CM"
••T v~i m
S S £
XO — — H
O — * t*-
oq CM CM
CO O* O"
OO CM —
•^- >n in
— • oo oo
co CM' CM
Tf O. 0
5 £ £
\O 00 —
r- co co
r- r- o
CM in NO
r^ co co
ON oo NO
OO NO NO
$ 55 R
— * CO OO
, — . VO NO
>n co co
•^t >T) »0
co o O
r^ ^ -^
co r- VN
*/"! ^" ^f
-* >n >n
0 CO CM
1 i 1
o; os ~
VI VS V>
SI— 1 t^l
OO OO
— ' r^ r^
R ' '
1 II
•* •* 10
51 2 2
*o W CQ
s § S
1 1
£ D
c-4 cj
S S
^ ^
Os ^O
(-- Os
en Os
? 0
-* o
^ ^
ON —*
ON 0
?j S
ON CM
CO Tt-
— * CO
"*" **
— CO
oq oo
£ S
OO NO
o o
to »o
w-i >o
in in
^-" O
— •*
ON OO
>n >n
co n-
S S
S 2
Wl >O
ft cs
od od
S '
NO*
V)
1 1
-* CO
i°, ^
W PQ
pj pj
11
S S
^jj- ^1
»n in
IS NO
CO CO
TT CO
in in
CO CO
CO CO
in *n
55 S
»n in
£ So
£ S
— o
u-i in
-* CO
in >n
ON ON
in in
S 2
NO NO
in m
ON ON
NO NO
m •*
CM C^
od od
NO •*
o O
ON ON
SOO
NO
in »n
~- CM
ON ON
O\ ON
} I
I 1
^j- in
CM CM
W pQ
a a
od
-
od
•^f
2
M
r^
>n
*
S
^
CO
od
5
°t
OO
oo
CO
in
od
CO
??
>n
oo
"*
o
ON
o
i-^
O
m
OO
o"
-*
CO
in
vo
CO
CM
OO
1
CM
S
^6
o
S
i i
8 ?
CM ON
T}- CO
ON oq
— CO
CO OO
NO NO
o* r^
"* CO
— . oq
CO CO
CO O
-* -a-
0 jo
^* §'
ON ON
— ON
g §
— * ON
R «
-^ od
oo ON
in CM
— ON
^ *°
O ON
0 NO
\o ^^
« o"
oo —
i' 5'
ON f-
CO CM
•^- 'd-
NO CM
in co
CM CO
1 '
1 1
S ts
0 S
OO (S
g S
11
vs r^
c^i o
9 ?!
c4 o"
<0 Os
>n oq
5 &
5 8
CM O"
{Q 8
CM O*
— vo
q CM
CO O
— ON
n
en '
?
CO
»n
o
&
o
CO
00
3
in
oo
O
oo
o"
£
o
r-
o
"*
r-^
o
^
ON
^
CO
ON
§
~
•^~
xn
en
5
CM
R
m
so
P3
a
^,
"O
I
i
E
c
.,
1
f
•S
g
0)
g
' §
1
1
.5
•§
i
§
*•- ^i
61
-------
I
•s
o
o
T3
C
C3
I
t
§
V*
3
!3
£3
i
c
^
£
^
Q
is
1
o\
s
3
0*,
CO
M
s
0\
i
0
—
§
0>
s
ft
O
§
1
§
I
O
u
s
55
S
o
eo
§
£J
?5
CO
s
eo
S
§
eo
i
eo
O
i
CO
1
«
1
O
CO
s
0
1?>
§
1
g
M
a
c
"T?i
£
VI — O\
vo — * ON
3!? S
Vi ON oo
CN co r-^
— vo vo
cn r>- — •
— VJ VJ
«•• \d vo
Kg 8
o in vi
, T T Tj-
CJ VJ
JCvo ^
?J? ^
O — CO
t"; ON en
O ^* V)
rrm m
SIS §
~- CO CO
"*)• "T ^3"
Sc5 0
O ON ON
S.8. S
3? 5
\o in o
5° °
\o 2 §
••^ O ON
"* 1 I
O
^r
^4- V)
S 5 I
o 5 5
tx] m U
^ S3
£ ^
5 iS
\O vo
en vo
in cn
vo vo"
g P
£ 3
«.
-a vj
3 s
SVD
^3"
CO
•a vo
1 ^f
^ «
5 ^ -
•§ c5
t*^ c^>
S o
5 "»
vn 5
r^ r-^
2 55
oo vd
•n rr
° -.
CO CO
•^r -^r
VJ VO
ON ON
vo O
° s
2 S
O O
S '
ON
rj-
1 1
\O
"«f in
O CN
Tl- 0
ON CD
& 3
r-cooQ1** p-cnovoo\ o cn vj CN o
voooTrvovj •sfr^-cN'^cn "nvjtnrj-c^
enoNOOo TfTtTfr^r^ r^r^r^-^r^
-•^•^j-vjvjvj. vjinvjinin vjinmvjvj
mencnm — • NO ON vj in •^•t^-ONinco
VOCNCNCN CO — O VO Vj VOVOVOVOCN
^•^invjin ininmvjin minininvj
— * O ON r- O — OOI^ON vor^OOvOO
vqinoNON -^t vqcncN cooooovjo
cnoNONOvo cn>»cnr^t-- r^r^r^ri-r^
Q
cNvovo-^-t-- CN —incn O-^OVOON
t~»»nVOOOOO —^ '^J-CN'— CNCNCNONO
cNONcdoocd cn>%cnvdvd vdvovdcnvo
"^ **• *sf ^ ^*- v> *- v> vj »n invjinvjin
KSroSS ooSSSoo S'SS'^g
*^* vj *^~ ^ ^ vj in v*i in vj vj in vj in vj
vovovovor- «cnoo« enr-covocn
o vo co co co ocncncor^ vovovocor—
^vivjinvj vjvjvjvjin vjvjvjvjin
en V) en CN en o en en *-^ *-* r~~ eN ^-j o ^*
^inv> vjvjvjvjvj >n»ninvjin
cnr^-*-**— >en cncNocnvo ON o o vo — «
r-vococooo T-.-^-TI-^HO o — - — CN CN
TT in in vi vj vi vj in vj vj in in in in v)»nvj vjvjvjvjv-j »nvjvjinvj
cNcOT—eNen vo-*cNT,t-co CN — « •*£ ON ON
•^- f -t o O O ^^ ^T ' *-1" OO t^- vo vo vo ~~* CN
Vj en CN CN cN t~~ r** I*- ON ON ON ON ON CO ON
•^•invjinxn invjinmin invjvjvjin
^-^incNin vjvjcNcNin o\-*inoNt^
—• ; O VJ VJ VJ ONCNCNVJVJ ONOOCOCO
T^ vj v> vj. *n *n in "n NO NO in vo NO in vj
r-t^-TfcNvj mr-^i-cNr^ cnvooNcovo
covocnenen tncocovjrr oooeno
Tj-tnviinvj invjinvovo vovovovjvo
ovooNenTj- vovovjcoo eN"^1^^'*
r— TJ- en en en ONONON— ••— » OOOONO
•^•vjvjvivj VJVJV>VONO VOVDVOVJVO
t loo'ineN r-cNen^-co ^COTJ-OO i
'r*-r-co co "-• «— > vo m mvjr^i— « *
mvjin invivjvovo vovovovj
1 1 vo eN ON co- — cn ON CN ^cnvoco •(
vjvjvj •— « O t~* en <— • ^Ooco
^ _; ,-i VJVlvdONON ONONONt"-
vjvjvj vjvjvjinin mvjvjvj
c^cM^oiSiS voinmmi£ i£i2S^°
d>ua>u a>
CQCQCQCQCQ CQ CQ pQ OQ CQ CO CQ CQ CQ CQ
sssss sssss sssss
62
-------
§
Cfl
I
•3
c
§
1
8
.S
OS
rj
i
s
Os
O
i
a
measurement
>8 01/27/99 03/05/99 04/08/99 OS/13/
CM r3
O Q
^ f4
Q 00
^
S
s
I
00
i
o
S
10/08/97 03/12/98 OS/06/
VJ
vo
Os
o
VJ
s
vo
CO
vd
VJ
CO
vo
VJ
OO
CO
CO
«n
VI
»
in
VJ
t-*
VI
oo
in
Os
d
o
OO
o
VO
1
*2 CQ
d-^r^oo os r- •'*'<* vj r^ — • os vo o. vo d r-» [ i-~ o icor*- vjcor^cNvj t^-ot^-t^vo coTj-r^-
ooosoo •— >— ir VD m >— • CN co o co coosot^o r-t^oo
vjinvdco coco^coTt r^vdood— ' — • *— • os vo* d — • »— • \d
vjinvjvj vivj VDVJ cococo'^j-^f •^•^••^vjvj vjxn^
O •— • r^ CN •^•VD>vVJCO VjVJCOCOTt -*J-OSCOOOO OOOO
vqvjr^-^t TJI'*(%C1™ ^ °°. ^ ~". ^ ^^^"^^ ^^P
vdvdvdco coco^covi r~^vicodd" dd'osvdd dd'r-^
vivjinvj vjvj VDVJ corocorj--^- -^-tt-^tvsvj invjTf
vpvoind OSCO>»OSCN — -mosvoos cNvor-t--co covjo
-TCOVOCO r— co ^r r* oo •^••^tt^Tj-cN cocoosvod r^r^-^f
r^r^rVco roco^dvi i-^vdoddd* d"d os» vd co CN CN t-^
V)VJV)ln VJVJ VOVJ COCOCO'^-'^- Tj- -^ TJ- VJ VJ - VJ VJ Tf
— co d os vod>>vjt-~ cooooscoco vj — co oo TJ- ooco
^^"^i^ °^ " o — ; "l oc?vc3aiT:tcn ^ "^t ^ ^i "^ ^^^
odoocoTj- TtTt^vdvd c^vdcdcNd dd'ovdco d*di-^
vjinvjvj vjin vovj cococo^t^t ^tTj-invsvj vjin^t
CO VO 1** O OO •— • J? ^VJ VSVOCO^i— • i-Hd^OOO OSCOO1
vjvjvjvj vjvj VDVJ cococo^j--* •^•^••^j-vjin invj1^-
coosvoos TTTJ->,CO— cor-osO'— covodr-co cO'—oo
Tfr-;r^oo cMoo*rcNin mvoos-^j-d d vj vj vo vj co co co TJ- ^ ^ ^f ^j~ in in vj in ^"
O oo d — • •^•osp-.Tf— ^-i o rr -^ vo coocococo osoo
vjvor-co coco^co— • vovoosvjco co-^-dvj— coosco
r^r^Kvi viin1-1-^^ KVOOOCNCN d'd'or^co cotood
vjvjvjvj vjvj VDVJ cococo-^t'^- -^t--^-mvjin vjvj1^-
dor*-d ovj>>cooo ooodd^t VD^-^-VJO m-^-co
r— coosi^ coos^rovo vocoost**m vjvococovo d d •— •
f^odr^vi vdvi^vir-- r-^vdodd'd cNCNor-^co •^••^j-os
os i— * in r~- vo *-* vo o oo co v> ^3* r^ o co vj co r~~ os ^^ to oo
odooodvo vo "sf vd co co covdosdd CN d" i— • co ^3- -^ ^ os
vjvjinin vjinviVDv) cococo-^j-^ -v>
ddco— Tj-^j-O'-'cN d-—ooTj-c^- o-*co— »nin
g 1 1 1 ' 1 I 1 1 1 | 0- g> Ng | | | | cj | | | | ' |
d t*^ Os co os
VO CO CO -^t VJ
£ 1 ! 1 1 1 I f t f 1 ~ 1 1 I 1 ' I £ 1 III
CO Os Os
V) CO VJ
QOVOCO osor^r^-oo os vo ?*• oo >— • dcO'— «,— ,d vj-^vo
tO 4> (D CD 0) (D ID 4> d> (U d> a> SJ 1> qj n) 4J tU (D CL>
U3 Ul tlJ frl [j3 yj M [rt p] pi fri [ji £Jp pi (ji [T~t p[3 [ji po [j] yi yq
63
-------
Appendix D
64
-------
"O
e
a
£>
42
^
o
O
•+— *
c
<5
of
2
f_*
•o"
5
55
>
o
0
•4-*
3
o
Jj
o
•S
_g
i
fl>
I
+->
5
<1>
w
o
£
<§
*o
o
'o
0
CO
*! S
u S B * •J
V U S ^* m
jg. 3 'K E w
1-i'it
B vi
oT Jg
u fc w
22s
S" ^J,
I-s
- t) B „
'UsT^ "S
STT! o s
Q fe « OT
b O V -
S 'S ° "
rt -Q »*•
^ "S 3
t> CU V3
'•C "*" 13
•8 tf i
1/1 S- S
t^ aT a>
o w va
£ » f «
g -o « —
M ^J
QJ M 4*
n O. "5
*"* S M
u51
tl
ll
Q
I
0
CS
u
o
o>
•Q
|
Q
B
e
•a
S
ir> vo t- m c*> f> "* — . '
<^- \o *O VO vi t^-
oocootnoNO^-
^d t-* oo" \o t-^ o*
\O O CS t** CS CO CO CO
VkCSlOVONO1— «^J-— •
ooN^^r^irJ"^^
— orov> — covot-
covoto^covocooo
CO
^ ^r o .^. ^ °2 2 o
M ^ (S — ^ — ~
^^ ^^ oo ^o ^O ^O O ^
g~ggg|g»
§§55§«5§
oooooooo
sssg§,§§2
cSfSfSfSmmmcS
2SS22222
— ' CS — • • CS CS CO r>oo— '— '
^Ili^^is
oooooooo
f^r^ooooosCTloooo
ssssssas
U^VOO; — — Ol-mooo-
to vi ^"7 ^x ^^ . . ^^ .
COONVN o^rot-omt-.
00 ON VO 10 "^ CS CS" CS* CO
'
^* ^ Si "^ C? JS ^- ?^ S S
r^- ^-. ^- u~> Tf CS "^ CS "™
— CO ON 1 1 1 I I I I
VO -Sf CO
OCOOO "OOOOOOOOOOOOOO
coTj--^t 5csco^j-v^\or^oo
g. - - - « -
•o
I
M; 1 H
C5 ,jj 5
PJ g a
E
g|l
l.a s
§n.
_B
SB
B ._
5^ »g
cf fc*
.i .SP o>
cc '£ "
4>
»5 &•
C B fc«
.3 *- -2
« = *•
-S T3 W
O C O.
ON g-
H
| 2|
•3 S i.
«I S.
•* "3
E" S
s § u
Is j?S
o£« t>
S C O-
-
S
J &1
o s
•3 ^ s
^ S B.
•S
"H.-O
11
Q
.
fe
A
\
I
O
P^ — 00 — O^colo1 ' '
^ r— .
0000*0000 I 1 1
CO "-• O *•"" CO O "^ O
— . v — V o V ON
^f VO — •
c^ — -^j-t^.ooor--oo i | f
— * — CS^CSCS — CO
C-^ON'^'rocO'^'ON^' 1 1 I
CS - CS ^ CS CO Tf
-TfONcooo — r^csvi i I I
r*- ^" cs ON co oo co
r^-^j-i— 'ON— •t~-^*'O | ! I
*""* cs \d TI- cs co cs
cs— *c^o^odoo" — "*
oooooooooooocooooooooo
ON ON ON ON ON ON ON ON ON ON <^
CSCSCSCSCSCSCSCSCSCSCS
ooooooooooo
^ iTi ON CS ON O "O ^ CS r*- OO f-I
ONONOOONON^^— «^-OO Jg
O4>4>1>4>4>1>4>3J4>4> -*S
tnoaoa«m«cQcnfflcaoa «
aaaaaaaaaaa-a
V"
. I
g
1
-S
Q.
1
CO
O
C
r
«~~.
65
-------
3
^
c
o
o
!
o
1
e>
a
oo
I
O
8
.§
OJ
a
i
cu
5i S3 Si'-.
< =3 S
1 •
If!
z"0]
in .-
-win.
Z °
S o«
.¥1 s-
.
« 2
| Z
S 3
"• .55.
1 12
?•§ g" s s
"5*1
II'
ilk
<•§
Jl
^oooooooooo
O\ l—» OO *-^
i i i i i
i i t i
i i i i i i
t i i i i i i
. .
000
i "z
1 1 I 1 I I i
0
V0 tf) O
00-^
OOCOOOOOCOCOOOCOOOOOOO
^N ^^ ^ ^^ ^N ON O\ O\ O\ O\
^" ^™
^ o o o ^ ^ o
o o o o o o o
ro Tt- in so t~~- oo o\
66
-------
Appendix £
67
-------
CO
Si
3
3
l-t
J
E
CO
I
1
JS
o
w
SB J3JH 43d
suiEjSjugui ui
snisjao sooaSsp gj
SU3UIOIS04DJ1U UI
'30UE)DnpUOD
oupsdg
sngspo SD3j3op ui
PUO33S JD(1
pUOODS43d
S40J3UI ojqna u;
§
1
Si
o a
T: §
lo 1 o —
-
ON — od o\voood —
\O O OO —•« VO *^J" t"*-
o q o o o o o o a o ^; ci cs ~; u ^. ^: A — ^ ^.' ^ « ^ 5
OOOOOOOOOOOOOOC5OOOG5OOOOOO
— — o — o — o —
OOOO^OOOOON^OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
o o o — o o o
oooooooo
a
S"cLf
S
^^^^^^^^ >:>>>>>>:>>>:>
oooooooooooooooooooooooooo
1
. I
!
68
-------
sc J3ji| jad
SlUEjSl|HUl Uj
s
f
00
PQ
1
"3
I
a
w
I
SJIUIU pJCplIEJS Ut
__ _ ON oo ON "" ON
'USSAXO p3A|OSSIQ
SniS|33 S33J§3p SI
JB J3|3UIpU33 J3d
SU3UI3JSO13IUI UI
o r-^ —:
-- - M cs — —
UI
puo33Sjad o%>oTf'*^--*Tr-a--^-ro^oro oo1^ - - -• - o d
)33j siqns ui
'3§JEIJ3S1(J
oc^oooooooooooocnooocJooocJoo
S43J3U1 3iqn3 Uj
LiEipsi
gl.
1
— SoSoSooooooooS — — — — 0060
11111111111S111 i 11111 § i I ?
g !!!!!!i!iiS!!!iS!!i!Si!gg
es
e
c
o
•c
3
CD
!i
II
ooooooooooooooooooooooooo
69
-------
T3
|
•a
I
I
£
-------
•o
I
00
t
I
|| 11
«!glo.
s" .£ to
&s 2 is
111 *
£•
If 11
" — .S°*ju
«
T3
a>
f
w
B
M
1
P .
"I fe
P
ON
OO
—
CS
CO
OO
r^
co
^
v->
«
§
^
o
Q
i
1
H
COW C NR CHES
OS
O
ro
Os
0
— o
— CO
CO CO
Os OO
r~~ cs
vo co
co in
OS OO
r~- vo
1/0 -*
Os O
Os O
S §
£2 5
ON CO
O O
Q Q
S 2
aS w"
g|
C^ prj
E— 1 f-^
COW C NR CHES
COW C NR CHES
ON ON
O O
CO CO
ON ON
•3- T
O 0
1 NO
NO CS
*3* ON
•* CS
ON —
ON OO
CS OO
CS •*
•* "3"
NO NO
VN. OO
OO OO
ON ON
ON ON
1 1
O O
Q a
S 2
3 3
CHESTERVE
CHESTERVE
z z
09 CQ
CHESTERVEXE
CHESTERVELLE
o o
CO CO
ON ON
0 O
^j.
OO
„_
cs
rt
•0
oo
vo
t^
Os
Os
CO
S
Q
S
i
&
CQ
CHESTERVILLE
o
CO
O
r~^ t~~
•— •
o cs
in os
Os
CO OO
cs
r- vo
\O if)
VO m
O oo
1 1
o o
Q Q
5 %
CHESTERVE
CHESTERVE
1 5
CQ CQ
CHESTERVELE
CHESTERVELLE
o —
ro CO
Os Os
O O
ro ^J-
— vo
— CS
ro 0
CS
CS vo
vo r-
OO CO
ro co
Os CO
m —
oo oo
Os Os
1 1
o o
Q
Q m"
S3
^ >
CHESTERVtt
NR CHESTER
5 15
CQ CQ
CHESTERVILLE
CHESTERVEXE
o
_ —
Os Os
O 0
— OO
cs —
t~- o
CS •—'
— 00
^
OO CO
CO
>o in
co in
oo oo
Os Os
Os OS
^2 —
l§ X
O O
Q
.
"| Q
> 2
NR CHESTER
CRUMPTON,
H Z
CQ 09
CHESTERVEXE
CHESTERVEXE
CO
— cs
CO CO
ON ON
0 O
1
1
1
1
1
1
1
OO
2
<3
i
o
i
CRUMPTON,
&
CQ
CHESTERVILLE
rs
ro
O
cs r*
r^- oo
*—• "~*
OO O
r~~ os
ro »O
t^ OS
\n oo
ro -^r
CO Os
os r^
"o in
Os Os
OO OS
Q CJ
m vo
O O
a Q
2 2
CRUMPTON,
CRUMPTON,
oi oi
Z Z
09 09
CHESTERVEXE
CHESTERVEXE
cs cs
CO CO
°r ^
0 O
1
1
1
1
1
1
1
oo
ON
ON
1
O
Q
CRUMPTON,
oi
CQ
CHESTER VEXE
cs
1
o
1
1
1
1
i
1
1
OO
ON
ON
35
G>
Q
CRUMPTQN,
Z
03
CHESTERVEXE
cs
CO
ON
0
NO
•"*
O
oo
06
00
CO
^
v>
"*
NO
OO
ON.
Q
Z
%
oi
Z
09
CHESTERVEXE
cs
co
ON
0
1 0
*— '
1 0
cs
1 ON
1 — .
^
I ON
rj-
1 CO
Ti"
1 "^
OO OO
ON ON
ON ON
rsl —
» g
*o o
Q Q
2 S
CRUMPTON,
CRUMPTON,
oi oi
Z Z
CO CQ
CHESTERVELLE
CHESTERVILLE
CS CS
co ro
Os Os
O 0
CO O
"^
o rs
ro in
Tj- OS
OO
r- t^
CO CO
^. _
*n in
\O vo
^r ^r
in in
oo oo
Cs Os
Os Os
~- oo
cs o
S °
Q Q
2 2
CRUMPTON,
CRUMPTON,
Z Z
m 09
CHESTERVEXE
CHESTER VEXE
cs cs
CO CO
°r ?p
o o
1
1
1
1
1
1
1
oo
ON
ON
5;
O
§
CRUMPTON,
oi
CQ
CHESTERVELLE
CO
CO
O
1
1
1
1
1
1
1
oo
ON
ON
§
Q
2
CRUMPTON,
oi
CQ
CHESTERVELLE
cs
CO
ON
0
1
1
1
1
}
1
1
00
ON
35
Q
CRUMPTON,
Z
CQ
i
I
CS
CO
ON
Tf
0
o
^—
CS
OO
ON
NO
CO
NO
V)
^
*
NO
oo
1
Q
CRUMPTON,
Z
09
CHESTERVEXE
(S
CO
ON
O
1 1-
CS>
i 0
"~
1 0
1 CO
CO
1 rl-
*rt
] •*
*
i >2
oo ON
ON ON
ON ON
i 1
§i
CRUMPTON,
CRUMPTON
DH (X
Z Z
09 oa
CHESTERVE.LE
CHESTERVEXE
rs cs
CO CO
ON ON
O 0
71
-------
T3
1
8
!
"p.
CO
6
*V
o
r— t
Q
42
Chemical ,
w
1
S
QJ
is s "
C g
B-rf 1 s
i s §>s
-1 2 £
K ij^ f*i
=5 E
.2 10 53
J| 2 E
i g S.
J" ** to
E ~ P u
a -o i «
1 s £s
Bgg b
js „
"o "3= fe
CO jo -g B.
if B 10
§ 5 " «
fif !
a
Cj ^ *S ^* '
41
"3,
t
CO
a
C
s
a'
£
g-g
li
w
IOVOI 101 1 | | lOOIOsCsl
en oo cs c- oo oo
Igo' 'So' ' ' ' I <» i f; •* i
loscnl Icsl I I I | — |
en en 10 -^j-
lencsl lenl 1 I I \ u-> \ t~-_ — \
"} V) V} «n ^J- to
1 "} vj 1 !•*! 1 1 1 lenlessql
"* "* * •*'#•*
1 — 12' '21 ' ' ' lr~lenT)-|
ON ON ON o\ ON ON o\ ON ON ON ON ON ON ON ON ON
ON ON ON ON ON ON ON ON ON ON ON ON ON O\ O> ON
ONCNONONONONONONONONONONON.ONON.ON
^^S^^o^tnr^oo^vB^cnxS^fo^
C^ O O ^D ^5 O C^ O O ^3 O G> O — ^ «— * •— *
QQDQQQQQQQQQQQQQ
555555'SS:'5'S^'S>3'5^;1^;
^rfS^a^^^^:^;^^;,^^;,^;^:^:,^;
222292PPPooooooo
!r5r'!rSrt'fcHf"t'^"f-'t"f-i^&"f~i-1
CUCU[XiCUCUCUCUCUCXtCUCUCUCUCl«CUClH
^^^^^^^^•s^^^islsAfA
DDD3.3DDD35DD55ID3
oleioiajGialoiQioicrfciSSoiSS
ouuuoououuouSouo
&*Q£Q£C£O£C£KI(£Q£C£Q£&£CXZC£ c£ ^
•y-y^y^-^lyy^^r^^^y^^:
SQCQpQpQpqpqpf^p^p^fQCQrQfyjp^pqpfj
StQWWpJWIiJijJtUttJtQtQjiatQijjfij
iHHHHHHHH'H'^^H'-^^H^
ddddddddddddrft^dd
>:>>>>>>>>>>>>>>>
oioic^oioic^erfeiioierfoicsicrftticdM
ywwwwpjLjjwtatiiujtQEtQtxiw
"" * c"* c™* r^ E"^ c™* r* E"^ r™^ E"^ E~^ E""* t~* E~* E*^ E^
WC/JCOCOCOCOCOCOCOCOCOCX3COWCOCO
^ PI P^ T-T4 [T'7 tU rT1 tTl fT1 fT1 fT] TT1 W PI TT1 fT1
EErCX3C2CffiEDCE3CEEE3CEffi
uouuuuocjoouuoGGo
SdtNescscsr-»CNicSfScscNC4r^(NfS
•nfncncncncncocncncnmfnmcnfnco
7^ O\ ON ON ON ON ON ON ON ON o\ ON ON ON ON ON
**T ^ ^ Tf ^- '^•Tj-^^J-^J-^-'^-'^J--^- -^- ^jj-
OOOC^^OO^OOOOOOOO
Ifsr-r^cnvofs-^-ooM— • P
— ^212^^:^22^^ — S 1
•—•
lmooooot-~ooomo !
t— •* en CS •* so "» I
1
oooooom'S-rrt-Oscsints '
" ^ Cs so "" t^- CS "" "^ so »
IcNenenenen^— — ^(-Tfi/->oo
enensoos^r^y^-^-odoo'en !
,
[
1-OscSOso^-en-enso
|n>/Soo'.r>Tl-v>soenio ,
'
, ^ „ ^-sooosenos^ I
*"" i
!
ONOOOOOOOOOOOO<»OCX)
ON^OON^ONON^ONONCDON I
ONOOONOONONOONONOON t
oo^SvS^^^S^ — S" '
iliiliililll
i
I
Q Q Q i i i :
S ?! S O O O •>*... Q i
2? 15 ^ '---Hj^j^j '
2 S 2 ^ ^ [v^ p^ j^ Q Q Q ^ |
C^^^^^IDOO^^^^ '
JS^^^^CJOO^^*^ •
fQCQja^^^OOOoio^oiPj i
>>>UUCJUOUUOU
sc § S o'o o o o o o o o i
oDS'SSSSSSSSS i
^•^ *~* "^ ON ON ON ON o\ ON ON ON ON '
t^M,—,^}.^.^..^..^.,^..^..^..^.
g^gjmcncnfnmfnmmfnm
'^••^•Tf-^fTf-'^'rf'^'^'^-'^-1*^- ,
oooooooooooo
72
-------
T3
I
1
O
S
"&
I
«
m
>-.
~ca
3
J3
O
w
I
I
3
oT B »>
v '"•' H i*
c "S 2 ••§
If "l ~ °*
11 E *
.S e
-•a' « J>
-Us.
=3 E
c „, 35
"1 B »
II ii
•s 1 fe
^
.£• .S ui
£ ~ E i.
s •» 1 .2
1 " !> =
I si |
a, 5 S
III!
cc jfl ^ a.
H *** e
1 1 ||
e „
If II
'3 S en1:"
3J1&
»
1
f
I
e
•c
M
2 i-
fi
v>
g
™*
CO
CO
*0
oo
VO
oo
T^-
m
oo
OS
Os
c>
Q
MNEDYVILLE, M
i
s<
o
1
i
CO
ON
O
CO
CO
04
04
t-;
t**
r-
VJ
—
^_,
x^-
_
o
0
CM
O
Q
MNEDYVELLE, M
§
%
CJ
1
S
CO
Os
O
Os -S-
OO — *
CM --a-
So
CM
VI
*"""
ro »—
v-j
OO VO
o
o o
*r> co
O4 CO
O (-;
w
VO OO
v> co
OJ vo
vj co
oo v>
oo *«r
O OO
O ON
O ON
r5 ^?
M
U
ii
£ 2
O CJ
§ S
ii
o
04
OO ON
ON ON
CO CO
ON ON
•3- ^j-
O O
0 J
04 0
VO ON
CO ^f
"" 06
ON TJ-
^
r—
I 00
"O
1 cs
i •*:
i —
OS O
OS O
5= §
(S CM
S S
D O
MNEDYVILLE, M
MNEDYVELLE, M
s a
2 2
erf erf
Z Z
U O
S §
ii
493500
493500
O O
73
-------
•is
I
I
S3
I
E
a
A 2°-
!~JE
5-fbS
3 J 3 ==
5°-l|
oT „
s§ Is
w
5 •»
- g-o | fe
a ff S &S
Sfilis
_B
. .«» If i
I— e
> c •—
at _!• '
o&5"g|
g g os OUT*
1-llS.
"'••S.B 5
1
.HI
§0 = Jj
8=5 fi.
=3 B
- 2
£TT|S
^ J5 & =
3 ° S S
«Ji &•
*t3
^"•2 a
If II
1-3-fe
3 S = o.
5 T3 C
c
g
1
OT
1
ta
ON r^ ^3
o p p
o o V
o o
O C5 CS
o o
es NO m 1^- oo o o
p — ppppppppp
o o v V o o o o o v V
— oo -^j-
O ^- ro
V o" o
— o
§ o
V
2 2 — —
o o o S
o V V V
o —•
p
o
o p p
V O V
o o o
O O O O
V O V
t^- os r- oo
r- ^- o oo
—•csos — — os — r-»
<-• — Os — — Os —
co en O O ^J-
cs ^ — — ^r
o" o o v" o
O •*
in in
r- oo »n
oooooooooo
— — cslmcsl
O O O O O O
~M —. vq
OS ^™ OS
I m
t I I
OS OS ^D OS OS OS
OS OS ^5 OS OS OS
oo oo oo oo oo oo oo oo oo oo oo oo oo oo oo oo oo oo oo
OS OS OS O\ OS OS OS OS O\ OS
^ «"^ M ^ •"* CO ^" "^ "^ ^ ^^ ^ O OO
—«^O — C^"-CN — CS — (S — rocN
ssssssssssssss
Q Q
OS OS OS
oo ON r^
£3 e> '
S S £
^- oo os ON ON O\ ON O\ O\ Os
ooooooooooooSooooooooooooooo
74
-------
• <* *•
|.s i s
||||
O eg.
Vt v,
"3 e 2 ™
K en
•3 § '•=" 1 1
.2 e» S> £ •—
o ® « w) —
^ £ 1 = b
g .5 ">
B. S "B" 1 S
1 1 ess
AS s -w •= O.
Z "S E
..
« •£ .5 Q
'S « £ a S
g £? 5 2 *!
| o g>«I
E ") j2 a «
B.
!•»" 1 a
i .1 «=
SlP
T3 C
.5 »
If II
*+* > COD "•"•
1|| 1
s c »>
•£ - •£f':I
.allS-
CQ ^
5
«
T3
V
"5.
t
«
.
il
1'
o
s
V
CO
0
cs
vo
oo
o
od
»o
v^
o
1
1
1
a
ON
^
o
Q
s
CRUMPTON
CHESTERVILLE B NR
cs
s
o
CO
8
o
2
o
oo
00
ON
"^
^-
U1
O
CO
vo
T^-
oo
cs
ON
ON
VO
O
Q
s
CRUMPTON,
CHESTERVILLE B NR
04
S
O
CS
o
o
o
o
no
00
\O
cs
OO*
„
CO
o
J^
cs
•*
cs
cs
ON
ON
^
o
Q
s
CRUMPTON,
CHESTERVILLE B NR
cs
1
o
1
0
i
o
1
1
o
V
1
1
•3-
CO
ON
ON
vo
f
o
Q
CRUMPTON,
CHESTERVILLE B NR
cs
s
o
o
o
V
o
o
oo
o
vO
OO
in
oo
CM
O
1
1
CO
CO
ON
ON
^
§
o
Q
S
Z
OS
o
CHESTERVILLE B NR
cs
CO
o
CO
0
CO
o
0
vo
vo
5-
vd
,-J.
cs
o
CO
r~
CO
^
CO
ON
ON
ON
s
o
Q
s
CRUMPTON,
CHESTERVILLE B NR
cs
CO
o
CO
8
0
CO
o
o
r-
oo
vd
cs
rs
o
1
1
1
ON
ON
p;
Ti
O
D
S
CRUMPTON
CHESTERVILLE B NR
cs
CO
o
i
0
1
o
Tf
>n
E
"S
cs
CO
o
1
1
Tf
TJ-
ON
ON
So
1
o
Q
S
CRUMPTON,
CHESTERVILLE B NR
CS
1
o
I
o
0
oo
in
^
cs
CO
o
1
1
1
ON
ON
o
o
Q
S
CRUMPTON,
CHESTERVILLE B NR
CM
1
o
o
o
o
o
ON
•"3-
VO
"*
in
co
o
1
1
0
"*
ON
ON
VO
O
o
Q
S
CRUMPTON,
CHESTERVILLE B NR
cs
1
o
^3-
O
O
C)
o
t~-
CO
ON
CS
CO
cs
o
1
1
^.
'*
§N
ON
•
O
Q
CRUMPTON,
CHESTERVELLE B NR
CS
1
o
0
o
o
o
wo
ON
V~t
o
cs
oo
cs
CO
"""
ON
ON
CO
I
o
Q
S
CRUMPTON,
CHESTERVILLE B NR
CM
1
O
•i
O
-.
ON
ON
0
vo
C>
ON
^
O
[
a
ON
vo
CTv
O
Q
CRUMPTON,
CHESTERVILLE B NR
cs
1
o
s
o
0
CM
•^J-
O
"*
oo
CM
O
CO
oo
^
^
CO
&
ON
O
o
Q
S
CRUMPTON,
CHESTERVILLE B NR
CS
1
O
o
o
V
CO
o
o
oo
"*•
CO
cs
o
CO
vo
•n
^
CO
ON
ON
CO
i
Q
s
CRUMPTON,
CHESTERVILLE B NR
cs
1
o
i
CD
o
o
oo
vo
VI
ON
•n
ON
O
I
1
ON
CO
SN
o\
0
~*
Q
s
CRUMPTON,
CHESTERVILLE B NR
cs
CO
o
i
o
CO
O
o
vo
n
CO
n
—
CM
CO
—
^"
^
CO
0
o
o
CO
1
Q
MORGAN CTRNRG4
ON
CO
ON
O
S
O
—
O
CM
ON
CM
OO
cs
^J.
^J"
o
CO
CO
oo
ON
ro
oo
ON
ON
5
1
Q
ST GROVE, N
MORGAN C NR LOCU;
n
CO
o
•o
o
cs
1
Q
Elf
NNEDYVILL
MORGAN CTRNRKE
0?
CO
ON
O
CO
cs
o
o
CO
o
CM
CO
oo
CM
(^
vq
O
^.
CO
vd
oo
oo
ON
ON
-------
•o
u>
3
C
o.
CO
CQ
<<-i
o
eo
S
1
6
I
I
•L tg ™
J-S 1 3
fffl
tn vt
c E ^"
"go & £
o cu« ,oy "
•fi "p ^
a
•* *"" f? *
3 | -g | a
s2||f|
a ~ S
*5> ** ^ a ™
IT'S g 2 —
S i: ">.g?T
£ -g g = s
2 "5 £
M ~ B a
"5 n. B1 a a
0 M g 5 ±1
llfll
o. c
^""L § i-
!i ^^>?
ol| S,
c
r "^ ^ *•*
« 1 2 S
Is « B°T
^IfS,
=3 B
~— a
Jl'liJ
" "S= h
S Bl e.
pa -3
•£
et
•o
6>
R
CO
i
S
1
cl
a ^
n S
yi E
OD
rvi vo
o o
o o
vj r~
o
« cn
04 CN
g S
JEDYVILLE, MD
JEDYVILLE, "MD
g g
#C< X-*
a a
£< 5
O 0
o o
1 i
t- r-
ON ON
en S
ON ON
Tf Tf
O 0
ft vo
o o
o o
CO O
— 01
o o
oo cn
« Tj-
"* 2
Sm
vo
cn ^
o
Tj- OO
o o
o —
^- r«
-^ cn
oo cn
»n oi
TJ- OO
OO OO
ON ON
ON ON
»n —
i 1
Q Q
5? 3
ENNEDYVILLE, 1
ENNEDYVILLE, 1
3 3
S S
O O
§ S
ii
oo oo
cn cn
Tf TJ*
o o
VO vo
O O
o o
— vo
.— . vo
01 0
O 0
— 01
ON ^3
>n o
« 01
^r cn
Tj- OO
in t-^
o «
Ol Ol
o -^
"™* t~»
ON ON
m cn
oo oo
ON ON
ON ON
S ""^
O O
ORGNEC, MD
ORGNEC, MD
2 2
oi oi
& &
O O
0 0
oi oi
ii
O O
OO OO
cn cn
•^f Tt*
o o
o oo
o o
V O
IO VO
*o — •
o o
VO 01
ON ol
•^t- oo
S 9
Tf VO
ON oo
00
o
CN Tt
CN cn
to —
00 ~
ON ON
CS CN
O OO
8 §;
*-: C
cn in
o o
ii
2 S
oi oi
Z Z
H H
O U
0 0
oi oi
o o
2 2
o o
i i
TJ- -^j-
cn cn
ON ON
TJ- ^J-
O O
0 ON
O 0
V 0
OO »O
r- 01
O r-
O O
01 t-
to oo
10 cn
•^ o\
o -<3-
»o cn
oo oo
O 0
t-* vo
cn —
VO C-;
"~* vo
»O F—
cn cn
§00
ON,
cj ^t
o o
0 g"
^ ^
)-J £^
i i
^ §
0 U
1 1
S S
0
00 ON
cn cn
ON ON
Ttf" •*^-
0 0
oo «— •
0 S
0 0
ON ^-r
0 °
0
cn cn
f- ON
o4 cn
cn vo
cn cn
oi cn
TJ- 0-
o
r- —
— Ol
cn 10
c^
ON ON
*o cn
oo o
g; 8
— to
ON ?5
O O
Q Q
2 2
U
oi oi
ii
oi oi
o o
9 9
i i
ON ON
?P ?
cn cn
ON ON
^ ^l"
0 0
ON 0
O O
0 V
*
o «—
— 0
O 0
01 cn
ON — *
cn -<3-
- ^r
cn cn
— ON
O O
VO 1
-~
r- 1
VO
f- 1
m
oo oo
ON ON
•* o
O 0
•JEDYVILLE, MD
•JEDYVILLE, MD
Z Z
a a
Z Z
U 0
o o
o o
O O
O O
n
— 0 O
O O 0
OO OO (S
vo O ON
CN cn CN
vo ON vo
CN VO VO
CN CN CN
CN ON vo
0 O O
1 vo |
1 — 1
cn
1 vo |
oo oo oo
ON ON ON
ON ON ON
r- — • ON
(N JN ^
OO ON O"
00 —
4EDYVILLE, MD
•JEDYVILLE, MD
4EDYVILLE, MD
g g g
oi oi oi
u u o
§ § §
iii
o o o
o o o
VO
-------
Appendix F
77
-------
Appendix F. Basin Characteristics of Sites from the Base-flow Synoptic Surveys Used in the Landscape Analysis for the
Locust Grove Study Area, Kent County, Maryland ;
Site ID „. .
number , Station name
01493491 MORGAN CREEK TRNR GALENA MD
01493495 MORGAN CNR LOCUST GROVE MD
01493496 MORGAN CTRNR KENNEDYVILLE MD
01 493 497 MORGAN C AT KENNEDYVILLE MD
01493498 MORGAN CTR AT KENNEDYVILLE MD
01493 49810 MORGAN C TR NR MORGNEC MD
0149349820 MORGAN C TR NR LYNCH MD
01493 499 MORGAN C NR WORTON MD
01493 500 MORGAN C NR KENNEDYVILLE MD
01493 109 COW C NR CHESTERVILLE MD
01493 1 10 CHESTERVILLE B NR CHESTERVILLE MD
01493 1 1 1 CHESTERVILLE B AT CHESTERVILLE MD
01493 1 1 1 10 CHESTERVILLE B TR NR CHESTERVILLE MD
01493112 CHESTERVILLE BNR CRUMPTON MD
= •5
if
11
Site ID „. ., £> =
number Station name *. &
|g
& S
§1
53 2
a. -a
01493 491 MORGAN CREEK TR NR GALENA MD 1 1 .43
01493495 MORGAN CNR LOCUST GROVE MD 13.52
01493496 MORGAN CTRNR KENNEDYVILLE MD 6.45
01493497 MORGAN CAT KENNEDYVILLE MO 12.78
01493498 MORGAN CTR AT KENNEDYVILLE MD 19.18
0149349810 MORGAN CTRNR MORGNEC MD 13.81
0149349820 MORGAN CTRNR LYNCH MD 42.27
01493499 MORGAN CNR WORTON MD 25.53
01493500 MORGAN CNR KENNEDYVILLE MD 14.67
01493109 COW CNR CHESTERVILLE MD 33.88
01493110 CHESTERVILLE BNR CHESTERVILLE MD 42.66
01493 1 1 1 CHESTERVILLE B AT CHESTERVILLE MD 32.40
0149311110 CHESTERVILLE B TRNR CHESTERVILLE MD 36.17
01493112 CHESTERVILLE BNR CRUMPTON MD 32.36
Drainage
area
(square
miles)
0.45
3.18
1.33
6.16
0.99
0.32
0.54
10.62
11.99
0.35
1.29
4.74
0.93
6.12
Percentage of type B
(moderately well-drained)
soils in watershed
Percentage of type C
86.45
74.11
72.86
65.69
49.86 '
41.77
42.74
5&38
58.77
59.50
48.59
51.18
49.78
48.64
Regional Local Stream
basin basin length
index index (miles)
(moderately poorly drained)
soils in watershed
0.18
6.31
12.76
12.10
17.39
24.32
11.81
15.63
15.42
3.55
3.98
4.93
6.53
7.93
-2.53
-2.08
-2.14
-1.87
-1.76
-2.35
-2.40
-2.13
-1.76
-2.59
-2.31
-2.06
-2.06
-1.98
Percentage of type D (poorly
drained) soils in watershed
0.00
, 0.86
3.28
2.93
5.77
3.14
0.00
1.56
3.51
0.00
0.09
1.46
0.42
1.40
. 1.75
0.03
0.39
-0.81
-1.19
-0.18
-0.22
-1.40
-1.57
1.28
0.92
0.07
-0.30
-0.76
Percentage of agricultural
land in watershed
Percentage of confined
100.00
94.27
97.66
92.23
92.49
94.32
96.22
90.60
87.74
100.00
97.32
94.56
92.60
93.07
[Stations 01493 500 and 01493 112, in bold, are gaging stations; the regional and local basin indexes are
Base Flow section of the report.}
78
i
0.35
1.94
1.03
3.10
0.95
0.51
1.04
3.75
5.38
0.53
1.06
2.19
0.73
3.64
animal feeding operations
in watershed
0.00
0.00
0.00
0.00
0.00
0.00
0.00
12.51
0.45
0.00
0.00
0.00
0.00
0.72
Minimum
altitude of
stream i re<|m
(feet above ***
sea level)!
55.00
35.00
40.00
25.00
20.06
45.06
20.00
10.00
2.00
40.00
40.001
25.CK)
i
30.00
5.00
i
Percentage of nursery
in watershed
JPercentage of urbanjand in _ _
watershed
0.00 0.00
4.49 b-00
0.00 11.74
2.24 J3.77
0.00 0.00
0.00 j).00
0.00 i.31
0.00 0.00
1.19 1.32
0.00 0.00
29.95 0.00
26.49 0.00
18.20 6.00
21.17 0.00
14.41 '
12.89 i
14.59
11.31 •
36.73 I
19.46 '
38.65
12.00 ;
10.77
38.02 !
14.11
13.71 '
41.15
13.73
Percent of forest
in watershed
0.00
5.41
2.34 :
6.85
7.51 '•
0.00
2.46 , '
3.81 !
8.21
0.00 ;'
2.68 :
5.10
7.74 -
6.70
explained in the Nutrient Yields from '
i
i
i
i
-------
-------
United States
Environmental Protection
Agency
National Risk Management
Research Laboratory
Cincinnati, OH 45268
Official Business
Penalty for Private Use
S300
Please make all necessary changes on the below label,
detach or copy, and return to the address in the upper
left-hand corner.
If you do not wish to receive these reports CHECK HERED;
detach, or copy this cover, and return to the address in the
upper left-hand corner.
PRESORTED STANDARD
POSTAGE & FEES PAID
; EPA
PERMIT No. G-35
i
EPA/600/R-02/008
January 2002
------- |