Hydrogeology_and_Water_Quality_at_Bent_Creek_Research_Station,_Buncombe_County,_North_Carolina,_2002_-_2009

i

Hydrogeology and Water Quality at Bent Creek
Research Station, Buncombe County, North
Carolina, 2002 - 2009
Prepared in cooperation with the U.S. Geological Survey, North
Carolina Water Science Center
Groundwater Bulletin 2011-01
N.C. Department of Environment and Natural Resources
Division of Water Quality
ii

Hydrogeology and Water Quality at Bent Creek
Research Station, Buncombe County, North
Carolina, 2002 - 2009
Prepared in cooperation with the U.S. Geological Survey, North Carolina
Water Science Center
Groundwater Bulletin 2011-01

By Ted R. Campbell
iii

North Carolina Department of Environment and Natural Resources
Dee Freeman, Secretary

North Carolina Division of Water Quality
Colleen Sullins, Director

For additional information or additional copies of this report write to:

North Carolina Department of Environment and Natural Resources
North Carolina Division of Water Quality, Aquifer Protection Section
Asheville Regional Office
2090 Highway 70
Swannanoa, North Carolina 28778

Suggested Citation:
Campbell, T.R., 2011, Hydrogeology and Water Quality at Bent Creek Research Station,
Buncombe County, North Carolina, 2002 - 2009: North Carolina Division of Water Quality
Groundwater Bulletin 2011-01, 107 p. (available online at
http://portal.ncdenr.org/web/wq/aps/gwp/groundwater-study-publications)
iv
CONTENTS

Abstract …………………………………………………………………………………………………….. 1
Introduction…………………………………………………………………………………………………. 2
Background and Previous Studies……………………………………………………………… 3
Purpose and Scope………………………………………………………………………………. 5
Study Area Location and setting ……………………………………………………………….. 6
Methods of Investigation………………………………………………………………………… 9
Acknowledgments………………………………………………………………………………… 11
Hydrogeologic Setting…………………………………………………………………………………….. 12
Regional Geology………………………………………………………………………………… 12
Local Geology…………………………………………………………………………………….. 17
Surface Water…………………………………………………………………………………….. 29
Groundwater ……………………………………………………………………………………… 32
Regolith flow system…………………………………………………………………….. 34
Properties ………………………………………………………………………. 34
Storage …………………………………………………………………………. 35
Bedrock flow system ……………………………………………………………………. 38
Seasonal and storm-induced groundwater level fluctuations ……………………… 41
Vertical head gradients between upper regolith, lower regolith, and bedrock
flow systems……………………………………………………………………. 46
Seasonal and storm-induced groundwater temperature fluctuations..……………. 48
Aquifer response to 72-hour pumping in regolith …………………………………… 60
Aquifer response to 45-hour pumping in bedrock ………….………………………. 71
Age………………………………………….……………………………………………. 72
Water Quality………………………………………………………………………………………………. 75
Groundwater Type and Comparison to Regional Water Quality .…….…………………….. 76
Changes in Groundwater Quality with Season .………………………….…………………… 78
Changes in Groundwater Quality with Depth and Topographic Setting ..…………………. 78
Changes in Groundwater Quality with Short-Term Rainfall Events ………………………… 88
Nutrients and Bacteria……………………………………………………………………………. 90
Volatile Organics, Semi-Volatile Organics, Pesticides, Herbicides, and Other
Constituents and Measures……………………………………………………………. 90
Naturally Occurring Radionuclides and Arsenic………………………………………………. 90
Comparison of Groundwater Chemistry to Rock Chemistry ………………………………… 91
Summary……………………………………………………………………………………………………. 91
Selected References………………………………………………………………………………………. 93
Appendices
1A. Geologic map showing lithologies and structures in study area, modified from
Merschat and Carter, 2002 ……………………………………………………………. 98
1B. Field reconnaissance map of geology in study area, modified from W. Burton,
written communication, July 25, 2008 ………………………………………………… 99
2. Geophysical logs and images showing structure and fracture zones tapped by
bedrock wells …………………………………………………………………………. 100-106
3. Well construction details ………………………………………………………………….. 107

Figures

1. Map showing location of Piedmont-Mountains hydrogeologic research stations,
including Bent Creek Research Station, and North Carolina physiographic
provinces …………………………………………………………………………………………… 3
2. Schematic showing regolith storage and bedrock fracture conduits in the fractured rock
aquifer system of the Piedmont and Mountains of North Carolina, from Harned and
Daniel, 1992………………………………………………………………………………………… 4
v
3. Map showing location of the Bent Creek Research Station in the Boyd Branch sub-basin,
Bent Creek Experimental Forest, North Carolina, and location of National
Oceanic and Atmospheric Administration rain gages 0246C and 310724 and U.S.
Geological Survey stream gages 03447894 and 0344789265………………………………. 7
4. Map showing topographic setting, well locations, and areas of recharge and discharge,
Bent Creek Research Station, North Carolina…………………………………………………. 8
5. Map showing geologic formations in the Ashe Metamorphic Suite and Tallulah Falls
Formation (modified from North Carolina Geological Survey, 1985) and physiographic
provinces of North Carolina……………………………………………………………………… 13
6. Map showing geologic belts in North Carolina (North Carolina Geological Survey, State
map, 1:500,000 scale, 1985) and hydrogeologic units mapped by Daniel and Payne in
Buncombe County (1990), as modified from Huffman and others, 2006…………………… 14
7. Stereonet projection of poles to foliations, fold axes, crenulation axes, and mineral
lineations within the Bent Creek watershed, Asheville, NC, modified from Merschat
and Carter, 2002…………………………………………………………………………………. 15
8 A. Photograph showing example of well developed, steeply NW-dipping S1 compositional
layering in biotite gneiss/metagraywacke on Blue Ridge Parkway about two miles west
of Bent Creek Research Station, North Carolina…………………………………………….. 15
8 B. Photograph showing example of S1 schistocity (compositional layering) in mica schist,
in northwest part of Bent Creek Research Station, North Carolina………………………… 15
9 A. Photograph showing example of locally dominant, SE-dipping S2 schistocity with
iron-sulfide precipitation, on Blue Ridge Parkway about two miles west of Bent Creek
Research Station, North Carolina……………………………………………………………… 16
9 B. Photograph showing example of medium- to fine-grained mica schist, S dipping S2
schistocity dominant foliation (strike at N83E, dipping at 43 degrees), transposing
N-dipping S1 compositional layering (strike at N87W, dipping at 50 degrees) to
produce intersection lineation on S2 surface, in northwest part of Bent Creek
Research Station ……………………………………………………………………………….. 16
10. Stereonet projection of poles to joints within the Bent Creek basin, Asheville, North
Carolina, modified from Merschat and Carter, 2002………………………………………… 17
11. Photograph showing example of interlayered gneiss/metagraywacke, schist, and
migmatite from rock sample at core site BC-1, Bent Creek Research Station, North
Carolina ……….………………………………………………………………………………… 18
12. Photograph showing samples from rock core sites BC-1, -3, -4, -6, and -7, with
general descriptions, Bent Creek Research Station, North Carolina……………………… 19
13 A. Photograph showing example of interlayered gneiss/metagraywacke, schist, and
migmatite in outcrop located on Blue Ridge Parkway about two miles west of Bent
Creek Research Station, North Carolina…………………………………………………….. 20
13 B. Photograph showing isoclinally folded migmatitic gneiss, S-dipping (axial plane) folds
(likely F2), Bent Creek Research Station, North Carolina………………………………….. 20
14. Photograph of sample from rock core BC-1 showing biotite gneiss/metagraywacke and
quartzofeldspathic leucosome, Bent Creek Research Station, North Carolina…………… 21
15. Schematic of near-surface geology in seven cored locations at Bent Creek Research
Station, North Carolina………………………………………………………………………….. 25
16. Photograph showing core recovery of regolith illustrating even, gradational weathering
at BC-3, Bent Creek Research Station, North Carolina……………………………………... 26
17. Photograph showing core recovery of regolith illustrating near absence of saprolite at
low-lying site BC-1 adjacent to Bent Creek, Bent Creek Research Station, North
Carolina …………………………………………………………………………………………….. 27
18. Photograph showing core recovery of regolith illustrating interlayered transition zone at
BC-5, Bent Creek Research Station, North Carolina…………………………………………… 28
19. Graph showing monthly rainfall amounts at National Oceanic and Atmospheric
Administration weather station 0246CA, 1.2 miles N-NE of the Bent Creek Research
Station, North Carolina, January 2003 to June 2009…………………………………………… 30
20. Graph showing mean daily discharge in Bent Creek at U.S. Geological Survey gage
03447894, November 2001 to July 2009 ………………………………………………………… 31
vi
21. Graph showing discharge in Boyd Branch at U.S. Geological Survey gage 0344789265,
Bent Creek Research Station, for period of record between July 2004 and December
2005, in relation to rainfall …………………………………………………………………………. 32
22. Generalized hydrogeologic cross section along well transect at the Bent Creek Research
Station, modified from Huffman and others, 2006………………………………………………. 33
23. Water level elevations observed in upper regolith wells on May 24, 2005, in vicinity of
BC-4, Bent Creek Research Station, North Carolina……………………………………………. 37
24. Graph showing maximum and minimum water levels as measured monthly between 2003
and 2008, and depth to water measured on August 14, 2008, Bent Creek Research
Station, North Carolina……………………………………………………………………………… 41
25A-F. Graphs showing monthly water level measurements in clustered wells at sites BC-1
to BC-7, Bent Creek Research Station, North Carolina, February 2003 to August 2008,
in relation to monthly rainfall……………………………………………………………………… 42-44
26. Graphs showing example of storm-induced water level fluctuations in upper regolith well
BC-7S (recharge area), Bent Creek Research Station, North Carolina…………………….. 47
27. Graphs showing storm-induced water level fluctuations in upper and lower regolith wells
BC-2S and -2I (discharge area), Bent Creek Research Station, North Carolina…………... 48
28. Graphs showing seasonal temperature fluctuations in upper regolith well BC-4S
and bedrock well BC-4D (recharge area), in relation to air temperatures, August 2008 to
January 2009, Bent Creek Research Station, North Carolina……………………………….. 49
29. Graph showing storm-induced temperature fluctuations in upper regolith well BC-4S
and bedrock well BC-4D (recharge area), in response to summer storm event, August 26
to 28, 2008, Bent Creek Research Station, North Carolina…………………………………. 49
30. Graph showing seasonal temperature fluctuations in upper regolith well BC-7S (recharge
area), in relation to air temperatures, May 2008 to June 2009, Bent Creek Research
Station, North Carolina……………………………………………………………………..……… 52
31. Graphs showing (A) diurnal temperature fluctuations in upper regolith piezometer P8S,
lower regolith piezometer P8I, Boyd Branch stream water, and streambank sediment pore
water, August 10 to 14, 2008, and (B) and (C) seasonal temperature fluctuations
recorded in P8S, P8I, Boyd Branch stream water, and streambank sediment pore water,
August 2008 to February 2009, in relation to air temperatures, Bent Creek Research
Station, North Carolina…………………………………………………………………………….. 53
32. Graphs showing storm-induced temperature fluctuations in upper and lower regolith wells
BC-1S and -1I and bedrock well BC-1D (discharge area), in response to (A) hourly rainfall
and air temperatures during a summer storm event, August 26, 2008, and (B) hourly
rainfall and air temperatures during an early winter storm event, December 8 to 17, 2008,
Bent Creek Research Station, North Carolina.…………………………………………….. 55
33. Seasonal temperature fluctuations in upper regolith well BC-1S (discharge area), in
relation to air temperatures, July 2008 to January 2009, Bent Creek Research Station,
North Carolina. .………………………………………….……………………………………….. 56
34. Map of monitor wells and groundwater ages at site BC-1 located in low-lying discharge
area, Bent Creek Research Station, North Carolina…………………………………………. 56
35. Graph showing (A) storm-induced temperature fluctuations in upper and lower
regolith wells BC-2S and -2I (discharge area), and (B) storm-induced water level
fluctuations at BC-2S and BC-2I, in response to late summer storm event, September
26 to October 3, 2004, Bent Creek Research Station, North Carolina…………….………. 58
36. Graph showing storm-induced temperature fluctuations in upper and lower regolith
wells BC-2S and -2I (discharge area), and (B) storm-induced water level fluctuations
at BC-2S and BC-2I, in response to winter storm event, January 12 to January 22,
2005, Bent Creek Research Station, North Carolina ……………………….……………… 59
37. Graphs showing seasonal temperature fluctuations in upper and lower regolith wells
BC-2S and -2I and bedrock well BC-2D (discharge area), in relation to air temperatures,
August 2004 to December 2005, Bent Creek Research Station, North Carolina……….. 60
38A Plan view map of selected observation wells measured during 72-hour aquifer
pumping test, Bent Creek Research Station, North Carolina……………………………. ….. 63
38B. Generalized cross section A-A‟ showing selected wells screened in regolith, Bent
vii
Creek Research Station, North Carolina .……………………………………………………… 64
38C. Generalized cross section B‟-B showing selected wells screened in regolith, Bent
Creek Research Station, North Carolina ……………………………………………………… 65
38D. Generalized cross section C-C‟ showing selected wells screened in regolith, Bent
Creek Research Station, North Carolina ……………………………………………………… 66
39. Graph showing time to initial drawdown in observation wells in vicinity of pumping
well BC-4I during aquifer test of October 11 to 14, 2004, Bent Creek Research
Station, North Carolina …………………………………………………………………….. ….. 69
40. Graph showing drawdown and recovery in observation wells during 72-hour
aquifer pumping test conducted at BC-4I, October 11 to 17, 2004, Bent Creek
Research Station, North Carolina.……………………………………………………………… 70
41. Map showing approximate lateral extent of influence due to pumping well BC-4I at
20 gallons per minute for 72 hours, October 11 to 14, 2004, Bent Creek Research
Station, North Carolina…………………………………………………………………………… 71
42. Graph showing drawdown in bedrock pumping well BC-4D during 45-hour aquifer test… 71
43. Graph showing mean age dates (chlorofluorocarbon piston-flow range) for stream
and groundwater at selected wells, Bent Creek Research Station, North Carolina ……… 73
44. Map showing location of ground- and surface-water quality sample locations, Bent
Creek Research Station, North Carolina ………………………….…………………………. 76
45. Stiff diagrams showing major ion milliequivalents in water samples collected from
(A) streams, (B) upper regolith wells, (C) lower regolith wells, and (D) open-hole
bedrock wells, June 2006, Bent Creek Research Station, North Carolina ….…………… 79
46. Trilinear Piper diagrams showing water chemistry data from April 2003, at (A)
regolith wells and stream sites, and (B) open hole bedrock wells, Bent Creek
Research Station, North Carolina, from Huffman and others (2006)…………………….. 80
47. Graph showing water quality results for selected parameters measured in
open-hole bedrock well BC-4D at various sample dates from 2003 to 2007, Bent
Creek Research Station, North Carolina ……………………………………………………. 83
48. Line plots showing range and median values for (A) pH, (B) specific conductance,
and (C) dissolved oxygen in wells and streams during periodic sampling events
between 2003 and 2007, Bent Creek Research Station, North Carolina………………. . 85
49. Line plots showing range and median values for (A) calcium, (B) magnesium, and
(C) sodium in wells and streams during periodic sampling events between 2003
and 2007, Bent Creek Research Station, North Carolina………………………………… . 86
50. Line plots showing range and median values for (A) bicarbonate, (B) chloride, and
(C) sulfate in wells and streams during periodic sampling events between 2003
and 2007, Bent Creek Research Station, North Carolina …..…………………………….. 87
51. Graph showing median silica concentrations observed in stream and well samples
during periodic sampling events between 2003 and 2007, Bent Creek Research
Station, North Carolina ……………………………………………………………………….. 88
52. Graphs showing hourly recordings in lower regolith well BC-2I for (A) specific
conductance, (B) dissolved oxygen and water levels, and (C) ground-water
temperature, in response to storm events from February 15 to March 5, 2005, Bent
Creek Research Station, North Carolina…………………………………………………….. 89
53. Graph of median dissolved radon-222 concentrations observed in well samples
during periodic sampling events between 2003 and 2007, Bent Creek Research
Station, North Carolina ……………………………………………………………………….. 91

Tables

1. Description of rock types common in Bent Creek Research Station, North Carolina ……… 22
2. Dominant foliations within the Bent Creek Research Station, as mapped by
W. Burton (written communication, July 25, 2008) and foliation attributes in
bedrock wells at Bent Creek Research Station, North Carolina ……………………………… 23
viii
3. Summary of relative thicknesses of saprolite and transition zone material that comprise
the regolith at seven core locations, Bent Creek Research Station, North Carolina………… 29
4. Median values for hydrologic budget components in the Blue Ridge and Piedmont
Provinces, in inches per year, as estimated using stream flow recession and baseflow
analysis, from Rutledge and Mesko, 1996………………………………………………………. 32
5. Geologic and hydrogeologic summary information obtained from rock coring and well
testing at Bent Creek Research Station, North Carolina………………………………………. 36
6. Estimated aquifer properties in the regolith flow system, Bent Creek Research Station,
North Carolina .…………………………………………………………………………………….. 39
7. Horizontal and vertical hydraulic gradients at Bent Creek Research Station, North
Carolina …………………………………………………………………………………………….. 40
8. Low and high water levels, average water levels, and total fluctuation during the period
of record, and depth to water on August 14, 2008, in wells at Bent Creek Research
Station, North Carolina …………………………………………………………………………… 45
9. Summarized water level fluctuations during the period of record for regolith wells and
bedrock wells at Bent Creek Research Station, North Carolina …………………………….. 46
10. Temperature fluctuations in different portions of flow system in response to seasonal
climate and short-term rainfall events, Bent Creek Research Station, North Carolina …… 50
11. Position, depth, and screen information for observation wells measured during aquifer
pumping test conducted at lower regolith well BC-4I, Bent Creek Research Station,
North Carolina …………………………………………………………………………………….. 67
12. Drawdown in relation to time in observation wells measured during aquifer pumping test
conducted at lower regolith well BC-4I, Bent Creek Research Station, October 11 to
14, 2004 ……………………………………………………………………………………………. 68
13. Apparent age at time of sampling in 2005 of stream and groundwater at selected
locations, Bent Creek Research Station, North Carolina …………………………………….. 74
14. Statistical summaries of water quality results for regional studies and sample locations
at the Bent Creek Research Station, North Carolina …………………………………………. 81
15. Median water quality results for locations within the Bent Creek Research Station, North
Carolina, 2003 to 2007, and data compiled from regional studies………..…………………. 82
16. Whole-rock chemistry oxide analyses for rock samples obtained from core holes BC-1,
BC-3, and BC-7, Bent Creek Research Station, North Carolina………….………………….. 83

Appendices

1 A. Geology and structural features in the vicinity of Boyd Branch, Bent Creek
Experimental Forest watershed, North Carolina, as modified from 1:12,000
scale map by Merschat and Carter, 2002 ……………………………………………………… 97
1 B. Geology and structural features in the vicinity of Boyd Branch, Bent Creek
Experimental Forest watershed, North Carolina, written communication,
W. Burton, July 25, 2008 …………………………………………………………………………. 98
2. Geophysical and optical televiewer logs for bedrock wells BC-1D, BC-2D, BC-3D,
BC-4D, BC-5D, and BC-7D, Bent Creek Research Station, North Carolina ……………….99-104
3. Well construction details for wells drilled at Bent Creek Research Station, North
Carolina ….………………………………………………………………………………………… 105

Hydrogeology and Water Quality at the Bent Creek Research Station, NC, 2002-2009 1
Hydrogeology and Water Quality at Bent Creek Research Station, Buncombe
County, North Carolina, 2002 to 2009

by Ted R. Campbell

ABSTRACT

A comprehensive multi-year hydrogeologic investigation (September 2002 to August 2009) was
conducted in the Bent Creek Experimental Forest watershed in Buncombe County, North Carolina, as
part of the North Carolina Department of Environment and Natural Resources and U.S. Geological
Survey (USGS) cooperative Piedmont-Mountain Resource Evaluation Program (REP). The Bent Creek
Research Station (BCRS) was installed to evaluate groundwater availability, movement, and quality in a
regolith-fractured rock flow system dominated by mica schist geology. The site lies within the Blue
Ridge physiographic province characterized by complex geology consisting of metamorphosed,
deformed, fractured, and weathered meta-sedimentary and intrusive rocks. The BCRS is one of ten
hydrogeologic research stations installed in the Piedmont and Mountains as part of the REP.
The BCRS is comprised of one stream gage, 7 rock cores, 15 piezometers, and 6 well clusters
along a mile long transect from topographic highs in recharge areas to topographic lows in discharge
areas. Each well cluster consists of three wells open to different zones within the groundwater system:
upper regolith, lower regolith, and fractured bedrock. Two aquifer pumping tests (72 and 45 hour) were
conducted in different portions of the groundwater flow system to evaluate hydraulic interconnectivity.
Groundwater occurred within a three-part system: upper regolith, lower regolith, and fractured
bedrock. The upper regolith system (2 to 38 feet thick, median = 23 feet) has a high porosity and
variable permeability and consisted of sandy, clayey, silty residuum-saprolite. The lower regolith
system (5 to 46 feet, median = 19 feet) has a variable porosity and high permeability and consisted of
partially weathered transition zone material with both primary porosity (pore spaces) and secondary
porosity (rock partings and veins). Fractured bedrock occurred at a depth of 33 to 60 feet below land
surface and generally contained few primary openings. Interlayering of saprolite, transition zone
material, and un-weathered bedrock occurred at 3 of 7 cored locations and was due in large part to
interlayered parent rock.
The lower regolith (median well yield = 5 gallons per minute (gpm) and hydraulic conductivity =
8 feet per day) had a much higher permeability than the upper regolith (median well yield = 1.5 gpm and
hydraulic conductivity = 2 feet per day). Bedrock well yields ranged from less than 0.5 to 40 gpm
(median = 1 gpm and average = 7 gpm). Rock coring, heat pulse flow measurements, and optical
televiewer imaging indicate relatively few primary (open and significant) fractures at 7 well locations.
Results from a 45-hour aquifer pumping test support this assessment and demonstrated a lack of
hydraulic connectivity between the pumped bedrock well and two bedrock wells located within 1500
feet, and minimal, sluggish connectivity between the pumped bedrock well and overlying regolith wells.
The dominant lithology at the BCRS is well-foliated mica schist, interlayered with lesser
amounts of poorly-foliated biotite gneiss/metagraywacke and migmatite. Bedrock foliations strike
predominantly to the NE, dipping to the NW and SE at angles between 25 and 85 degrees. Bedrock
fractures generally cross cut foliation and strike predominantly NW, dipping to the NE or SW at angles
between 50 and 70 degrees. Less often, fractures are parallel to foliation or are of stress relief type.
Groundwater flow at BCRS generally is from topographic highs to lows, consistent with
conceptual models of flow in the Blue Ridge and Piedmont physiographic provinces. Vertical gradients
generally are downward in upslope and midslope recharge areas and upward in lower lying discharge
areas. Rainfall events produced a rapid, temporary increase in potentiometric heads in upper and lower
regolith wells that dissipated within about 7 days. Water temperature and vertical head gradients in
clustered wells were used to demonstrate upwelling from the deeper to the shallower groundwater
system during rain events in a low-lying discharge area. Data indicate that at least some recharge
infiltrates to the water table within 0.5 to 2 days of rainfall events in two low-lying discharge areas. Age
estimates of groundwater in four bedrock wells ranged from about 20 to 50 years, and, within a given
well, a mixture of ages was common (including water less than 2 years old in at least one upslope
bedrock well). Observing a mixture of older and younger waters within bedrock wells is consistent with
Hydrogeology and Water Quality at the Bent Creek Research Station, NC, 2002-2009 2
the regional conceptual model in which multiple potential flow paths (and source waters) contribute to
the total volume of water in a bedrock well.
Bent Creek, a 3rd order stream that drains the larger watershed, is a major control on
groundwater flow direction in the Boyd Branch basin. Locally, groundwater discharges to some reaches
of Boyd Branch, a small, weakly incised 2nd order tributary to Bent Creek, and groundwater under-flow
occurred in one measured reach of the stream. The water table occurred at a depth of about 3 to 32
feet below land surface and varied seasonally and with topographic setting, and vertical head gradients
between the shallow and bedrock flow system typically were 1 to 2 ft except in one discharge location
with a 12 ft difference. Water levels tended to be lowest in late summer when evapotranspiration
effects tend to be greater than recharge, and highest in late winter.
The dominant water type for bedrock wells is calcium bicarbonate, and for streams and regolith
wells is sodium/magnesium-bicarbonate. Ground and surface water are relatively soft, of high quality,
and suitable for most domestic and industrial-use purposes. Values of pH, calcium, magnesium, and
sulfate were higher than typical regional levels in some of the BCRS bedrock wells, due in part to the
rock type and local geochemistry. Deeper groundwater contained higher amounts of dissolved ions
than shallow groundwater due to increased residence time for the dissolution of minerals.

INTRODUCTION

This investigation is part of a multi-year cooperative partnership with the N.C. Department of
Environment and Natural Resources Division of Water Quality (NCDENR DWQ) and the U.S.
Geological Survey (USGS) to characterize ambient quality and movement of groundwater in the
Piedmont and Blue Ridge physiographic provinces of North Carolina (Daniel and Dahlen, 2002). The
study was conducted at the Bent Creek Research Station (BCRS) (fig. 1) located in the Pisgah National
Forest in southwest Buncombe County, a setting characterized by mica schist and biotite gneiss. About
5 percent of the hydrogeologic units within the Piedmont-Mountains are classified as schist, and about
18 percent are classified as mafic (to include biotite) gneiss (Daniel and Dahlen, 2002).
The BCRS site was selected to better understand groundwater quality and movement within the
weathered and unweathered parts of the flow system in a mountain setting underlain by interlayered
mica schist and biotite gneiss. Results of this study will be integrated with findings from nine other sites
(fig. 1) across the Piedmont-Mountains to better understand, map, and protect the State and Nation‟s
water resources and to help advance a number of key objectives (Pippin and others, 2008; Chapman
and others, 2005; Daniel and Dahlen, 2002) of the DWQ-USGS partnership. Specifically, the BCRS
findings will be used to develop a better working understanding of:
the overall hydrogeologic framework of the Piedmont and Blue Ridge physiographic provinces;
the three portions of the fractured rock groundwater flow system and their hydraulic connectivity
in different geologic and geomorphologic settings;
recharge and discharge in fractured rock systems and their effect on ground- and surface-water
quality;
potential groundwater storage in fractured rock settings; and
the interconnection between surface- and groundwater flow systems and their effects on water
quality.
Findings will also be incorporated into a comprehensive groundwater database for the region.

Hydrogeology and Water Quality at the Bent Creek Research Station, NC, 2002-2009 3

Background and Previous Studies

Groundwater in the Piedmont-Mountains of North Carolina occurs in complex hydrogeologic
terranes that consist of metamorphosed igneous, sedimentary, and volcanic rocks. Rocks within these
terranes have undergone multiple deformations and often are interlayered at varying scales (inches to
miles). Local stratigraphy is shaped by the movement (tectonic, erosional, alluvial, and fluvial
processes) and alteration (chemical and mechanical weathering) of these rocks.
Piedmont-Mountains groundwater generally occurs in a two-part flow system - shallow flow
through weathered regolith and deeper flow through fractured bedrock. Some researchers have
referred to a three-part flow system in which the highly permeable transition zone is defined as a
separate unit (Chapman and others, 2005). In their hydrogeologic characterization of the Piedmont,
Harned and Daniel (1992) described the nature of the transition zone between regolith and bedrock and
the relationship between storage in the granular, highly porous regolith and fracture conduits in the
underlying bedrock (fig. 2). In a summary of findings, Daniel and Dahlen (2002) noted regolith
thicknesses in the region from zero (at rock outcroppings) to more than 150 feet (ft), depending on
parent rock type, topography, climate, and groundwater chemistry, and observed that the number of
fractures and size of fracture openings in consolidated bedrock decreases with depth (openings tend to
be very small below about 750 feet due to pressure from the overlying material).

Figure 1. Location of Piedmont-Mountains hydrogeologic research stations,
including Bent Creek Research Station, and North Carolina physiographic
provinces.
. 1 2
3
5
7
6
8
9
10 . 4
Blue Ridge Coastal Plain
Bent Creek
Research
Station
Buncombe
County .
Research Station
7 Coweeta Hydrogeologic Research Station
6 Allison Woods Hydrogeologic Research
8 N.C. Zoological Park Research Station
9 Duke Forest Research Station
Station
10 Western Carolina University Research Station
Map number Research station
1 Langtree Peninsula at Lake Norman
2 N.C. State University Lake Wheeler Road
3 N.C. State University Upper Piedmont
4 Bent Creek Experimental Forest
5 Raleigh Hydrogeologic Research Station
Field Laboratory
Map number Research station
70 miles
N
. . . . . .
Piedmont
.

Coastal Plain
Hydrogeology and Water Quality at the Bent Creek Research Station, NC, 2002-2009 4

Various researchers have described the hydrogeology and conceptual groundwater framework
of the Piedmont-Mountain region of North Carolina, including LeGrand (1967), Heath (1983, 1984),
Swain and others, (1996), and Daniel and Dahlen (2002). Mesko and others (1999) published a
description of Blue Ridge and Piedmont hydrogeology and hydrogeologic terranes, including geologic
and water-bearing properties. Daniel and Payne (1990) compiled a hydrogeologic unit map for the
region. LeGrand (2004) published a conceptual model of the region that summarized a number of
commonly understood hydrogeologic generalizations. His model describes groundwater flow as
occurring within individual slope-aquifer compartments that correspond to areas between topographic
divides and that are open-ended down slope. These slope-aquifer systems can serve to isolate
ambient (un-pumped, gravity-drained) flow in one slope-aquifer compartment from that of a neighboring
compartment. Seaton and Burbey (2005) described the influence of ancient thrust faults on the
hydrogeology of the Blue Ridge Province in southwestern Virginia and documented the occurrence of a
shallow saprolite aquifer separated from a high-permeability, high storage deeper fault-zone aquifer by
a low-fracture permeability bedrock confining unit.
Daniel (1992) and Daniel and others (1997) statistically analyzed fracture permeability with
depth in fractured rock of different topographic settings within the Piedmont of North Carolina. Daniel
(1989) compiled well information from existing data in the Piedmont-Mountains and determined
relationships between well yield and construction, topographic setting, hydrogeologic units, geologic
belts, and other characteristics.
Chapman and others (2005) published findings from a PMREP station representing a well
foliated, steeply dipping, felsic gneiss (Lake Wheeler Road, Wake County). The groundwater flow
system was described, and hydraulic, flow, and water quality data were reported. Results generally
support the conceptual model of groundwater flow in the region, with local variations in vertical
gradients. Local upwelling of groundwater from the transition zone into the shallow regolith was
observed after recharge events and subsequent rise in stage at a nearby tributary. Elevated nitrate
concentrations were observed in groundwater, likely due to past agricultural activities in the study area.

Figure 2. Schematic showing regolith storage and bedrock fracture conduits in the fractured
rock aquifer system of the Piedmont and Mountains of North Carolina (from Harned and Daniel,
1992).

Hydrogeology and Water Quality at the Bent Creek Research Station, NC, 2002-2009 5
Huffman and others (2006) compiled five years of hydrogeologic and water resource data for 4
PMREP research stations in the Piedmont and Blue Ridge physiographic provinces, including the Lake
Wheeler Road station in Wake County, Langtree Peninsula in Iredell County (moderate to steeply
dipping quartz diorite and weakly foliated biotite gneiss), Upper Piedmont in Rockingham County (gently
to moderately dipping amphibolite and interlayered felsic gneiss), and Bent Creek in Buncombe County
(moderate to steeply dipping schist and biotite gneiss). Data obtained from well installation at the 4
stations (110 wells), rock coring, borehole geophysics, periodic groundwater levels, water quality
measurements, stream flow measurements, and slug tests were presented.
Pippin and others (2008) presented hydrogeologic, groundwater flow, and groundwater quality
findings for the Langtree Peninsula PMREP research station in Iredell County. Results from six years
of data support the general conceptual model of groundwater flow in the region - groundwater generally
flowed from high to low topographic settings, groundwater discharged to a surface water boundary, and
vertical hydraulic gradients generally were downward in recharge areas and upward in discharge areas.
The dominant groundwater type - calcium bicarbonate - was similar in all three zones (regolith,
transition zone, and bedrock) of the groundwater system. Recharge may occur seasonally over a
period of several months or, after heavy rainfall, over a shorter period of a few to several weeks.
McSwain and others (2008) presented two years of hydrogeologic and water resource data for
the PMREP Raleigh Hydrogeologic Research Station in Wake County. The site is underlain by the
Rolesville batholith, a massive to weakly foliated granitic intrusion, which is, in turn, intruded by near
vertical tabular basaltic diabase dikes. Data obtained from well installation, borehole geophysics,
periodic groundwater levels, water quality measurements, and slug tests were presented.
Groundwater-surface water relationships were evaluated using temperature and water quality
measurements from drive points along a cross section of the adjacent Neuse River discharge area.
Groundwater recharge rates for the Piedmont and Blue Ridge of North Carolina were estimated
by Heath (1994), Daniel and Sharpless (1983), Harned and Daniel (1987), and Mew and others (2002).
Studies of groundwater recharge in the Central Piedmont have been made by Daniel (1996) and Daniel
and Harned (1998). And Rutledge and Mesko (1996) evaluated hydrologic characteristics of shallow
aquifer systems in three provinces, including the Blue Ridge, based on stream flow recessions and
baseflow analyses.
Briel (1997) published a comprehensive evaluation of regional water quality data obtained from
various sources from the Valley and Ridge (3734 wells), Piedmont (n = 13,498 wells), and Blue Ridge
(n = 776 wells) physiographic provinces of the eastern U.S. Briel described the dominant chemical type
in Blue Ridge groundwater as a mixed water containing calcium magnesium bicarbonate and sodium
chloride. Trapp (1970) published an evaluation of geology and groundwater resources and quality in
Buncombe (18 wells), Henderson (24 wells), Transylvania (14 wells), and Madison (14 wells) Counties,
North Carolina. Trapp described groundwater in Buncombe County as having a dominant bicarbonate
anion signature and varying cation ratios of calcium, magnesium, and sodium. Reid (1993) presented
results of the National Uranium Resource Evaluation Program in which uranium and other inorganic
constituents were analyzed from nearly 6000 wells across North Carolina. Campbell (2006a, 2008b,
2008) conducted regional studies in the Piedmont-Mountains to evaluate naturally occurring
radionuclides in groundwater, including uranium, radium-226, radium-228, and radon-222; several
areas underlain by granitic rocks were associated with elevated dissolved radon and, to a lesser
degree, elevated dissolved uranium. Combined radium-226 + 228 were below regulatory limits in all
sampled wells.
Geologic mapping of the Bent Creek watershed was published by Merschat and Carter (2002).
Local, site-scale geologic mapping was conducted in the Boyd Branch sub-basin within the Bent Creek
Experimental Forest by W. Burton (written communication, July 25, 2008). Unpublished soil maps and
descriptions are available for the Bent Creek watershed (H. McNab, written communication, February 5,
2004).

Purpose and Scope

The purpose of this report is to present findings from a hydrogeologic investigation conducted
from September 2002 to August 2009 at the BCRS, located within the Bent Creek watershed. The
study focuses specifically on one part of the Bent Creek watershed – the Boyd Branch sub-basin (fig.
Hydrogeology and Water Quality at the Bent Creek Research Station, NC, 2002-2009 6
3). The sub-basin, hereinafter referred to as BCRS or study area, represents one geologic type area in
the Piedmont-Mountains region. A type area is defined here as a representative geologic setting
commonly found in other parts of the region and having transferability value. Over time, the DWQ-
USGS partnership will compare groundwater availability, movement, and quality data from one type
area to other type areas across the region.
Data used in this investigation were obtained from existing sources (for example, soil and
geology maps) and from fieldwork conducted by the NCDWQ and USGS scientists between 2002 and
2009. Fieldwork included the following activities:

rock coring was conducted at 7 locations
6 well clusters and an additional bedrock well (18 wells in all) were installed along a mile-long
transect from topographic highs to lows to evaluate groundwater conditions; the well clusters
were designed to measure different depths within the groundwater system and consist of an
upper regolith, lower regolith, and bedrock well.
7 piezometer clusters were installed (15 piezometers in all) to monitor drawdowns during
aquifer pumping tests
borehole geophysical logs and direction adjusted optical televiewer images were collected in 6
bedrock wells
stream and groundwater-quality samples were collected and analyzed periodically at various
locations over 5 years
monthly water levels were measured in study wells over a 6-year period
age dates were analyzed for selected groundwater samples
continuous, hourly groundwater level and groundwater quality monitoring was conducted over
a 1-year period at one well cluster in a discharge area
continuous, hourly temperatures were recorded in 3 well clusters, 1 shallow well, stream water,
and streambank sediment over a 9-month period
aquifer slug and pumping tests were conducted to evaluate the water-bearing properties of the
groundwater system.

Study Area Location and Setting

The BCRS is located within the Boyd Branch sub-basin of the Bent Creek watershed, 11 miles
south of the city of Asheville, in southwest Buncombe County in the Blue Ridge physiographic province
of western North Carolina (fig. 3). The Blue Ridge physiographic province is part of the Appalachian
Highlands, a mountainous range that extends from southern Canada to central Alabama with elevations
ranging from about 6,800 to 2,000 feet above mean sea level (asl). The BCRS watershed is comprised
of rugged, pristine, forested land that is owned and managed by the U.S. Forest Service for research
and recreational uses.
The larger Bent Creek watershed, of which BCRS is part, covers an area of 8.7 mi2 and is NE
trending and trough-shaped, with steep slopes near the ridgeline drainage divide (about 3,600 feet asl)
and milder slopes in the lower elevations along the valley floor (about 2,100 feet asl) (fig. 4). The
natural erosion of weaker rock over very long periods of time has shaped topography in the basin. The
weaker rock tends to occur in areas where faults, joints, and (or) fractures are present or in
the vicinity of contacts between different rock types. These areas tend to erode, form, and shape
natural drainage features within the basin. Streams and tributaries often are visible expressions of
these underlying structures and the erosion process. Slope instability is not uncommon in steep
terrains like Bent Creek, and Pomeroy (1991), Witt (2005), and Wooten and others (2009) have
documented historic landslides within the Bent Creek watershed.
The BCRS watershed drains to Bent Creek, a tributary of the French Broad River, just south of
Asheville (fig. 3). The French Broad is located on the western side of the Eastern Continental Divide
and is thus part of the large network of rivers that form the Tennessee, Ohio, and Mississippi Rivers,
which ultimately empty into the Gulf of Mexico.

Hydrogeology and Water Quality at the Bent Creek Research Station, NC, 2002-2009 7

BCRS
study area,
Boyd
Branch
sub-basin
2000 ft N
N
.
1 - USGS stream gage 03447894
2 - USGS stream gage 0344789265
3 - NOAA weather station 0246CA
4 - NOAA weather station 310724
.
Bent Creek
watershed
. 1/4
2
3
2 mi
Figure. 3. Location of the Bent Creek Research Station in the Boyd Branch sub-basin, Bent Creek
Experimental Forest watershed, North Carolina, and location of National Oceanic and Atmospheric
Administration rain gages 0246C and 310724 and U.S. Geological Survey stream gages 03447894
and 0344789265.

contour interval = 40 ft
Schematic, not to scale
N
. .
2
Hydrogeology and Water Quality at the Bent Creek Research Station, NC, 2002-2009 8

.
.
.
.
.
.
. .
.
7
6
5 P7
2
3
P4
P3 . P2
4
. P8
1 .
1000 ft
N
.
.
.
.
.
.
. .
.

P7

P4
P3 . P2
. P8
.
1000 ft
N Monitor wells and
core locations

4
USGS
stream gage
0344789265
Recharge area, vertical hydraulic gradient
from regolith wells to bedrock well
Vertical contour
interval = 40 ft

Discharge area, vertical hydraulic gradient
from bedrock well to shallow regolith wells 2
BC-4S
BC-4I
BC-4D P5I
P5S
P6S
P6IP1SP1I
18 ftN
B
o
y
d

B
r
a
n
c
h

R
o
a
d
BC-4S
BC-4I
BC-4D P5I
P5S
P6S
P6IP1SP1I
18 ftN
B
o
y
d

B
r
a
n
c
h

R
o
a
d
4
Figure 4. Topographic setting, well locations, and areas of recharge and discharge, Bent Creek
Research Station, North Carolina.
BC-1
BC-2
BC-3
BC-5
BC-6
BC-7
Piezometer cluster
consisting of upper
regolith piezometer (ex.
P7S) and lower regolith
piezometer (ex. P7I)
. BC-7 Well cluster & core
P4 Piezometer .
Recharge area, downward vertical
hydraulic gradient
Discharge area, upward vertical
hydraulic gradient
Hydrogeology and Water Quality at the Bent Creek Research Station, NC, 2002-2009 9
The Boyd Branch sub-basin is situated in the north-central part of the Bent Creek watershed
(fig. 3). It is a 1.1 mi2 oblong-shaped basin, trending northwest, approximately perpendicular to Bent
Creek (fig. 4). Elevations range from about 3,200 ft asl at the ridgeline to about 2,180 ft asl at the valley
floor where Boyd Branch discharges into Bent Creek.
The Bent Creek watershed is underlain predominately by mica schist and muscovite-biotite
gneiss (referred to as metagraywacke) of the Ashe Metamorphic Suite and Tallulah Falls Formation
(Merschat and Carter, 2002). Within the BCRS area of investigation, mica schist is the predominate
rock type, with slightly lesser amounts of interlayered migmatitic biotite gneiss. The site therefore
represents a schist-gneiss “type area”, one of several geologic type areas being compared as part of a
larger systematic evaluation of hydrogeologic settings across the region.

Methods of Investigation

The methods used for well installation, rock coring, borehole geophysical testing, water level
measurements, water quality sampling, and slug and aquifer testing conducted during this investigation
are described in the report of standard operating procedures for conducting hydrogeologic field
investigations (NCDENR, DWQ, 2008). Additional site specific methods are presented here.
The BCRS consists of 6 well clusters and an additional bedrock well (18 wells in all) positioned
along a high to low topographic transect within a “slope-aquifer” system similar to that described by
LeGrand (2004). The well clusters were designed to evaluate groundwater conditions from recharge to
discharge areas within a slope-aquifer sub-basin (Boyd Branch). The clusters consist of an upper
regolith (saprolite) well, a lower regolith (transition zone) well, and an open hole bedrock well.
The well naming convention used in this study is as follows: all wells designated with “S” (for example,
“BC-3S”) are upper regolith wells screened across saprolite material (with the exception of BC-1S,
which is screened across alluvium/colluvium/transition zone material). All wells designated with “I” (for
example, “BC-3I”) are lower regolith wells screened across transition zone material. And all wells
designated with “D” (for example, “BC-3D”) are open borehole bedrock wells with surface casing seated
about 3 to 10 feet into competent bedrock. Lower regolith well BC-6I was vandalized shortly after
installation and could not be used, and upper regolith well BC-2S was damaged during construction and
could be used only for measurement of water levels and selected field parameters. Seven piezometer
clusters (consisting of an upper regolith piezometer and a lower regolith piezometer) (15 piezometers in
all) were installed along two approximately perpendicular transects in the vicinity of an aquifer test site
to measure detailed drawdown information and to map lateral groundwater movement in one
representative recharge area of the BCRS.
To aid in well construction design and to better understand site geology, continuous soil and
bedrock cores were collected at each well cluster location. Cores were collected using a wire-line
coring rig and a 5-ft long, 2.5 inch inside diameter core barrel. Core descriptions were conducted on
and off site, and included soil characteristics, rock type, fracture and foliation type and orientation,
weathering, and other geologic features.
Monitoring wells were constructed using mud-rotary, air-rotary, and hollow-stem auger methods.
Coarse sand was used as a filter pack around, and about 2 feet above, the screened interval to improve
the hydraulic connection between the well and the aquifer. At least 2 feet of bentonite was placed as a
seal on top of the sand pack, and the remaining annular space was filled with Portland cement grout to
land surface. Steel protective casings with locking covers were installed on all wells.
Borehole geophysical logging was conducted in 6 bedrock wells and included caliper, natural
gamma, temperature, fluid resistivity, electrical resistivity, and heat pulse flow meter logs. Direction-
adjusted optical televiewer images were obtained to determine foliation and fracture orientations.
Water levels were measured to an accuracy of 0.01 ft approximately monthly using electronic
water level meters referenced to a measuring point at the top of casing. Measuring points were
surveyed and referenced to North American Vertical Datum of 1988. Hourly water levels (and
temperature) were recorded using a float tape or pressure transducer at wells BC-2S, -2I, and -2D from
August 2004 to December 2005 and at BC-7S from May 2008 to August 2009. In addition, hourly water
quality parameters (pH, temperature, specific conductance (SC), and dissolved oxygen (DO)) were
measured with a downhole multimeter during this period to determine seasonal fluctuations and to
Hydrogeology and Water Quality at the Bent Creek Research Station, NC, 2002-2009 10
evaluate short-term fluctuations and response to recharge. The multimeter probe was cleaned and
calibrated at least once per month following USGS guidelines (Wagner and others, 2006).
Submersible temperature data loggers were placed in selected cluster wells, stream, and
streambank locations to evaluate groundwater-surface water relationships in recharge and discharge
areas and to evaluate the relationship between the shallow regolith and deeper bedrock flow system.
The loggers were placed on the stream bottom, 2 to 5 feet into the stream bank sediment, or were
suspended downhole within the screened interval, about 2 feet from the bottom of the well, or near a
primary fracture in bedrock wells. Hourly temperature readings were recorded during selected periods
between July 2008 and February 2009. Hourly air temperatures and hourly rainfall data were obtained
from the National Oceanic and Atmospheric Administration weather stations 0426CA (about 1.2 miles
N-NE of BCRS) and KAVL (about 5.9 miles SE of BCRS).
Slug tests were conducted at selected wells to evaluate horizontal hydraulic conductivity in the
vicinity of the well. The slug consisted of a sealed 5-ft long, 3-inch diameter PVC pipe that was quickly
submerged (falling head test) or removed from the well water (rising head test). A pressure transducer
and data logger recorded the change in water level over time in response to the displaced water until
levels returned to within 95 percent of the pre-test level. The time-water level data were analyzed using
the Bouwer and Rice (1976) method which accounts for partial penetration effects and changing aquifer
thickness. Assumptions inherent in the method are that the aquifer is porous, isotropic (no directional
variation in properties), effects of elastic storage are negligible, and the position of the water table does
not change during the test. The aquifer thickness was approximated by the total regolith thickness for
tests on upper and lower regolith wells or the length of the open borehole for tests on bedrock wells.
A 72-hour constant-rate aquifer pumping test was conducted at BC-4I (fig. 4) to evaluate aquifer
characteristics in the regolith and to determine the degree of connectivity between upper regolith, lower
regolith, and fractured bedrock in this area. A 4-inch diameter, variable speed submersible pump was
placed in BC-4I to a depth of 41 ft, and a 1.5 inch diameter polyethylene hose was used as a discharge
line. A digital, in-line flow meter was used to estimate pump discharge rates, and the rates were
manually verified using a graduated 5-gallon bucket and stopwatch. Care was taken to ensure a
steady, continuous, uniform flow rate throughout the test, and discharge water was routed away from
areas of local groundwater recharge and into Boyd Branch approximately 300 feet downgradient to the
east. Water levels were measured by an In-Situ® Hermit data logger attached to In-Situ® pressure
transducers, In-Situ® miniTROLLs, or, for outlying wells, by handheld electric tapes. Conditions were
dry throughout the test and a minor seasonal drift in water levels was noted in distant background wells
BC-3 and BC-7 (fig. 4) for comparison to test wells in proximity to pumping. Staff gage measurements
made during the test did not reveal observable drawdown in Boyd Branch. No attempt was made to
measure potential changes in flow rates in Boyd Branch during the aquifer test.
Regolith wells installed within about 50 feet of the pumping well were used to observe the delay
in the time-drawdown curve at individual wells and to analyze aquifer characteristics in this area. These
wells were positioned in a radial pattern and along perpendicular transects to observe potential
anisotropy and the shape and extent of the cone of pumping influence. If weathered-in-place parent
rock retains relict foliations and these foliations are preferred pathways, then both the saprolite and
transition zone likely will be anisotropic, both horizontally and vertically. More distant wells were used
to observe the distance that the cone of depression expanded during the test.
A 45-hour constant-rate aquifer pumping test was also conducted at BC-4D, an open-hole
bedrock well, from June 20 to 22, 2006 to evaluate the relative hydraulic connectivity between the
deeper fractured rock and the shallow regolith flow system. The pump and discharge line configuration
were the same as that used in the aquifer test at BC-4I, and the well was pumped at a constant flow
rate of about 6 gpm. Well BC-4D is cased to 61 feet bls and is 501 feet deep. Water levels were
measured throughout the test in the same 19 wells and piezometers used during the aquifer test at BC-
4I. Water levels were measured by an In-Situ® Hermit data logger attached to In-Situ® pressure
transducers, In-Situ® miniTROLLs, or, for outlying wells, by handheld electric tapes. Flow rates were
measured at regular intervals throughout the test using manual and instrumented totalizer methods.
Conditions were mostly dry throughout the test and minor seasonal drift in water levels was adjusted
during data analysis using water level trends in distant background wells for comparison.
Groundwater quality samples were collected using variable-speed, submersible 2-inch diameter
pumps and smaller, single-speed plastic submersible pumps. At least 3 well volumes of water were
purged and field parameters were stable prior to sample collection at regolith wells. Removing 3 well
Hydrogeology and Water Quality at the Bent Creek Research Station, NC, 2002-2009 11
volumes from the deep, 6-inch diameter bedrock wells was impractical using the low-rate sample
pumps, so at least one well volume of water was removed and field parameters were stable prior to
sampling. To obtain a representative sample, the sample pump was lowered to a mid-screen depth in
regolith wells. Through strategic pump placement, water quality samples collected from bedrock wells
were taken at a depth corresponding to the most predominant fracture in the borehole (Chapman and
others, 2005). Even still, sample water collected from a bedrock well is a composite of water from
multiple fractures, with each fracture representing a flow path of varying length originating in one or
more recharge areas.
For age dating purposes selected groundwater samples were analyzed for chlorofluorocarbons
(CFCs) and tritium-helium. CFCs are stable anthropogenic compounds used as refrigerants since the
1930s and are commonly used to date the age of relatively young (less than about 50 years)
groundwater. Three CFCs (CFC-11, -12, and -113) were measured and used in age dating models to
determine an apparent mean-estimated age. The CFC-based age estimate represented the mean
residence time composite for the sample and typically was presented as a range. Unless noted, the
estimated ages do not account for mixing scenarios that can occur in wells with large open intervals or
multiple producing fractures.
Tritium and helium were sampled and used to estimate the youngest component of groundwater
ages. Tritium is a radioactive element that entered groundwater systems after atmospheric testing of
nuclear devices began in 1952. Tritiogenic helium-3 is derived from the radioactive decay of tritium,
and the ratio of tritium-helium in the water sample may be used to estimate the age of younger
groundwater to within one year.
As part of the age dating methodology, dominant fracture sets were isolated and sampled using
an inflatable packer system at two bedrock wells. For quality control, a halon tracer had been mixed
with compressed air during the original air-rotary well drilling activities to indicate whether or not modern
air containing CFC‟s was introduced to the groundwater during well installation. If the halon tracer was
not present in the groundwater sample, then the sample generally is considered representative of the
natural flow system and an age was estimated. Because of the complex three-dimensional nature of
fractured groundwater flow, bedrock wells inherently draw water that is a composite of various source
waters, and a range of ages may be expected. Sampling and analysis methods are provided in the
USGS Reston Chlorofluorocarbon Laboratory website (http://water.usgs.gov/lab/cfc/).

Acknowledgments

The author wishes to thank Melinda Chapman and Brad Huffman of the USGS for their active
participation throughout the project and manuscript preparation (including the generation of selected
figures) and review. Their expertise and input have been critical to the success of the project and
report.
The geologic mapping and personal communications of Carl Merschat (North Carolina
Geological Survey (NCGS), Mark Carter (formerly with NCGS), and Bill Burton (USGS, Reston) were
instrumental in developing an understanding of the geologic framework of the study area. The detailed
core descriptions made by Arthur Merschat were important in the lithologic and structural analysis of the
site. Special thanks also to Ken Gillon and Nick Bozdog (NCGS) for their assistance in the
mineralogical characterization of selected core samples.
The author wishes to acknowledge the U.S. Forest Service for their cooperation in this project
and for providing access to suitable drill sites. Henry McNab (U.S. Forest Service) provided expertise
and logistical support during the project, and his assistance is greatly appreciated.
The field work and drilling activities associated with this investigation were substantial and
required the effort of many individuals, including Ray Milosh, Billy Casper, Jesse Martin, Chuck Pippin,
Rick Bolich, Don Geddes, Shuying Wang, Andrew Pitner, Joju Abraham, and Landon Davidson of the
NCDENR DWQ and Brad Huffman of the USGS. Finally, special thanks are extended to Melinda
Chapman, Landon Davidson, Rick Bolich, Brett Laverty, Shuying Wang, and Ken Gillon for their review
of the data interpretations and (or) manuscript. Their comments and suggestions greatly improved the
report.

Hydrogeology and Water Quality at the Bent Creek Research Station, NC, 2002-2009 12

HYDROGEOLOGIC SETTING

The BCRS is located within the Blue Ridge physiographic province, a dynamic, rugged,
mountainous landscape that has undergone continual evolution during various episodes of
metamorphism, igneous intrusion, folding, faulting, and erosion throughout its billion-plus year history.
Groundwater in this region occurs in complex hydrogeologic terranes that vary locally with local
geology. The geology through which groundwater flows in these regions consists of (1) an unsaturated
zone of soil (organic rich), alluvium, colluviums, and weathered bedrock; (2) a saturated zone of highly
weathered (residuum and saprolite) bedrock; (3) a saturated zone of partially weathered bedrock that is
a transition between weathered and unweathered rock; and (4) fractured crystalline bedrock. Relatively
shallow alluvium is common in areas proximate to active or buried stream channels, and colluvium
occurs in areas associated with debris fans and flows. Saprolite is typically clay-rich, soft, friable, and
porous, and is derived from in-place weathering of parent rock; as such, it retains the fabric and
fractures of the parent rock but with a much lower bulk density. Transition zone material consists of
partially weathered, highly permeable bedrock. Interlayering of weathered and unweathered rock within
the regolith is common depending on the composition and texture of the parent rocks from which they
are derived.
Taken together, the regolith is comprised of layers of soil, alluvium, colluvium, saprolite, and
transition zone material, and grain sizes range from clay to boulders, with porosities ranging from 35 to
55 percent (Heath, 1980). Because of its high porosity, the regolith is the storage reservoir for water to
the underlying fractured bedrock flow system. Porosity decreases with depth in the regolith as
weathering decreases (Stewart, 1962). Porosity of bedrock in the region ranges from about 1 to 3
percent. Most water bearing openings in bedrock are associated with joints, faults, and stress relief
fractures rather than pores.
Chemical and mechanical weathering produces a regolith layer that varies in thickness from 0 to
150 feet in the North Carolina Piedmont and Mountains (Daniel and Dahlen, 2002). Many factors affect
the weathering of parent rock, including climate, topographic setting, and rock composition, texture,
structure, and degree of foliation. For example, feldspars and micas weather to produce clay-rich
residuum and saprolite with high porosities and low hydraulic conductivities. The type and quality of
weathering can affect the properties of groundwater flow and quality in both the shallow regolith system
and the deeper fractured-flow system. As such, an understanding of the rock type, texture, and
structure is important in evaluating groundwater occurrence, flow, and quality at a site.

Regional Geology

The state is divided into geologic belts with similar rock types and geologic history, but
significant variability occurs at the local scale. The BCRS is located within the Blue Ridge Belt (NCGS,
1985), a terrain that consists of a complex mixture of metamorphosed sedimentary and igneous rocks
that have been folded, fractured, faulted, and altered during multiple high temperature and pressure
metamorphic events. Its major rock units occur as northeast trending belts, and its tectonically derived
fractures tend to be oriented to the northeast and northwest. Hydrogeologic units in the Blue Ridge Belt
include felsic gneiss, mafic gneiss, schist, quartzite, and phyllite (Daniel, 1989).
Bent Creek is underlain by complex, highly metamorphosed sedimentary and, to a lesser
extent, mafic volcanic rocks of the Ashe Metamorphic Suite and Tallulah Falls Formation, and include
mica schist, metagraywacke, paragneiss, and amphibolite (Hatcher, 1971; NCGS, 1985). These rocks
are of late Proterozoic age and have been metamorphosed to sillimanite grade. The mica schists are
derived from metamorphosed siltstones and shales and are medium- to coarse-grained, lepidoblastic,
and composed mostly of quartz, feldspar, biotite, muscovite, and garnet. The metagraywackes are
derived from metamorphosed muddy sandstones and are fine- to medium-grained, granoblastic, and
composed mostly of quartz, feldspar, biotite, muscovite, garnet, epidote, and apatite. The paragneisses
include biotite-muscovite gneiss and biotite gneiss that extend across a large portion of central
Buncombe and northern Transylvania Counties (Hatcher, 1971). Other rocks include compositional and
textural variations of these (for example, metasiltstone and metaconglomerate) and other minor,
discontinuous bodies such as migmatite, pegmatite, and granofels.
Hydrogeology and Water Quality at the Bent Creek Research Station, NC, 2002-2009 13
The NCGS state geologic map (1:500,000 scale, 1985) places Bent Creek within a unit of
muscovite-biotite gneiss (Zam) (fig. 5), and regional mapping by Daniel and Payne (1:500,000 scale,
1990) places Bent Creek within the quartzite (QTZ) hydrogeologic unit (fig. 6). However, site-scale
mapping places BCRS in a strongly foliated mica schist setting with interlayered (scales from inches to
feet), weakly foliated metagraywacke/gneiss and migmatite (1:12,000 scale, Merschat and Carter, 2002
(map in appendix 1a); W. Burton, written communication, July 25, 2008 (map in appendix 1b). A small
number of amphibolite bodies also occur locally. Foliations mapped by Merschat and Carter (2002)
tend to strike to the NE and are moderate to steeply NW or SE dipping (fig 7). Two generations of
schistocity are present: a steeply-dipping S1 that is parallel to bedding and primary compositional
layering (photos in figs. 8a and 8b), and a gently to moderately-dipping S2 (photos in figs. 9a and 9b)
that is axial planar to locally recumbent map-scale folds that dominate the structural framework
(Merschat and Carter, 2002). Joints measured by Merschat and Carter (2002) generally are steeply
dipping with a NW strike (fig. 10), roughly perpendicular to the axes of local-scale folds. The nearest
regional fault is the NE trending Brevard Fault about 8 miles to the southeast (fig. 5).

-
,
,
Figure. 5. Geologic formations in the Ashe Metamorphic Suite and Tallulah Falls
Formation (modified from North Carolina Geological Survey, 1985) and physiographic
provinces of North Carolina.
Muscovite biotite
gneiss, Zatm
Metagraywacke,
Zatw
Biotite gneiss,
Zatb
Amphibolite, Zata
Bent Creek
Research
Station
Modified from NCGS (1:500,000 scale, 1985)
Brevard Fault
20 miles N
Blue Ridge

Piedmont

aFigure 7.
Stereonet
projection
of poles to
foliations,
fold axes,
crenulation
axes, and
mineral
lineations
within the
Bent Creek
watershed,
Asheville,
North
Carolina,
modified
from
Merschat
and Carter,
2002.
sFigure 8a.
Example of
well
developed,
steeply
NW-dipping
S1
compositio
nal layering
in biotite
gneiss/met
agraywack
e on Blue
Ridge
Parkway
about two
miles west
of Bent
Creek
Research
Station,
North
Carolina.
.
Piedmont Coastal Plain
Hydrogeology and Water Quality at the Bent Creek Research Station, NC, 2002-2009 14

Figure. 6. Geologic belts in North Carolina (North Carolina Geological Survey, State map,
1:500,000 scale, 1985) and hydrogeologic units mapped by Daniel and Payne in Buncombe
County (1990) (upper inset), as modified from Huffman and others, 2006.

Hydrogeology and Water Quality at the Bent Creek Research Station, NC, 2002-2009 15

Figure 7. Stereonet projection of
poles to foliations, fold axes,
crenulation axes and mineral
lineations within the Bent Creek
watershed, Asheville, North
Carolina, modified from Merschat
and Carter, 2002.
Figure 8A. Example of well
developed, steeply NW-dipping
S1 compositional layering in
biotite gneiss/metagraywacke on
Blue Ridge Parkway about two
miles west of Bent Creek
Research Station, North Carolina.
W. Burton, written
communication, July 25, 2008.
Photo by B. Burton, July 2008
Figure 8B. Example of S1
schistocity (compositional layering)
in mica schist (ms), in northwest
part of the Bent Creek Research
Station, North Carolina (just north
of outcrop observation station BC-
11 shown in appendix 1b), facing
west. Well-defined layers reflect
varying quartz content. W. Burton,
written communication, July 25,
2008.

Photo by B. Burton, July 2008
Hydrogeology and Water Quality at the Bent Creek Research Station, NC, 2002-2009 16

Figure 9A. Example of locally
dominant, SE-dipping S2
schistocity showing iron-sulfide
precipitation (above hammer), on
Blue Ridge Parkway about two
miles west of Bent Creek
Research Station, North Carolina.
Remnants of steeply-NW-dipping
S1 compositional layering visible
at lower left. W. Burton, written
communication, July 25, 2008.
Figure 9B. Example of medium-
to fine-grained mica schist, S
dipping S2 schistocity dominant
foliation (strike at N83E, dipping
43 degrees), transposing N-
dipping S1 compositional layering
(strike at N87W, dipping at 50
degrees) to produce intersection
lineation (at pen) on S2 surface,
in northwest part of Bent Creek
Research Station, North Carolina
(outcrop observation station BC-4
shown in appendix 1b), facing
west. W. Burton, written
communication, July 25, 2008. Photo by B. Burton, July 2008
Photo by B. Burton, July 2008
Hydrogeology and Water Quality at the Bent Creek Research Station, NC, 2002-2009 17

Local Geology

The BCRS contains two dominant lithologies (minerals listed in order of increasing abundance):
biotite (minor)-muscovite-garnet-quartz-plagioclase-chlorite schist (photos in figs. 8b, 9b, 11, 12, and
13a; table 1) and muscovite-biotite-quartz-plagioclase gneiss/metagraywacke (photos in figs. 8a, 11,
12, 13a, and 14; table 1). Interlayering (at scales ranging from inches to feet), compositional variation,
and migmatization are common (photos in figs. 8b, 11, and 13a). Accessory minerals include garnet,
sillimanite, and pyrrhotite. Both lithologies contain interspersed layers and veins of migmatite and (or)
leucosome. Minor bodies of amphibolite occur locally. This description is consistent with mapping by
Merschat and Carter (2002) and Burton (W. Burton, written communication, July 25, 2008) and with
observations of seven rock cores.
Merschat and Carter mapped two dominant lithologies within the Boyd Branch sub-basin as part
of their work in Bent Creek (1:12,000 scale): 1) a very light gray to greenish gray to medium gray,
strongly-foliated, fine-grained, lepidoblastic to porphyroblastic quartz-chlorite-garnet-muscovite schist
(Zas), with thicknesses ranging from less than an inch to several inches and commonly interlayered
with sillimanite-garnet-chlorite-mica metasiltstone; and 2) a light gray to medium dark gray, non- to
weakly-foliated, medium- to coarse-grained, granoblastic to lepidoblastic, locally migmatitic muscovite-
biotite-potassium-feldspar-quartz-plagioclase metagraywacke (Zag), with thicknesses ranging from
several inches to several feet. Their interpretation was based on observations of outcrops and two thin
sections of rocks collected in the study area. Burton described the lithologic setting similarly (W.
Burton, written communication, July 25, 2008) based on his field reconnaissance of the Boyd Branch
study area, but referred to the metagraywacke as a gray biotite gneiss (bgn) and listed slightly different
relative percentages of mineral compositions (appendix 1b). Burton also described a minor occurrence
of black and white layered migmatite gneiss (photo in fig. 13b) that corresponds to Merschat and
Carter‟s migmatitic metagraywacke. Minor bodies of amphibolite were noted in both surveys.
Eighty eight to 192 feet of rock core was recovered from each of 7 locations in BCRS (fig. 4) to
augment information obtained from outcrop observations and the geologic map prepared by Merschat
and Carter (2002). In describing these cores, three dominant lithologies were noted (A. Merschat,
written communication, August 25, 2003): 1) a light gray to greenish gray to dark gray, migmatitic, fine-
to coarse-grained, porphyroblastic to lepidoblastic chlorite-garnet-muscovite-biotite schist with variable
Lower hemisphere equal area (Schmidt)
stereonet projection of poles to joints
Figure 10. Stereonet projection of poles to joints within the Bent Creek
basin, Asheville, North Carolina, Merschat and Carter, 2002.
Hydrogeology and Water Quality at the Bent Creek Research Station, NC, 2002-2009 18
composition, texture, and migmatite content; 2) a very light gray to medium gray, fine- to medium-
grained, muscovite-biotite-feldspar-quartz gneiss, referred to as metagraywacke, with variable
migmatite content; and 3) a very light gray to medium gray, medium- to coarse-grained, foliated- to
weakly-foliated, granoblastic muscovite-biotite-quartz-plagioclase and potassium feldspar gneiss,
referred to as migmatite. A hybrid of types 1 and 2 also was noted, as well as white to very light gray,
coarse-grained leucocratic pegmatite or leucosome, dominated by quartz and potassium feldspar with
biotite, muscovite, and garnet, referred to as leucosome/pegmatite. Other very minor lithologies
observed in the cores included amphibolite, quartzite, calc-silicate, and trondhjemite.

Figure. 11. Example of interlayered gneiss/metagraywacke, schist, and migmatite from rock sample
at core site BC-1, Bent Creek Research Station, North Carolina.
Core BC-1, 176 to
185 ft bls
Medium gray, medium-grained, weakly
foliated muscovite-biotite-quartz-plagioclase
gneiss (referred to as metagraywacke by
Merschat and Carter, 2002), 184 ft bls
Migmatite layer, quartzofeldspathic,
177.5 ft bls
Dark gray, migmatitic, well-foliated garnet-
biotite-muscovite-chlorite-plagioclase schist,
178.5 ft bls and 177.5 ft bls
Hydrogeology and Water Quality at the Bent Creek Research Station, NC, 2002-2009 19

Core 6
Core 4
Core 7
Core 1
Core 3
Depths in feet below land surface: Core 1: 94; Core
3: 83.5; Core 4: 57.5; Core 6: 114.5; Core 7: 103.
Core site 1: very light to medium gray, medium- to coarse-grained, moderately to strongly foliated garnet-
biotite-muscovite-quartzofeldspathic-chlorite SCHIST, interlayered with foliated leucosome
Core site 3: dark gray, fine- to medium-grained, weakly to moderately foliated garnet-chlorite-
quartzofeldspathic GNEISS/METAGRAYWACKE, with very thin layers of leucosome
Core site 4: light to medium gray, medium to coarse grained, weakly foliated biotite-muscovite-garnet-
quartzofeldspathic GNEISS/METAGRAYWACKE, with very thin layers of garnet-sillimanite-biotite-
chlorite SCHIST
Core site 7: medium to dark greenish gray, fine to medium grained, moderately foliated biotite, muscovite,
biotite, quartzofeldspathic- SCHIST
Core site 6: medium to dark gray, fine- to medium-grained, strongly foliated sillimanite, muscovite,
biotite, quartzofeldspathic GNEISS/METAGRAYWACKE, with leucosome
Figure. 12. Samples from rock core sites BC-1, -3, -4, -6, and -7, with descriptions, Bent Creek
Research Station, North Carolina.
Hydrogeology and Water Quality at the Bent Creek Research Station, NC, 2002-2009 20

biotite gneiss/
schist
metagraywacke
Photo by B. Burton, July 2008
Figure. 13A. Example of
interlayered
gneiss/metagraywacke,
schist, and migmatite in
outcrop located on Blue
Ridge Parkway about two
miles west of Bent Creek
Research Station, North
Carolina. W. Burton,
written communication,
July 25, 2008.

Figure. 13B. Isoclinally
folded migmatitic gneiss,
S-dipping (axial plane)
folds (likely F2), looking
west, at outcrop
observation station BC-8
(shown in appendix 1b),
Bent Creek Research
Station, North Carolina.
W. Burton, written
communication, July 25,
2008.

.
Photo by B. Burton, July 2008
Hydrogeology and Water Quality at the Bent Creek Research Station, NC, 2002-2009 21

Dominant foliations within the study area are an S1, sub-inch to inches scale, compositional
layering, generally parallel to foliation (photos in figs 8a, 8b, and 13a; table 2), and an S2, a later
schistocity, most commonly seen developing within the mica schist (photos in figs. 9a and 9b; table 2)
(W. Burton, written communication, July 25, 2008). The S2 foliation overprints S1 and is axial planar to
F2 folds that fold S1 compositional layering. The S1 layering is typically NE-striking and steeply NW
dipping (typically 50 to 80 degrees), but can be gently NW-dipping or overturned to steeply SE-dipping
(about 70 to 90 degrees) by F2 folds. S2 schistocity is also NE-striking and typically moderately SE-
dipping (typically 30 to 80 degrees). These observations are in agreement with those mapped by
Merschat and Carter (2002), with rock cores, and with data obtained using an oriented optical
televiewer in bedrock wells (appendix 2).

Figure 14. Sample from rock core BC-1 showing biotite gneiss/metagraywacke and
quartzofeldspathic leucosome, Bent Creek Research Station, North Carolina.
Light gray to white, medium- to
coarse-grained, muscovite-
quartz-feldspar leucosome layer,
187.5 and 189 ft bls
Medium gray, medium-
grained, migmatite,
quartzofeldspathic,
186.5 ft bls
Medium gray, medium-
grained, muscovite-biotite-
quartz-plagioclase
gneiss/metagraywacke,
185.5 ft bls
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
22

Dominant rock
types Descriptiona Occurrence
Accessory
minerals
Schist Light gray to greenish gray to dark gray; strongly foliated;
migmatitic; fine- to coarse-grained; quartz-biotite-sericite-
chlorite-muscovite; variable composition, texture, and
migmatization; thickness of layering ranges from millimeters to
centimeters; migmatite may be interlayered or occur as lenses
parallel to foliation
Interlayered with
gneiss/metagraywacke rock,
often at scales of a centimeter
to meters
Sillimanite,
garnet, pyrrhotite
Gneissb /
metagraywackec
Light to medium gray; variably migmatitic; non to weakly
foliated; fine- to medium-grained, muscovite-biotite-feldspar-
plagioclase-quartz; thickness of layering ranges from
millimeters to decimeters; occasional millimeter to centimeter
thick micaceous layers create stripped pattern; migmatite may
be interlayered or occur as lenses and knots parallel to foliation
Interlayered with schist rock,
often at scales of about a
centimeter to meters
Garnet,
pyrrhotite
Minor rock types Descriptiona Occurrence
Accessory
minerals
Migmatite White to very light to medium gray; non to weakly foliated;
medium- to coarse-grained; muscovite-biotite-quartz-k feldspar-
plagioclase feldspar; occurs in thin layers and lenses within
other rock types (most commonly metagraywacke) at
thicknesses ranging from millimeters to decimeters
Interlayered and (or) vein-like,
derived from local melting
during regional high-grade
metamorphism
Pyrrhotite,
chlorite,
chalcopyrite
Leucosome White to very light gray; coarse grained; muscovite-biotite-
quartz-k feldspar; occurs in thin layers and lenses parallel to
foliation within other rock types at thicknesses ranging from
millimeters to decimeters; occasionally cross-cuts foliation; less
biotite and muscovite than the migmatite listed above
Interlayered and (or) vein-like,
derived from local melting
during regional high-grade
metamorphism
Pyrrhotite,
chlorite,
chalcopyrite
Amphibolite Black to dark gray to greenish gray; non to strongly foliated;
variably migmatitic; medium- to coarse-grained; varying mineral
composition
Sparse, discontinuous bodies,
possible interlayering
Pyrrhotite,
chalcopyrite
a minerals listed in order of increasing abundance
b referred to as gneiss by B. Burton, written communication, July 25, 2008
c referred to as metagraywacke by Merschat and Carter, 2002
Table 1. Description of rock types common at Bent Creek Research Station, North Carolina, modified from Merschat and Carter, 2002
and A. Merschat, written communication, July 1, 2003.
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
23

Table 2. Dominant foliations within the Bent Creek Research Station, as mapped by W. Burton (written
communication, July 25, 2008) and foliation attributes in bedrock wells at Bent Creek Research Station,
North Carolina.

Rock type Strike Dip angle and dir Location of outcrop
ms, mica schist NE 32o NW BC-14, west of Boyd Branch
ms, mica schist NE 73o NW BC-12, west of Boyd Branch
ms, mica schist NE 57o NW BC-11, west of Boyd Branch
ms, mica schist NE 53o NW BC-9, west of Boyd Branch
ms, mica schist NE 78o SE BC-23, west of Boyd Branch
ms, mica schist NE 73o SE BC-5, west of Boyd Branch
ms, mica schist NE 83o NW east of Boyd Branch
ms, mica schist NE 57o NW BC-1, east of Boyd Branch
ms, mica schist NE 30o NW BC-18, east of Boyd Branch
ms, mica schist NE 80o NW BC-16, east of Boyd Branch
ms, mica schist NE 13o NW BC-15, east of Boyd Branch
bgn, biotite gneissc NE 88o SE BC-25, west of Boyd Branch
bgn, biotite gneissc NE 84o SE BC-24, west of Boyd Branch
bgn, biotite gneissc NE 42o NW BC-10, west of Boyd Branch
bgn, biotite gneissc NE 56o NW BC-13, west of Boyd Branch
bgn, biotite gneissc NE 78o NW BC-2, east of Boyd Branch
bgn, biotite gneissc NE 76o NW BC-20, east of Boyd Branch
bgn, biotite gneissc NE 42o NW BC-19, east of Boyd Branch
bgn, biotite gneissc E 38o N BC-38, east of Boyd Branch
Rock type Strike Dip angle and dir Location of outcrop
ms, mica schist SE 28o SW BC-21, west of Boyd Branch
ms, mica schist NE 78o SE BC-23, west of Boyd Branch
ms, mica schist NE 60o SE BC-6, west of Boyd Branch
ms, mica schist E 43o S BC-4, west of Boyd Branch
ms, mica schist NE 12o SE BC-37, east of Boyd Branch
Monitor wells Rock type
BC-1D interlayered, migmatitic ms-bgn
BC-2D interlayered, migmatitic ms-bgn
BC-3D interlayered, migmatitic bgn-ms
BC-4D migmatitic ms
BC-5D interlayered, migmatitic bgn-ms
BC-7D interlayered, migmatitic ms-bgn
a associated with compositional layering; b overprints S1 & is axial planar to F2 folds that fold S1compositional layering;
c bgn also mapped as metagraywacke by Merschat and Carter, 2002; d determined from oriented OTV logs
NE strike, shallow-steep dip to NW & SE
NE strike, shallow-steep dip to NW & SE
NE & N strike, moderate-steep dip to NE & N
NE strike, moderate dip to NW & SE
NE strike, mod-steep dip to SE
NE strike, mod-steep dip to SE
Foliation attributesd
S1 schistocitya
S2 schistocityb
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
24
Moderate to well-developed foliation was noted in the recovered rock from all seven core
locations. Foliation strike orientations measured using optical televiewer data were to the NE in all
bedrock wells and are moderate to steeply dipping (table 2; maps in appendix 1a and 1b). Grain sizes
vary from fine to medium with lesser amounts of coarse material.
Mild, regional post-F2 warping may deform both S1 and S2, but no map-scale folds (F2 or later)
could be identified. Map-scale lithologic contacts should follow the NE strikes of S1 foliation, with minor
inflections caused by topography and F2 folding. Due to the complexity of lithologic interlayering, no
map-scale contacts could be mapped with confidence in the study area (W. Burton, written
communication, July 25, 2008). Faults or fracture zones were not observed in outcrops.
Jointing is not common in the study area but would be expected to increase in abundance as
the predominant rock type increases from mica schist to biotite gneiss to migmatite gneiss to
amphibolite. Twelve joints measured in the Boyd Branch sub-basin by Merschat and Carter (2002) are
steeply SW or NE dipping (typically 50 to 80 degrees) and strike NW roughly perpendicular to the axes
of their local-scale folds (appendix 1a). Based on observations of rock core, rock outcrops, and
geophysical logs and optical televiewer data (appendix 2) rock partings in the study area generally are
foliation parallel and moderately to steeply dipping.
Depth to bedrock ranged from 33 to 60 feet thick (median = 48 ft) in the seven cored locations,
although outcrops occur sparsely along Boyd Branch, Bent Creek, and in other local areas of the basin
(appendix 1a and 1b). The regolith is variably thick and consists of a thin mantle of organic-rich silty,
sandy, clayey loam, highly weathered residuum and weathered saprolite, and a transition zone of
partially weathered rock. The residuum generally is more weathered than saprolite and tends to lack
textural fabric. The saprolite retains relict textures, structures, and fractures from its parent bedrock.
Because of its granular and often clayey, silty character, the residuum-saprolite tends to be very porous
and have a lower hydraulic conductivity than the underlying transition zone or bedrock fractures.
The transition zone is partially weathered material of varying grain sizes from silt to cobbles to
large rock fragments, is very permeable, and retains the structure of its parent bedrock. It tends to
weather unevenly, particularly in settings characterized by interlayered parent rocks, and often consists
of inter-layered sequences of saprolite and (or) partially weathered and un-weathered bedrock.
Alluvium and colluvium may occur locally, depending on the geomorphic setting.
The thickness of the soil, alluvium, colluvium, residuum, saprolite, and transition zone layers
varied with location. Because of uneven weathering, a clear delineation between layers was not always
apparent, and a subjective interpretation was made as necessary. Thicknesses of soil, saprolite, and
transition zone material observed in the 7 BCRS cores are shown in the schematic in fig. 15.
Three types of weathering were noted in the study area: (1) an even grading from soil to
residuum to saprolite to transition zone was noted at sites 2, 3, 4, and 6 (see example in photo of core
BC-3, fig. 16); (2) soil/alluvium/colluvium grading directly to transition zone material (saprolite is nearly
absent) at site 1 located in a low lying setting adjacent to the basin-draining third order Bent Creek (see
example in photo of core BC-1, fig. 17); and (3) a repeating sequence of unweathered rock interlayered
within intervals of saprolite and (or) transition zone material occurs at sites 5 and 7 (see example in
photo of core BC-5, fig. 18)1.

1 The occurrence of interlayered sequences of unweathered and weathered bedrock has important implications for
well design, particularly those associated with contaminant transport investigations, due to the potential for
insufficient seating of surface casing material and (or) contaminant underflow through unanticipated, partially
weathered, highly transmissive channels that can occur several feet beneath the surface of otherwise competent
bedrock
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
25

10
20
30
40
50
60
70
80
90
100
110
189’
1 2 3 4 5 6 7
Core Holes
Soil, organic matter
Transition zone of
partially weathered
bedrock
Saprolite
Crystalline bedrock
Depth
Below
Land
Surface
(feet)
168’192’186’
This interval
includes alluvium /
colluvium
* Depth to
bedrock in well
BC-4D, located
75 ft distant, is
about 45 ft
*
10
20
30
40
50
60
70
80
90
100
110
189’
1 2 3 4 5 6 7
Core Holes
Soil, organic matter
Transition zone of
partially weathered
bedrock
Saprolite
Crystalline bedrock
Soil, organic matter
Transition zone of
partially weathered
bedrock
Saprolite
Crystalline bedrock
Depth
Below
Land
Surface
(feet)
168’192’186’
This interval
includes alluvium /
colluvium
* Depth to
bedrock in well
BC-4D, located
75 ft distant, is
about 45 ft
*
Figure 15. Near surface geology in seven cored locations at Bent Creek
Research Station, North Carolina.
Bedrock
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
26

Figure. 16. Core recovery of regolith illustrating even, gradational weathering at BC-3, Bent
Creek Research Station, North Carolina.
Similar rock
type to depth
of core
Saprolite zone
begins here
Transition zone
begins here
Bedrock
begins here
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
27

Figure. 17. Core recovery of regolith illustrating near absence of saprolite at low-lying site BC-1
adjacent to Bent Creek, Bent Creek Research Station, North Carolina.
Similar rock
type to depth
of core
Alluvium/transition
zone and interlayered
bedrock begins here
Bedrock
begins
here
Saprolite/alluvium
zone begins here

Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
28

Saprolite ranged from about 10 to 38 feet thick (median = 23 ft), except at the low-lying site BC-
1 adjacent to Bent Creek, where long term scouring has resulted in a 2 feet thick layer (fig.17; table 3).
Scouring also may have thinned the saprolite layer (12 feet thick) at site BC-2, another low-lying site
adjacent to Boyd Branch near the base of the watershed. The thickness of the saprolite layer relative to
the total regolith thickness was smallest at BC-1 (fig. 15; table 3). As observed in the cores, the
uppermost saprolite consists primarily of reddish to golden brown, well-drained, fine- to coarse-loamy
residuum, with varying clay, silt, and fine sand content. Deeper saprolite consists of orange-brown to
medium brown to dark brown, clayey to sandy silt material that is weathered and retains the relict
structure of its parent rock. Saprolite soils in BCRS are usually moderately to strongly acidic (4.5 to 6)
Figure. 18. Core recovery of regolith illustrating interlayered transition zone at BC-5, Bent Creek
Research Station, North Carolina.
Similar rock
type to depth
of core
Saprolite zone
begins here
Transition zone
begins here
Bedrock
begins
here
Transition zone
intervals
interlayered
within saprolite
and bedrock
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
29
and often are coated with iron oxides that produce a reddish complexion (soil maps, H. McNab, written
communication, February 5, 2004). Alluvium/colluvium is present at sites BC-1 and -2. Five monitoring
wells and eight piezometers were screened across saprolite to measure water level fluctuations and
water quality in the upper regolith flow system.

A transition zone of partially weathered rock occurred at all sites and ranged in thickness from 5
to 46 feet (median of 19 ft) (table 3). The thickness of the transition zone relative to the total regolith
thickness was greatest at sites BC-1 and -2 (figs. 15 and 17; table 3). The transition zone was defined
as having both primary porosity (pores) and secondary porosity (rock partings). The transition zone
material typically was overlain by an interval of dark brown sandy-clayey-silty saprolite.
Three of the 7 cores (BC-1, BC-5, and BC-7) were characterized by repeating sequences of
transition zone material interlayered with saprolite and (or) unweathered rock (figs. 15 and 18). Uneven
and interlayered weathering appeared to be associated with interlayering of parent rocks. Iron and
manganese leachate stains were common on the fragments and fractures of most transition zone
material, indicating oxidation-reduction reactions in the weathered, aerobic portion of the aquifer.
Seven monitoring wells and 8 piezometers were screened across the transition zone to measure water
level fluctuations and water quality at the base of the regolith flow system.
Depth to competent bedrock ranged from 33 to 60 feet (median of 48 ft) (fig. 15; table 3).
Groundwater flow in bedrock occurs primarily within fractures and other rock openings. Based on core
and OTV logs, fractures are mainly parallel to foliation and occur at moderate to steep angles. Low
angle stress relief fractures that cross cut foliation occur less frequently. Where open, fractures often
contained dissolution stains with either orangish brown to yellowish orange iron oxide (from the
oxidation of iron sulfides, pyrrhotite, pyrite, or chalcopyrite) or black manganese oxide (from
manganese-bearing minerals garnet, biotite, chlorite, or amphibole, for example). Staining occurred
less frequently in the bedrock fractures than in the transition zone, particularly in anaerobic settings
such as discharge areas. Seven open-hole monitoring wells were installed in bedrock to measure
water level fluctuations and water quality in the deeper fractured rock groundwater flow system.

Surface Water

The Bent Creek watershed receives an average annual precipitation of 48 inches, distributed
fairly evenly throughout the year (National Oceanic and Atmospheric Administration (NOAA) weather
station 310724, located 2.5 miles from BCRS, fig. 3, 1949 to 2008, accessed from www.acis.sercc.com
on July 2, 2009). Average monthly rainfall is highest in March (4.9 inches) and lowest in October (3.4
inches). Monthly rainfall amounts for the study period (fig. 19) were obtained from NOAA weather
station 0246CA, about 1.2 miles N-NE of BCRS.

Table 3. Summary of relative thicknesses of saprolite and transition zone material that comprise the
regolith at seven core locations, Bent Creek Research Station, North Carolina.
BC-1 BC-2 BC-3 BC-4 BC-5 BC-6 BC-7
thickness of saprolite, in feet 2 12 38 26 33 10 23
thickness of regolitha, in feet 48 43 60 33 53 38 50
% of saprolite thickness to regolith thickness 4 28 63 79 62 26 46
thickness of transition zone, in feet 46 25 17 5 17 19 20
thickness of regolitha, in feet 48 43 60 33 53 38 50
% of transition zone thickness to regolith thickness 96 58 28 15 32 50 40
depth to competent bedrock 48 43 60 33 53 38 50
a As defined here, regolith includes soil, saprolite, and transition zone material.
Core locations
a As defined here, regolith includes soil, saprolite, alluvium/colluvium, and transition zone material
45
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
30

A portion of this precipitation runs off the land surface into Bent Creek and its tributaries. The
remainder evaporates, is transpired by plants and trees, or percolates downward through the
unsaturated zone and recharges the groundwater system, eventually discharging to tributaries to Bent
Creek or to Bent Creek itself. The surrounding ridgeline acts as the ground- and surface-water divide
and therefore represents the basin boundary that separates flow in the Bent Creek basin from flow in
other watersheds. A possible rare exception would be groundwater flowing through very deep fractures
that are connected to an adjacent basin.
The 8.74 mi2 Bent Creek watershed is drained by thirteen perennial 2nd order streams that flow
into Bent Creek (fig. 4). Bent Creek, a 3rd order stream, flows into Lake Powhatan and 2.8 miles
downstream where it empties into the French Broad River. Waters of the French Broad flow northwest,
joining the Tennessee River, the Ohio River, and the Mississippi River, eventually emptying into the
Gulf of Mexico.
From 2002 to 2008, annual mean stream flow in Bent Creek ranged from 6.7 cfs (2008) to 19.4
cfs (2005), based on discharge measurements at USGS gage 03447894 in Bent Creek just upstream of
the French Broad River (figs. 3 and 20) (USGS Surface-Water Annual Statistics for the Nation, web
accessed at http://waterdata.usgs.gov/nwis/sw on August 14, 2009). A peak flow of 3,000 cfs occurred
in September 2004, corresponding to a period of intense rainfall associated with hurricanes Ivan and
Frances. Highest flows tended to occur in April, and lowest flows tended to occur in August.
Mean streamflow for the period of record was estimated to be 15 cfs based on hydrograph
separation computations using stream flow records and the USGS PART program (Rutledge, 1998).
The program calculated baseflow in the pristine, largely undisturbed forest to be about 80 percent of
this amount, or 12 cfs. When applied to the Bent Creek watershed drainage area of 8.7 mi2, this
corresponds to about 19 in/yr. Subtracting this amount from an estimate of annual mean precipitation
of 48 inches gives an approximated value for direct runoff and evapotranspiration of 29 inches. These
values are comparable to regional hydrologic budget estimates by Rutledge and Mesko (1996) (table
4). Wide variations are possible within a given region, area, or watershed since recharge is controlled
by local precipitation, basin topography, land use (percentage of disturbed area), land cover, and soil
type and thickness.

Figure 19. Monthly rainfall amounts at National Oceanic and Atmospheric Administration weather
station 0246CA, 1.2 miles N-NE of the Bent Creek Research Station, North Carolina, January 2003
to June 2009.
Monthly rainfall at NOAA Rain Station 0246CA, 1.2 miles N-NE of BCRS (inches)
0
2
4
6
8
10
12
14
16
18
20
Ja
n
-
0
3
Ap
r
-
0
3
Ju
l
-
0
3
Oc
t
-
0
3
Ja
n
-
0
4
Ap
r
-
0
4
Ju
l
-
0
4
Oc
t
-
0
4
Ja
n
-
0
5
Ap
r
-
0
5
Ju
l
-
0
5
Oc
t
-
0
5
Ja
n
-
0
6
Ap
r
-
0
6
Ju
l
-
0
6
Oc
t
-
0
6
Ja
n
-
0
7
Ap
r
-
0
7
Ju
l
-
0
7
Oc
t
-
0
7
Ja
n
-
0
8
Ap
r
-
0
8
Ju
l
-
0
8
Oc
t
-
0
8
Ja
n
-
0
9
Ap
r
-
0
9
Mo
n
t
h
l
y

r
a
i
n
f
a
l
l
,

i
n
c
h
e
s
35.8
inches
Water
Year
2008
33.7
inches
Water
Year
2007
48.5
inches
Water
Year
2006
52.0
inches
Water
Year
2005
59.7
inches
Water
Year
2004
> 61
inches
Partial
Water
Year
2003
Monthly rainfall at NOAA weather station 0246CA, 1.2 miles N-NE of the BCRS, NC
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
31

Mean hourly stream flow in Boyd Branch ranged from 0.5 cfs to 56 cfs, with a median of 1.5 cfs,
based on discharge measurements from March 1, 2004 to November 30, 2005 (fig. 21) at USGS gage
0344789265 in Boyd Branch at Bent Creek Road near Lake Powhatan (fig. 3). The daily mean peak
flow of 56 cfs and the mean hourly instantaneous peak flow of 88 cfs occurred during September 2004,
a period associated with heavy rainfall (17.8 inches in September) caused by hurricanes Ivan and
Southerna Northernb Southerna Northernb
Precipitation 58 43 48 44
Direct runoff 8 4 6 5
Evapotranspiration 25 27 32 29
Stream flow 38 16 18 15
Ground water recharge 33 12 12 10
Ground water discharge 29 11 11 8
no. of basins measured 17 10 15 13
a South of latitude 37
b North of latitude 37
Blue Ridge Piedmont
Table 4. Median values for hydrologic budget components in the Blue Ridge and Piedmont
Provinces, in inches per year, as estimated using stream flow recession and baseflow
analysis, from Rutledge and Mesko, 1996.
0
10
20
30
40
50
60
70
80
90
100
No
v
-
0
1
Ma
r
-
0
2
Ju
l
-
0
2
No
v
-
0
2
Ma
r
-
0
3
Ju
l
-
0
3
No
v
-
0
3
Ma
r
-
0
4
Ju
l
-
0
4
No
v
-
0
4
Ma
r
-
0
5
Ju
l
-
0
5
No
v
-
0
5
Ma
r
-
0
6
Ju
l
-
0
6
No
v
-
0
6
Ma
r
-
0
7
Ju
l
-
0
7
No
v
-
0
7
Ma
r
-
0
8
Ju
l
-
0
8
No
v
-
0
8
Ma
r
-
0
9
Ju
l
-
0
9
Me
a
n

d
a
i
l
y

d
i
s
c
h
a
r
g
e
,

cf
s
0
100
200
300
400
500
600
700
Ou
t
l
i
e
r
s
,

Me
a
n

d
a
i
l
y

d
i
s
c
h
a
r
g
e
,
cf
s
Mean Daily
Q, cfs
Outliers,
cfs
Q data from
USGS Gage
03447894 at Bent
Creek Gap Road near Glen Bald
(shown in fig. 3)

Figure. 20. Mean daily discharge (Q) in Bent Creek at U.S. Geological Survey gage 03447894,
November 2001 to July 2009, modified from USGS stream records, accessed from webpage
http://waterdata.usgs.gov/nc/nwis/ on August 14, 2009.

Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
32
Frances. When applied to the Boyd Branch sub-basin watershed drainage area of 1.1 mi2, the daily
mean streamflow corresponds to 18.5 in/yr. The percentage of this amount represented by baseflow
was not estimated because only 21 months of flow record were available.

Boyd Branch is moderately incised, ranging in channel depth from about 2 to 6 feet depending
on location. Surface water depths range from about 0.5 to 2 ft, depending on recent rainfall events,
location, and time of year. In a portion of its mid to lower reaches the channel is characterized by a
relatively flat, wet, bog-like area of stream bifurcation prior to merging back into a single channel before
its discharge to Bent Creek. Because Boyd Branch is only minimally incised in certain areas, a portion
of groundwater underflows the stream and discharges directly to Bent Creek. As a result, this portion is
not captured by discharge measurements in Boyd Branch, and the measured stream flow at the basin
outlet very likely under-estimates the total amount of recharge and groundwater flow within the sub-
basin.

Groundwater

Groundwater in the Piedmont-Mountains occurs within two geologic flow regimes: the regolith
system and the underlying fractured crystalline bedrock. The regolith system is unconfined and made
up of porous, weathered or partially weathered rock that acts as a storage reservoir for water to the
underlying fractured bedrock system. The regolith system may be further subdivided into an upper
layer (referred to as the upper regolith) of highly porous, low to moderately permeable, weathered
residuum and saprolite, and a lower layer (referred to as the transition zone or lower regolith) of highly
permeable, partially weathered rock that grades into unconsolidated bedrock. The regolith system also
contains alluvium and colluvium, usually near surface in lower lying settings. The materials comprising
the upper and lower regolith may be interlayered or even absent in some settings, and the degree of
weathering may vary considerably across short distances. Because of its high porosity the regolith is
the primary reservoir for groundwater storage and provides slow recharge to bedrock fractures (Heath,
1983, 1994). The fractured bedrock system underlies the regolith system, may be partially confined or
unconfined, and is made up of very low porosity rock with partings that transmit water. Because the
three flow zones usually have different matrix and hydraulic properties, groundwater is said to occur in a
three-part system: upper regolith, lower regolith, and fractured bedrock.
Precipitation recharges the groundwater system at BCRS. A portion of the recharge moves
laterally along shorter flow paths within the regolith and discharges to Boyd Branch or Bent Creek.
Another portion moves laterally along somewhat longer flow paths within the regolith, underflows Boyd
Median value of all mean hourly discharges = 1.5 cfs
-20
0
20
40
60
80
Ju
l
-
0
4
Au
g
-
0
4
Se
p
-
0
4
Oc
t
-
0
4
No
v
-
0
4
De
c
-
0
4
Ja
n
-
0
5
Fe
b
-
0
5
Ma
r
-
0
5
Ap
r
-
0
5
Ma
y
-
0
5
Ju
n
-
0
5
Ju
l
-
0
5
Au
g
-
0
5
Se
p
-
0
5
Oc
t
-
0
5
No
v
-
0
5
De
c
-
0
5
Ho
u
r
l
y

D
i
s
c
h
a
r
g
e
,

c
f
s
0
1
2
3
4
Ho
u
r
l
y

R
a
i
n
f
a
l
l
,

i
n
c
h
e
s
Discharge, cfs
Hourly Rain,
inches, NOAA
raingage 0246C
Figure. 21. Mean hourly discharges in Boyd Branch at U.S. Geological Survey gage 0344789265,
Bent Creek Research Station, for period of record between July 2004 and December 2005, in
relation to hourly rainfall.

Median value of all mean hourly discharges = 1.5 cfs
Maximum value of all mean hourly discharges = 88 cfs, 09/2004
Maximum 24-hour composite of all mean hourly discharges = 56 cfs, 09/2004
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
33
Branch, and discharges to lower reaches of Boyd Branch or to Bent Creek. And a remaining portion
moves downward into underlying bedrock fractures along longer flow paths and eventually discharges
in lower reaches of Boyd Branch (as evidenced, for example, by artesian conditions at BC-2) or Bent
Creek. Flow direction and travel time will depend in part on topography, thickness, continuity, and
hydraulic properties of the regolith, and the number, size, orientation, and connectedness of bedrock
fractures.
The three part groundwater system at BCRS (upper regolith, lower regolith, and fractured
bedrock) was evaluated using data collected from 7 rock cores, 18 monitoring wells (grouped in 6 well
clusters), and 15 regolith piezometers (grouped in 7 clusters). The well clusters were installed along a
mile long NW-SE transect, from topographic highs to lows (figs. 4 and 22), and each cluster consists of
an upper regolith (shallow, S), a lower regolith transition zone (intermediate, I), and a bedrock (deep, D)
well. Upper regolith S wells have 15-foot screens across the water table within saprolite material (with
the exception of BC-1S that is screened across alluvium/colluviums/transition zone material due to the
absence of saprolite in this area). Lower regolith I wells have 15-foot screens across partially or highly
weathered transition zone material just above competent bedrock. Bedrock D wells are open holes
(with casing from land surface to the top of bedrock) in fractured bedrock to depths of 190 to 501 feet
bls. Piezometers have 5-foot screens in the upper and lower regolith. Wire line coring, borehole
geophysics, and downhole optical imaging were conducted at each well cluster location to understand
and map lithology, structure, and fracture density. Borehole geophysical logs and direction-adjusted
optical televiewer images are shown in appendix 2. Well construction details are shown in appendix 3.
Groundwater levels were measured approximately monthly over a 5-year period between February
2003 and December 2008.

BC-7
BC-6
BC-5
BC-4
BC-3 BC-2
Figure 22. Generalized hydrogeologic cross-section along well transect at the Bent Creek
Research Station, modified from Huffman and others, 2006.
Foliation depictions are approximate
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
34

Regolith Flow System

Two distinct flow zones – an upper and lower - occur within the regolith. The upper regolith flow
system has a high porosity and variable permeability and consists of sandy, clayey, silty residuum-
saprolite. The lower regolith flow system has variable porosity and high permeability2 and consists of
partially weathered transition zone material that varies in size from clay to boulder with both primary
porosity (pore spaces) and secondary porosity (rock partings and veins). Directional, anisotropic
hydraulic properties are common particularly in the lower regolith and are associated with relict
foliations, fractures, contacts, and veins.
The thickness of the two-part regolith system measured in the study area ranged from 33 to 60
feet (median = 48 ft) (table 5; fig. 15). The upper system ranged from 2 to 38 feet thick (median = 23 ft),
and the lower system ranged from 5 to 46 feet thick (median = 19 ft). Because of differential
weathering, the thickness, composition, and permeability of the upper and lower flow systems vary with
location, as does the depth to bedrock.
Groundwater generally follows topographic contours from higher to lower elevations. However,
horizontal and vertical head gradients suggest that Bent Creek is a major influence on groundwater
movement and discharge in the Boyd Branch sub-basin while Boyd Branch is, depending on location, a
relatively minor influence. Bent Creek is a 3rd order stream and is the major drain for the larger
watershed, while Boyd Branch is a 2nd order stream that drains the local sub-basin (fig. 4). Head
gradients obtained from a May 24, 2005 synoptic round of water level measurements in 12 upper
regolith wells/piezometers near P8 (figs. 4 and 23), a period of increasing evapotranspiration, suggest
that most groundwater at this location flowed beneath Boyd Branch and south toward Bent Creek. The
gradient from P8 to BC-4 is to the southwest, away from Boyd Branch. Boyd Branch does not incise
deeply into the aquifer here (the elevation of the channel bottom is higher than the average water table
elevation in this portion of the sub-basin). This observation suggests that only a small amount of
groundwater near the top of the water table may be discharging into Boyd Branch in this location, with
the remainder underflowing the stream3. Groundwater also underflowed a small intermittent tributary
between P2 and P3 (figs. 4 and 23) during this measurement period; this is an area of recharge so
unless the water table rises significantly, this is a losing reach of stream.
Areas of groundwater recharge and discharge were determined based on vertical head
gradients and water temperature differences between clustered wells open to the upper regolith, lower
regolith, and bedrock systems. As discussed in sections titled „Seasonal and Storm-Induced
Groundwater Level Fluctuations‟ and „Vertical Head Gradients‟, recharge occurs in the areas of BC-7, -
5, -4, and -3, and discharge occurs in the areas of BC-1, BC-2, and P8 (fig. 4). In recharge areas the
vertical head gradients were downward from the upper regolith to the lower regolith to the bedrock
system, resulting in a net downward flux of groundwater. In discharge areas the vertical head gradients
were upward from lower portions of the flow system to upper portions of the flow system.

Properties

Lower regolith wells were associated with higher yields and permeability than upper regolith
wells. This is due in large part to the more highly transmissive transition-zone material across which the
lower regolith well is screened. At a given well cluster, the yield of the lower regolith well generally was
higher by a factor of 3 to 10 than the yield of its clustered upper regolith well (table 5). In all, lower
regolith wells yielded from 0.5 to 20 gpm (median = 5 gpm) and upper regolith wells yielded from 0.5 to
3 gpm (median = 1.5 gpm).

2 Well-developed transition zones tend to be highly transmissive and may significantly affect contaminant transport
rates (Harned and Daniel, 1992).
3 This is significant when considering issues of contaminant transport and groundwater flow direction in mountain
watersheds, and demonstrates the importance of understanding several factors that may affect stream underflow,
including: depth to water, depth of incised stream, underlying geology and thickness of regolith, size and shape of
recharge boundaries, and relative topographic differences in both the sub-basin and larger watershed.
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
35
Displacement (slug) tests were conducted on regolith wells at sites 1, 2, 3, 4, 5, and 7 to
evaluate hydraulic conductivity (K) in the immediate vicinity of the tested wells. The K values were
about a half an order of magnitude higher in the lower regolith, transition zone wells than in the upper
regolith, saprolite wells, consistent with the fact that the transition zone material is more permeable and
transmissive. As shown in table 6, K values in shallow saprolite wells at BC-2S, -3S, -4S, -5S, and -7S
ranged from 0.2 to 3 ft/day (median = 2 ft/day), and K values in transition zone wells BC-1S, -3I, -4I, -5I,
and -7I ranged from 5 to 30 ft/day (median = 8 ft/day). A representative average K value for the upper
regolith was about 1 ft/day and for the lower regolith was about 10 ft/day. Slug tests also were
conducted in the bedrock well at three of these sites, however because the slug displacement volume
was relatively minor in comparison to the total volume of water in the well bore, bedrock test results
were considered to be accurate only to within an order of magnitude. In general, the K values in
bedrock wells were over an order of magnitude lower than the K values in regolith wells.
The approximate velocity of groundwater was estimated using Darcy‟s Law, Q=KA dh/dl, where
K = hydraulic conductivity and dh/dl = horizontal hydraulic gradient. Written another way and
accounting for actual flow paths (porosity), velocity v = K dh/dl/porosity. The actual velocity will vary in
different areas of the watershed and in different horizons of the groundwater flow system, and would be
expected to be higher in steeper, upslope portions of the watershed, due to higher horizontal hydraulic
gradients, than in flatter, lower lying areas, and higher in the lower regolith (transition zone) system, due
to higher permeability and lower porosities, than in the upper regolith (saprolite) portion.
Porosity in the upper regolith system (mostly saprolite) was estimated to range from 0.25 to 0.7,
with a representative value of about 0.5, based on published values for similar sediment types (Driscoll,
1986; Freeze and Cherry, 1979; and Roscoe Moss, 1990) (table 6). Porosity in the coarser grained
lower regolith was estimated to be somewhat lower, with a range of 0.15 to 0.4, and a representative
value of about 0.3. Hydraulic conductivity was estimated using slug test results and published values
(Fetter, 2001) for similar sediment types (silt, clay, fine and coarse sand, and partially-weathered rock
fragments of varying sizes, with the understanding that the percentage of coarser, less weathered rock
fragments at BCRS generally increases with regolith depth).
Measured horizontal hydraulic gradients ranged from 0.01 to 0.05, but somewhat higher values
are expected in upper, steeper portions of the watershed where water level data were not available
(table 7). Applying these ranges of representative values to the equation v = K dh/dl/porosity,
groundwater velocity v in the loamy sand, clay, and silt of the upper regolith is expected to be between
about 0.01 and 1 ft/day, with a median of about 0.1 ft/day. Groundwater velocity v in the partially
weathered, fragmented rock, cobbles, and silty sands of the lower regolith is expected to be between
about 0.2 and 20 ft/day, with a median of about 2 ft/day. Actual velocities at any given location will
depend on local geology, topographic setting, and rainfall activity (fluxes in the shape of the
potentiometric surface). Vertical hydraulic gradients measured in regolith wells were comparable to the
horizontal hydraulic gradients and ranged from 0.02 to 0.05 (table 7).

Storage

The volume of groundwater storage fluctuates with changes in rainfall, evapotranspiration, and
groundwater withdrawals. Changes in land use/cover (impervious areas, for example) can increase
rainfall runoff and thereby decrease recharge and storage, but these effects are minimal in the BCRS
watershed. Storage increases during periods of rising water tables when precipitation is above normal
(for example, during tropical storms in summer and autumn) and during periods of decreased
evapotranspiration (during cooler periods in early spring and winter). As defined here, storage is the
amount of water that would drain from the aquifer under the force of gravity, and is estimated by
multiplying the volume of saturated regolith by the regolith porosity4.
A very approximate estimate of total groundwater storage in the 1.1 mi2 Boyd Branch basin was
determined using observations of regolith thickness from seven rock cores and water levels in 27
regolith wells/piezometers measured from 2003 to 2008. For purposes of this estimation, it is assumed

4 This measure is also referred to as specific yield for unconfined aquifers, which is the volume of water released
from storage per unit surface area of the aquifer per unit decline in the water table (Freeze and Cherry, 1979).
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
36

BC-1 BC-2 BC-3 BC-4 BC-5 BC-7
Rock Type ms-bgn; interlayered and
migmatitic
ms-bgn; interlayered
and migmatitic
bgn-ms; interlayered
and migmatitic
ms-migmatite;
interlayered
bgn-ms; interlayered and
migmatitic
ms-bgn; interlayered and
migmatitic
Regolith
Thickness of saprolite, ft 2 12 38 26 33 23
Thickness of TZ, ft 46e 25e 17 5 17 20
Thickness of regolith, ft 48 43 60 50d 53 50
Type of weathering profile in regolith very thin saprolite even gradation even gradation even gradation interlayered saprolite-TZ interlayered saprolite-TZ
Bedrock
Depth to bedrock, ft bls 48 43 60 50 53 50 Foliated or non-foliated foliated foliated foliated foliated foliated foliated
Degree of foliation moderate moderate moderate moderate moderate moderate
Grainsize fine to medium medium fine to medium medium medium medium
Foliation strike direction NE NE NE NE NE and N NE
Foliation dip azimuth and angle SE / moderate-steep SE / moderate-steep NW and SE / shallow-
moderate-steep
NW and SE / shallow-
moderate-steep
W and NW / moderate-
steep NW and SE / moderate
Well characteristics
primarya / secondaryb primary / secondary primary / secondary primary / secondary primary / secondary primary / secondary
Number of fractures: 50 to 100 ft bls 0 / 3 0 / 7 1 / 2 0 / 7 0 / 6 0 / 3
Number of fractures: 100 to 150 ft bls 0 / 13 0 / 2 0 / 3 2 / 0 0 / 2 0 / 7
Number of fractures: 150 to 200 ft bls 1 / 7 0 / 3 0 / 7 0 / 3 0 / 3 0 / 8
Number of fractures: 200 to 250 ft bls 0 / 2 0 / 2 0 / 3 0 / 2 0 / 5 0 / 9
Number of fractures: 250 to 300 ft bls 0 / 0 0 / 9 0 / 8
Number of fractures: 300 to 350 ft bls 0 / 2
Number of fractures per foot 0.006 / 0.15 0 / 0.06 0.004 / 0.06 0.004 / 0.03 0 / 0.10 0 / 0.15
Predominant fracture strike direction N (primary)NE (secondary)NE (primary)NE (primary)N-NE (secondary)NE (secondary)
Predominant fracture dip azimuth & angle W / low SE / moderate-steep SE / moderate NW / moderate W-NW / moderate-steep SE and NW / mild-steep
Total well depth 221 300 300 501 300 285
Water level and flow characteristics
Total water level fluctuation in bedrock
well, February 2003 to August 2008 3.3 7.8 2.9 2.3 3.9 1
Hydraulic conductivity (ft/d) and well
yield (gpm) of upper regolith wells 30 / 5 3 / 1 0.5 / 0.5 2 / 2 2 / 3 0.2 / 1
Hydraulic conductivity (ft/d) and well
yield (gpm) of lower regolith wells 10 / 15 5 / 5 5 / 2 nd / 20 10 / nd nd / 1
Hydraulic conductivity (ft/d) and well
yield (gpm) of open hole bedrock wells 4 / 40 nd / 0.5 0.04 / < 0.5 0.2 / 6 nd / 1 nd / <0.5
Setting
Proximity to stream/drainage
70 ft SE to Bent Creek
(regolith wells); 170 ft SE
to Bent Creek (bedrock
well)
150 ft E to Boyd Br and
150 ft N to Boyd Branch;
600 ft E to Bent Creek
80 ft W to unnamed trib
to Boyd Br; 130 ft NE to
Boyd Br; 1100 ft SE to
Bent Creek
190 ft NE to Boyd
Branch; 2700 ft SE to
Bent Creek
30 ft S to unnamed trib to
Boyd Br; 250 ft to Boyd
Branch; 3400 ft SE to
Bent Creek
100 ft NE to Boyd Branch;
4600 ft SE to Bent Creek
Hydrologic setting discharge discharge recharge recharge recharge recharge
BC-1 BC-2 BC-3 BC-4 BC-5 BC-6
Rock Type ms-bgn; interlayered and
migmatitic
ms-bgn; interlayered
and migmatitic
bgn-ms; interlayered
and migmatitic
ms-migmatite;
interlayered
bgn-ms; interlayered and
migmatitic
ms-bgn; interlayered and
migmatitic
Regolith
Thickness of saprolite, ft 2 12 38 26 33 10
Thickness of TZ, ft 46e 25e 17 5 17 19
Thickness of regolith, ft 48 43 60 50d 53 38
Type of weathering profile in regolith very thin saprolite even gradation even gradation even gradation interlayered saprolite-TZ even gradation
Bedrock
Depth to bedrock, ft bls 48 43 60 50 53 38
Foliated or non-foliated foliated foliated foliated foliated foliated foliated
Degree of foliation moderate moderate moderate moderate moderate moderate
Grainsize fine to medium medium fine to medium medium medium medium to fine
Foliation strike direction NE NE NE NE NE and N nd
Foliation dip azimuth and angle SE / moderate-steep SE / moderate-steep NW and SE / shallow-
moderate-steep
NW and SE / shallow-
moderate-steep
W and NW / moderate-
steep nd
Well characteristics
primarya / secondaryb primary / secondary primary / secondary primary / secondary primary / secondary primary / secondary
Number of fractures: 50 to 100 ft bls 0 / 3 0 / 7 1 / 2 0 / 7 0 / 6 nd
Number of fractures: 100 to 150 ft bls 0 / 13 0 / 2 0 / 3 2 / 0 0 / 2 nd
Number of fractures: 150 to 200 ft bls 1 / 7 0 / 3 0 / 7 0 / 3 0 / 3 nd
Number of fractures: 200 to 250 ft bls 0 / 2 0 / 2 0 / 3 0 / 2 0 / 5 ndNumber of fractures: 250 to 300 ft bls 0 / 0 0 / 9 ndNumber of fractures: 300 to 350 ft bls 0 / 2 ndNumber of fractures per foot 0.006 / 0.15 0 / 0.06 0.004 / 0.06 0.004 / 0.03 0 / 0.10 nd
Predominant fracture strike direction N (primary)NE (secondary)NE (primary)NE (primary)N-NE (secondary)nd
Predominant fracture dip azimuth & angle W / low SE / moderate-steep SE / moderate NW / moderate W-NW / moderate-steep nd
Total well depth 221 300 300 501 300 190
Water level and flow characteristics
Total water level fluctuation in bedrock
well, February 2003 to August 2008 3.3 7.8 2.9 2.3 3.9 nd
Hydraulic conductivity (ft/d) and well
yield (gpm) of upper regolith wells 30 / 5 3 / 1 0.5 / 0.5 2 / 2 2 / 3 nd
Hydraulic conductivity (ft/d) and well
yield (gpm) of lower regolith wells 10 / 15 5 / 5 5 / 2 nd / 20 10 / nd nd
Hydraulic conductivity (ft/d) and well
yield (gpm) of open hole bedrock wells 4 / 40 nd / 0.5 0.04 / < 0.5 0.2 / 6 nd / 1 nd / 3
Setting
Proximity to stream/drainage
70 ft SE to Bent Creek
(regolith wells); 170 ft SE
to Bent Creek (bedrock
well)
150 ft E to Boyd Br and
150 ft N to Boyd Branch;
600 ft E to Bent Creek
80 ft W to unnamed trib
to Boyd Br; 130 ft NE to
Boyd Br; 1100 ft SE to
Bent Creek
190 ft NE to Boyd
Branch; 2700 ft SE to
Bent Creek
30 ft S to unnamed trib to
Boyd Br; 250 ft to Boyd
Branch; 3400 ft SE to
Bent Creek
400 ft E to Boyd Branch;
4400 ft SE to Bent Creek
Hydrologic setting discharge discharge recharge recharge recharge recharge
Notes: TZ, transition zone; ft bls, feet below land surface; nd, not determined; strikes and dips determined using optical televiewer data; aprimary fractures are defined as open; bsecondary
fractures are defined as weathered but not open; cupper regolith well BC-1S is screened across alluvium; ddepth to bedrock in core collected at site BC-4, about 75 ft distant, is ~ 45 ft; econtains alluvium/colluvium
Table 5. Geologic and hydrogeologic summary information obtained from rock coring and well testing at Bent Creek Research Station, North
Carolina.
45
45
c
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
37

Figure 23. Observed water levels (rounded) on May 24, 2005, in vicinity of BC-4, Bent
Creek Research Station, North Carolina.
200 ft
N

Enlargement
shown below.
Boyd Branch
Piezometer or piezometer cluster
Well cluster
51 Water level elevations (rounded) observed May
24, 2005. Note: Add 2200 to the displayed
value to obtain elevation in ft above mean sea
level (applies to map and inset map).
Temporary
well point
54
55
56
52 51
51 50
45
20
63
95
46
95
51
02
-16
157

Foliation
strike
direction
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
38
that storage occurs primarily within the regolith. Regolith thickness ranged from 33 to 60 feet (median =
48 ft) in the 7 core locations and is expected to be somewhat thinner at higher elevations that were not
cored; although regolith thickness will vary across the watershed, a uniform value of 35 feet was
selected to represent average conditions. The average depth to water for the 27 observation
wells/piezometers was 8 ft; depth to water is expected to be greater in higher elevations in the basin, so
an average value of 12 feet was selected to represent average conditions. As referenced in table 8, the
porosity of materials in the upper regolith (mostly saprolite) is estimated to range from about 0.3 to 0.6,
with an average value of 0.5; the porosity of materials in the lower regolith (transition zone) is estimated
to range from about 0.15 to 0.4, with an average value of 0.2. A representative value of 0.4 was
selected as an average for the full, saturated thickness of the saturated regolith system across the
basin. Using these values and assumptions, a typical value for total groundwater storage in the 1.1 mi2
Boyd Branch basin during the climatic conditions and water level fluctuations that occurred between
2003 and 2008, was estimated to be about 1 to 3 billion gallons. However, not all of the water in
storage may be released or mined, as hydraulic conductivities generally are much lower in clayey
materials (such as at well cluster BC-7, for example) where porosities may be high.

Bedrock Flow System

In the deeper consolidated rock, groundwater flow is dependent upon the number, size, and
aspect of fracture openings and their connectedness with fracture networks and the overlying regolith.
This is because crystalline bedrock mostly lacks primary porosity5. The abundance and size of
fractures are expected to decrease with depth, becoming less than 1 percent at depths below 750 feet
due to lithostatic pressure (Daniel, 1989, 1992); this trend could not be evaluated in the study area
because six of the seven bedrock wells were drilled to only 300 feet or less.
The bedrock flow system at BCRS is characterized by low porosity and permeability, with
relatively few primary fractures6 in rock cores and well bores. Well yields ranged from less than 0.5 to
40 gpm, with a median of 1 gpm (table 5) and an average of 7 gpm. These yields are far lower than the
average yield of 16 to 22 gpm reported in similar hydrogeologic units in the mountain and piedmont
regions of North Carolina (Daniel, 1989). This may be due to the fact that (1) foliation and fracture
openings intercepted by bedrock wells at BCRS tend to be moderately to steeply dipping, and thus are
less likely to intersect with and draw from the relatively few steeply dipping joints that are efficient
conduits to abundant groundwater storage supplies from the regolith, and (2) mica schist is the
dominant lithology underlying BCRS, and, conceptually, fewer rock openings may be expected in this
rock type whose laminated cleavage planes can readily fold and deform without fracturing. Yields did
not appear to be related to topographic setting. The K values obtained from slug tests in bedrock wells
(4, 0.04, and 0.2 ft/day, for BC-1D, -3D, and -4D, respectively, with a median of 0.2 ft/day) were lower
than the K values in regolith wells by over an order of magnitude.
Only 6 primary fractures were observed in the six bedrock wells using optical televiewer images
and borehole geophysical logs (two in well BC-4D, and one each in wells BC-1D and -3D) (table 5 and
appendix 2). The low number of primary fractures observed is consistent with conditions in a watershed
underlain by mica rich schist whose laminated cleavage planes can readily fold and deform without
fracturing. Of the six bedrock wells tested for ambient flow through the open borehole using a heat
pulse flowmeter, only two had measurable readings (appendix 2); an ambient inflow of 0.1 gpm was
observed in BC-2D at 53 ft, and an ambient outflow of 0.01 gpm was observed in BC-3D at 60 ft. It is
possible that these very low flows were associated with well construction anomalies and not the natural
formation because they occurred at the interface between the surface casing and bedrock, a position
commonly associated with potential leakage to or from the regolith.

5 Porosity in fractured bedrock typically ranges from 1 to 10 percent (Freeze and Cherry, 1979), with lower values
being far more common. 6 Primary fractures are defined here as open and significant.
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
39

Table 6. Estimated aquifer properties in the regolith flow system, Bent Creek Research
Station, North Carolina.

Porosity
Soil type Publisheda porosity, n, in %
Estimated regolith porosity, n,
in %
clay 40 to 70
silt 35 to 50
sand 25 to 50 40b
sand and gravel 15 to 35
gravel 20 to 40
Selected representative porosity = 0.5 (saprolite)
Selected representative porosity = 0.3 (TZ)
Hydraulic Conductivity
Well
Method used to estimate hydraulic
conductivity
Hydraulic conductivity, K, in
feet/day
BC-2S (Saprolite)slug test 3
BC-3S (Saprolite)slug test 0.5
BC-4S (Saprolite)slug test 2
BC-5S (Saprolite)slug test 2
BC-7S (Saprolite)slug test 0.2
BC-1S (TZ)slug test 30
BC-1I (TZ)slug test 10
BC-2I (TZ)slug test 5
BC-3I (TZ)slug test 5
BC-5I (TZ)slug test 10
Published value for soil types
Publishedc hydraulic conductivity,
K, values in feet/day
silt, sandy silts, clayey sandsd 0.003 to 0.3
silty sands, fine sandsd 0.3 to 3 3b
well-sorted gravele 30 to 3000
Aquifer test results in regolithf median K of 7 analyses 14
Selected representative K = 1 (saprolite)
Selected representative K = 10 (TZ)
TZ, transition zone
a Driscoll (1986), Freeze and Cherry (1979), and Roscoe Moss (1990)
b Chosen Representative value for study area, based on soil types and associated published values
c Fetter, C.W., Applied Hydrogeology, 2001, Prentis Hall, Inc., p.85.
d Soil types that make up upper regolith (saprolite) in study area
d, e Soil types that make up lower regolith (transition zone) in study area
f From 72-hour aquifer test conducted October 2004: median transmissivity ~ 640 ft2/day,
with a saturated thickness ~ 45 ft
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
40

The existence of at least one primary fracture in the bedrock well bore generally was a good
indicator that the well would produce significantly higher yields than wells without a primary fracture.
For example, the two highest yielding wells, BC-1D (yield = 40 gpm) and BC-4D (yield = 6 gpm), have
one primary and 25 secondary fractures, and two primary and 14 secondary fractures, respectively.
The three wells that lack a primary fracture, BC-2D, -5D, and -7D, had yields of only 0.5, 1, and less
than 0.5 gpm, respectively.
The number of observed secondary fractures was not predictive of relative bedrock well yield.
For example, low yielding wells, BC-5D (yield = 1 gpm) and BC-7D (less than 0.5 gpm), were
associated with the highest number of secondary fractures (25 and 35, respectively)(table 5). Similarly,
the number of fractures per foot was not predictive of relative bedrock well yield (table 5).
Based on borehole geophysical and heat pulse flowmeter data, a good qualitative predictor of
elevated well yield (defined here as 6 gpm or higher) was a combination of primary fracture(s) and a
sharp caliper increase at a given depth. The highest yielding BCRS well, BC-1D (40 gpm), has both a
primary fracture and a sharp caliper increase at 140 feet (appendix 2 and table 5). The second highest
yielding well, BC-4D (6 gpm), has two primary fractures and a sharp caliper increase at 110 ft. In both
cases flowmeter testing showed an increased flow rate at these depths. On the other hand, wells with
only one of the two predictive features (for example BC-3D and BC-5D) had low yields (table 5).
Regolith thickness7 was not predictive of relative bedrock well yield (table 5). For example, well
cluster BC-3 had the thickest observed regolith (60 ft) of all six well cluster locations but the lowest
bedrock well yield (less than 0.5 gpm). And well cluster BC-4 had the thinnest regolith (33 ft) of all six
locations but the second highest bedrock well yield at 6 gpm. Factors affecting well yield include
fracture size, density, and connectedness to overlying storage. Factors affecting long-term
sustainability of this yield include (1) long-term rainfall and drought patterns, (2) regolith thickness,
porosity, and permeability in upslope areas that recharge the well, (3) withdrawals from interconnected
bedrock wells, and (4) withdrawals from wells in upslope recharge areas.

7 Regolith thickness was determined from rock core collected from 30 to 70 feet away from the bedrock well, and
aquifer characteristics may vary significantly in relatively short distances in fractured rock terrains.
Table 7. Horizontal and vertical hydraulic gradients, Bent Creek Research Station, North Carolina.
Locations used in gradient
observation head difference distance
horizontal
hydraulic
gradient
P7 to P4 45 ft 1095 ft 0.04
P4 to BC-3 17 ft 1070 ft 0.02
BC-3 to BC-2 18 ft 785 ft 0.02
P8 to BC-4 2 ft 222 ft 0.01
BC-7 to P7 93 ft 1800 ft 0.05
Locations used in gradient
observation head difference distance
vertical
hydraulic
gradient
PC-3S to BC-3I 0.4 ft 20 ft 0.02
BC-4S to BC-4I 0.5 ft 18 ft 0.03
BC-5S to BC-5I 0.5 ft 23 ft 0.02
BC-7S to BC-7I 0.7 ft 20 ft 0.04
BC-1S to BC-1I 0.2 ft 13 ft 0.02
P2S to P2I 2 ft 37 ft 0.05
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
41
Seasonal and Storm-Induced Groundwater Level Fluctuations

Water levels were measured monthly from February 2003 to August 2008 in six well clusters
open to the upper regolith, lower regolith, and bedrock flow zones to observe seasonal changes and
horizontal and vertical head gradients. Water levels also were measured hourly for several months at
two representative well clusters (BC-7 in a recharge setting and BC-2 in a discharge setting), and at
selected intervals at four other well clusters following rain events, to observe the response to storm
events. Hourly rainfall data were obtained from NOAA gage 0246CA located 1.2 miles N-NE of BCRS.
The water table was about 2 to 15 feet bls during the study period depending on topographic
setting, rainfall, and season (table 7 and fig. 24). Noted exceptions to this were well BC-2D, a flowing,
artesian, open hole bedrock well in a discharge area with a water level about 1 to 9 feet above land
surface, and well BC-6D, a midslope open hole bedrock well on a topographic rise a relatively distant
300 feet from Boyd Branch with a water level of 31 to 33 feet bls. The water table was closer to land
surface in lower elevations of the basin (sites 1, 2, and 3, for example) than in higher elevations (site 7,
for example) (fig. 24).

Most wells fluctuated 1 to 2 ft per year in response to seasonal fluctuations in precipitation and
evapotranspiration, and water levels in the three parts of the groundwater system (upper regolith, lower
regolith, and bedrock) followed similar patterns of fluctuation (figs. 25A to 25F)8. Water levels generally
declined during late spring and summer months when soil moisture deficits tend to be high and
evapotranspiration rates tend to be higher than rainfall, with the lowest levels observed in late summer
and highest levels in late winter (figs. 25A to 25F).
Water level fluctuations usually were greater in regolith wells than in bedrock wells (figs. 25A to
25F and tables 8 and 9), due in part to the more immediate and direct response of regolith wells to
rainfall recharge and due to the fact that water tended to be under at least some pressure in the semi-
confined bedrock system. Water levels declined about 2 to 3 ft during the measurement period, with
highs occurring in May 2003 (a year in which some areas of the State received over 50 percent higher
rainfall than normal) and lows occurring in August 2008 (figs 25A to 25F), a year characterized by

8 The fluctuation measured at a well is dependent upon the hydrogeologic setting, climate, and time of
measurement.
Figure. 24. Maximum and minimum water levels (in feet below land surface) as measured monthly
between 2003 and 2008, and depth to water (shown as horizontal dash) measured on August 14,
2008, Bent Creek Research Station, North Carolina.

-10
-5
0
5
10
15
20
25
30
35
BC
-
1
S
BC
-
1
I
BC
-
1
D
BC
-
2
S
BC
-
2
I
BC
-
2
D
BC
-
3
S
BC
-
3
I
BC
-
3
D
BC
-
4
S
BC
-
4
I
BC
-
4
D
BC
-
5
S
BC
-
5
I
BC
-
5
D
BC
-
6
D
BC
-
7
S
BC
-
7
I
BC
-
7
D
De
p
t
h

t
o

w
a
t
e
r
,

f
e
e
t

b
e
l
o
w

l
a
n
d

s
u
r
f
a
c
e
Direction of increasing land surface elevation
BC-6D is located on an upslope rise.
Values shown here are based on only 8
measurements during a relatively dry
period between December 2005 and
August 2008.
Land surface
Dash represents value
measured on August 14,
2008.
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
42
regional drought conditions and historically low stream flows (USGS website accessed January 12,
2009; http://nc.water.usgs.gov/drought/).

2198
2199
2200
2201
2202
2203
2204
Fe
b
-
0
3
Ma
y
-
0
3
Au
g
-
0
3
No
v
-
0
3
Fe
b
-
0
4
Ma
y
-
0
4
Au
g
-
0
4
No
v
-
0
4
Fe
b
-
0
5
Ma
y
-
0
5
Au
g
-
0
5
No
v
-
0
5
Fe
b
-
0
6
Ma
y
-
0
6
Au
g
-
0
6
No
v
-
0
6
Fe
b
-
0
7
Ma
y
-
0
7
Au
g
-
0
7
No
v
-
0
7
Fe
b
-
0
8
Ma
y
-
0
8
Au
g
-
0
8
Wa
t
e
r

l
e
v
e
l

a
l
t
i
t
u
d
e
,

i
n

f
t

a
b
o
v
e

s
e
a

l
e
v
e
l
0
4
8
12
16
20
24
28
32
36
40
Mo
n
t
h
l
y

r
a
i
n
f
a
l
l
,

i
n
c
h
e
s
BC-3S
BC-3I
BC-3D
monthly rain, in

Land surface elevation at BC-3S ~ 2210.1 ft asl
Land surface elevation at BC-3I ~ 2209.5 ft asl
Land surface elevation at BC-3D ~ 2209.1 ft asl
2242
2244
2246
2248
2250
2252
2254
Fe
b
-
0
3
Ma
y
-
0
3
Au
g
-
0
3
No
v
-
0
3
Fe
b
-
0
4
Ma
y
-
0
4
Au
g
-
0
4
No
v
-
0
4
Fe
b
-
0
5
Ma
y
-
0
5
Au
g
-
0
5
No
v
-
0
5
Fe
b
-
0
6
Ma
y
-
0
6
Au
g
-
0
6
No
v
-
0
6
Fe
b
-
0
7
Ma
y
-
0
7
Au
g
-
0
7
No
v
-
0
7
Fe
b
-
0
8
Ma
y
-
0
8
Au
g
-
0
8
Wa
t
e
r

l
e
v
e
l

a
l
t
i
t
u
d
e
,

i
n

f
t

a
b
o
v
e

s
e
a

l
e
v
e
l
0
2
4
6
8
10
12
14
16
18
20
22
24
Mo
n
t
h
l
y

r
a
i
n
f
a
l
l
,

i
n
c
h
e
s
BC-4S
BC-4I
BC-4D
monthly rain, in
NOAA rain
gage 0246C
Land surface elevation at BC-4S ~ 2259.7 ft asl
Land surface elevation at BC-4I ~ 2258.8 ft asl
Land surface elevation at BC-4D ~ 2258.5 ft asl
Figure 25B. Monthly water level measurements at cluster wells at site BC-4, a recharge area, in relation to
monthly rainfall, Bent Creek Research Station, February 2003 to August 2008.

Figure 25A. Monthly water level measurements at cluster wells at site BC-3, a recharge area, in
relation to monthly rainfall, Bent Creek Research Station, February 2003 to August 2008.

Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
43
2292
2293
2294
2295
2296
2297
2298
Fe
b
-
0
3
Ma
y
-
0
3
Au
g
-
0
3
No
v
-
0
3
Fe
b
-
0
4
Ma
y
-
0
4
Au
g
-
0
4
No
v
-
0
4
Fe
b
-
0
5
Ma
y
-
0
5
Au
g
-
0
5
No
v
-
0
5
Fe
b
-
0
6
Ma
y
-
0
6
Au
g
-
0
6
No
v
-
0
6
Fe
b
-
0
7
Ma
y
-
0
7
Au
g
-
0
7
No
v
-
0
7
Fe
b
-
0
8
Ma
y
-
0
8
Au
g
-
0
8
Wa
t
e
r

l
e
v
e
l

a
l
t
i
t
u
d
e
,

i
n

f
t

a
b
o
v
e

s
e
a

l
e
v
e
l
0
4
8
12
16
20
24
28
32
36
40
Mo
n
t
h
l
y

r
a
i
n
f
a
l
l
,

i
n
c
h
e
s
BC-5S
BC-5I
BC-5D
monthly rain, in

2355
2356
2357
2358
2359
2360
Fe
b
-
0
3
Ma
y
-
0
3
Au
g
-
0
3
No
v
-
0
3
Fe
b
-
0
4
Ma
y
-
0
4
Au
g
-
0
4
No
v
-
0
4
Fe
b
-
0
5
Ma
y
-
0
5
Au
g
-
0
5
No
v
-
0
5
Fe
b
-
0
6
Ma
y
-
0
6
Au
g
-
0
6
No
v
-
0
6
Fe
b
-
0
7
Ma
y
-
0
7
Au
g
-
0
7
No
v
-
0
7
Fe
b
-
0
8
Ma
y
-
0
8
Au
g
-
0
8
Wa
t
e
r

l
e
v
e
l

a
l
t
i
t
u
d
e
,

i
n

f
t

a
b
o
v
e

s
e
a

l
e
v
e
l
0
4
8
12
16
20
24
28
32
36
40
Mo
n
t
h
l
y

r
a
i
n
f
a
l
l
,

i
n
c
h
e
s
BC-7S
BC-7I
BC-7D
monthly rain, in

Land surface elevation at BC-5S ~ 2300 ft asl
Land surface elevation at BC-5I ~ 2302.2 ft asl
Land surface elevation at BC-5D ~ 2204.8 ft asl
Figure 25C. Monthly water level measurements at cluster wells at site BC-5, a recharge area, in relation to
monthly rainfall, Bent Creek Research Station, February 2003 to August 2008.

Land surface elevation at BC-7S ~ 2368.2 ft asl
Land surface elevation at BC-7I ~ 2369 ft asl
Land surface elevation at BC-7D ~ 2369.9 ft asl
Figure 25D. Monthly water level measurements in cluster wells at site BC-7, a recharge area, in relation to
monthly rainfall, Bent Creek Research Station, February 2003 to August 2008.

Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
44
2196
2197
2198
2199
2200
Fe
b
-
0
3
Ma
y
-
0
3
Au
g
-
0
3
No
v
-
0
3
Fe
b
-
0
4
Ma
y
-
0
4
Au
g
-
0
4
No
v
-
0
4
Fe
b
-
0
5
Ma
y
-
0
5
Au
g
-
0
5
No
v
-
0
5
Fe
b
-
0
6
Ma
y
-
0
6
Au
g
-
0
6
No
v
-
0
6
Fe
b
-
0
7
Ma
y
-
0
7
Au
g
-
0
7
No
v
-
0
7
Fe
b
-
0
8
Ma
y
-
0
8
Au
g
-
0
8
Wa
t
e
r

l
e
v
e
l

a
l
t
i
t
u
d
e
,

i
n

f
t

a
b
o
v
e

s
e
a

l
e
v
e
l
0
4
8
12
16
20
24
28
32
36
40
Mo
n
t
h
l
y

r
a
i
n
f
a
l
l
,

i
n
c
h
e
s
BC-1S
BC-1I
BC-1D
monthly rain, in

2182
2184
2186
2188
2190
2192
2194
2196
2198
Fe
b
-
0
3
Ma
y
-
0
3
Au
g
-
0
3
No
v
-
0
3
Fe
b
-
0
4
Ma
y
-
0
4
Au
g
-
0
4
No
v
-
0
4
Fe
b
-
0
5
Ma
y
-
0
5
Au
g
-
0
5
No
v
-
0
5
Fe
b
-
0
6
Ma
y
-
0
6
Au
g
-
0
6
No
v
-
0
6
Fe
b
-
0
7
Ma
y
-
0
7
Au
g
-
0
7
No
v
-
0
7
Fe
b
-
0
8
Ma
y
-
0
8
Au
g
-
0
8
Wa
t
e
r

l
e
v
e
l

a
l
t
i
t
u
d
e
,

i
n

f
t

a
b
o
v
e

s
e
a

l
e
v
e
l
0
10
20
30
40
50
60
70
80
Mo
n
t
h
l
y

r
a
i
n
f
a
l
l
,

i
n
c
h
e
s
BC-2S
BC-2I
BC-2D
monthly rain, in

Land surface elevation at BC-1S ~ 2201 ft asl
Land surface elevation at BC-1I ~ 2202.5 ft asl
Land surface elevation at BC-1D ~ 2201.8 ft asl
Figure 25E. Monthly water level measurements at cluster wells at site BC-1, a discharge area, in relation to
monthly rainfall, Bent Creek Research Station, February 2003 to August 2008.

Land surface elevation at BC-2S ~ 2190.7 ft asl
Land surface elevation at BC-2I ~ 2191.7 ft asl
Land surface elevation at BC-2D ~ 2190.2 ft asl
Figure 25F. Monthly water level measurements at cluster wells at site BC-2, a discharge area, in relation to
monthly rainfall, Bent Creek Research Station, February 2003 to August 2008.

Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
45

Table 8. Low and high water levels, average water levels, and total fluctuation during the
period of record, and depth to water on August 14, 2008, in wells at Bent Creek Research
Station, North Carolina.
Well Low water table High water table Average water table
Range (difference
between high and low
water table)
Depth to water on
August 14, 2008
DTW, in ft bls DTW, in ft bls DTW, in ft bls ft ft bls
BC-1Sa 5.0 3.2 4.0 1.8 5.0
BC-1Ia 6.5 3.1 5.2 3.4 6.5
BC-1Da 6.0 2.7 4.7 3.3 6.0
BC-2Sa 8.4 4.9 6.7 3.6 8.4
BC-2Ia 9.1 6.1 7.6 3.0 9.1
BC-2Da [8.9]d [1.2]d [6.0]d 7.8 [7.26]d
BC-3Sa 11.1 7.0 8.6 4.1 11.1
BC-3Ia 10.5 6.9 8.3 3.6 10.5
BC-3Da 10.5 7.7 9.0 2.9 10.5
BC-4Sa 10.4 6.8 8.7 3.6 10.4
BC-4Ia 10.0 6.8 8.4 3.2 10.0
BC-4Da 13.2 10.9 11.9 2.3 13.2
BC-5Sa 7.6 3.2 5.1 4.4 7.6
BC-5Ia 9.9 6.0 7.7 3.9 9.9
BC-5Da 12.7 8.8 10.5 3.9 12.7
BC-6Db 33.9 31.0 32.1 2.9 33.5
BC-7Sa 12.1 9.2 11.1 2.9 12.1
BC-7Ia 13.2 12.0 12.7 1.2 13.2
BC-7Da 14.1 13.1 13.6 1.0 14.1
P1Sc 10.6 8.2 8.6 2.4 10.6
P1Ic 10.3 7.7 8.3 2.6 10.3
P2Sc 8.6 5.6 6.2 3.0 8.6
P2Ic 9.1 7.5 7.8 1.6 9.1
P3Sc 8.6 6.0 6.5 2.7 8.6
P4Sc 14.2 11.9 12.5 2.2 14.2
P4Ic 14.6 12.7 13.1 1.9 14.6
P5Sc 9.1 6.7 7.2 2.4 9.1
P5Ic 8.2 7.3 7.5 0.9
P6Sc 8.7 6.5 6.9 2.2 8.7
P6Ic 8.6 6.4 6.7 2.2 8.6
P7Sc 10.0 7.4 7.9 2.5 10.0
P7Ic 11.5 8.8 9.3 2.7 11.5
P8Sc 5.1 2.7 3.1 2.4 5.1
P8Ic 4.7 2.4 2.8 2.3 4.7
Average, all wells 10.5 7.8 8.8 2.8 10.5
Median, all wells 10.0 6.9 7.9 2.7 10.0
Minimum, all wells 4.7 2.4 2.8 0.9 4.7
Maximum, all wells 33.9 31.0 32.1 7.8 33.5
aBased on measurements made approximately monthly (n = 46 measurements) between February 2003 and November 2008;
bBased on measurements made approximately monthly (n = 20 measurements) between December 2005 and November 2008;
cBased on measurements made approximately monthly (n = 21 measurements) between February 2005 and August 2008;
dvalue shown is feet above land surface; DTW, depth to water; ft bls, feet below land surface
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
46

Storm events produced a rapid, temporary increase in water levels in regolith and bedrock
wells. At BC-7S, located in a midslope recharge area, storm-induced water level increases of about 1
foot were observed within 4 to 6 hours of numerous rainfall events throughout the period of hourly
record (examples are shown in figs. 26). At BC-2S and -2I, located in a low-lying discharge area,
storm-induced water level increases of about 1 to 3 ft were observed within 3 to 4 hours of numerous
rainfall events throughout the period of hourly record (examples are shown in fig. 27). In general, the
temporary storm-induced rise in both wells returned to within about 20 percent of the pre-storm levels
within about 4 to 8 days. The amount of water level rise is a function of atmospheric pressure, rainfall
amount, rainfall intensity, antecedent soil moisture, and geologic conditions. Water levels measured
manually up to a day after storm events also showed similar rises in upper regolith, lower regolith, and
bedrock wells at clusters P89, BC-4, and BC-1 (written communication, T. Campbell, January 20, 2011).

Vertical Head Gradients Between Upper Regolith, Lower Regolith, and Bedrock
Flow Systems

Vertical head gradients observed at clustered wells were used to evaluate vertical movement of
groundwater between different zones of the flow system. Vertical head gradients from the regolith
system downward to the bedrock system were observed at sites BC-3, -4, -5, and -7 (areas of
recharge) (fig. 4 and fig. 25A-D). At BC-2, a discharge area associated with the flowing, artesian
bedrock well BC-2D, vertical head gradients were strongly upward from the bedrock system to the
regolith system, with a potentiometric surface about eight feet above land surface (fig. 25F). The
vertical head gradient between the upper and lower regolith wells BC-2S and -2I tended to be very
slightly upward (differences were typically less than a tenth of a foot), except in response to storm
events when the gradient could be upward or downward and up to 0.7 ft (fig. 27). These general trends
are common in the Piedmont and Blue Ridge Provinces and follow the conceptual model described by
LeGrand (2004), Harned and Daniel (1992), and Heath (1983, 1984).
At BC-1, a flat, low-lying highly permeable discharge area adjacent to Bent Creek (fig. 4), a
slight upward vertical head gradient was observed between upper and lower regolith wells BC-1S and -
1I (typically about 0.3 ft, fig. 25E). Head gradients were slightly downward (typically about 0.2 ft) from
the lower regolith well BC-1I to the bedrock well BC-1D, illustrating the fact that the potentiometric head
in a bedrock well is a composite of the hydraulic heads of all fractures tapped by the well and, as such,
may not represent conditions throughout the local discharge area. This suggests that care should be
taken when evaluating recharge versus discharge conditions solely on the basis of vertical head

9 A bedrock well not installed at P8.
Table 9. Summarized water level fluctuations during the period the period of record
for regolith wells and bedrock wells at Bent Creek Research Station, North
Carolina.

regolith wells open hole bedrock wellsb
maximum 4.4 (BC-5S)3.9 (BC-5D)
minimum 0.9 (P5I)1.0 (BC-7D)
median 2.3 1.9
a Water levels as measured approximately monthly during period from February 2003 to November
2008; b Does not include BC-2D, an artesian, flowing well which had a range of 7.8 ft during the
period of record.
Range (Maximum Minus Minimum for a Given Well) in Water Levels Over Period of Recorda, in feet
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
47
gradients in a single cluster of wells. A more complete discussion of vertical flow at this site is
presented in the section titled „Seasonal and Storm-Induced Groundwater Temperature Fluctuations‟.

0.00
0.05
0.10
0.15
0.20
0.25
0.3010
10.5
11
11.5
12
12.5
13
13.5
14
12
:
0
0

A
M
2:
0
0

A
M
4:
0
0

A
M
6:
0
0

A
M
8:
0
0

A
M
10
:
0
0

A
M
12
:
0
0

P
M
2:
0
0

P
M
4:
0
0

P
M
6:
0
0

P
M
8:
0
0

P
M
10
:
0
0

P
M
12
:
0
0

A
M
2:
0
0

A
M
4:
0
0

A
M
6:
0
0

A
M
8:
0
0

A
M
10
:
0
0

A
M
12
:
0
0

P
M
2:
0
0

P
M
4:
0
0

P
M
6:
0
0

P
M
8:
0
0

P
M
10
:
0
0

P
M
Ho
u
r
l
y

r
a
i
n
f
a
l
l
,

i
n
c
h
e
s
Wa
t
e
r

L
e
v
e
l
,

f
e
e
t

b
e
l
o
w

l
a
n
d

s
u
r
f
a
c
e
November 9-11, 2009
Storm-induced groundwater fluctuations in BC-7S
0
1
210.5
11.0
11.5
12.0
12.5
Au
g
-08
Se
p
-08
Se
p
-08
Oc
t
-08
No
v
-08
No
v
-08
De
c
-08
Ho
u
r
l
y

R
a
i
n
f
a
l
l
,

i
n
c
h
e
s
Wa
t
e
r

L
e
v
e
l
,

f
e
e
t

b
e
l
o
w

l
a
n
d

s
u
r
f
a
c
e
Storm-induced groundwater fluctuations in BC-7S

Fig. 26. Example of storm-induced water level fluctuations in upper regolith well BC-7S
(recharge area), Bent Creek Research Station, North Carolina.
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
48
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
2183
2184
2185
2186
2187
2188
Au
g
-04
Au
g
-04
Se
p
-04
Oc
t
-04
No
v
-04
De
c
-04
Ja
n
-05
Fe
b
-05
Ma
r
-05
Ap
r
-05
Ma
y
-05
Ho
u
r
l
y

r
a
i
n
f
a
l
l
,

i
n
c
h
e
s
Wa
t
e
r

l
e
v
e
l

e
l
e
v
a
t
i
o
n
,

f
e
e
t

a
b
o
v
e

s
e
a

l
e
v
e
l
Storm-induced groundwater fluctuations in BC-2S and -2I
Hourly rainfall
0.0
0.2
0.4
0.6
0.8
1.0
2180
2181
2182
2183
2184
2185
2186
2187
26
-Se
p
-04
27
-Se
p
-04
28
-Se
p
-04
29
-Se
p
-04
30
-Se
p
-04
1-Oc
t
-04
2-Oc
t
-04
3-Oc
t
-04
Ho
u
r
l
y

r
a
i
n
f
a
l
l
,

i
n
c
h
e
s
Wa
t
e
r

l
e
v
e
l

e
l
e
v
a
t
i
o
n
,

f
e
e
t

a
b
o
v
e

s
e
a

l
e
v
e
l
Storm-induced groundwater fluctuations in BC-2S and -2I
BC-2S
BC-2I
BC-2S
BC-2I
Hourly
rainfall

Seasonal and Storm-Induced Groundwater Temperature Fluctuations

Hourly water temperatures were recorded during various periods in several well clusters to
observe seasonal and storm-induced response and inferred recharge, groundwater-surface water
interactions, and vertical groundwater movement. Hourly water temperatures also were recorded in
stream water and streambank sediment pore water of Boyd Branch near P8 (fig. 4). Storm response
was observed using pressure transducer data loggers (recording both water levels and temperatures)
for several months at two representative well clusters (BC-7 in a recharge setting and BC-2 in a
discharge setting).
While water levels in the three flow zones fluctuated seasonally, water temperatures fluctuated
seasonally only in upper regolith wells and in some lower regolith wells. Water levels in the three flow
zones fluctuated in response to storms, but water temperatures responded to storms only in upper
regolith wells. Water temperatures in all measured bedrock wells were stable throughout the
measurement period.
Water temperatures in all upper regolith wells, regardless of hydrologic setting (recharge or
discharge), fluctuated seasonally by at least 3o (and up to 14o F) in response to seasonal air
Fig. 27. Example of storm-induced water level fluctuations in upper and lower regolith
wells BC-2S and -2I (discharge area), Bent Creek Research Station, North Carolina.
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
49
temperatures. However, water temperatures in lower regolith wells (data available only in discharge
areas) fluctuated seasonally by only 1 to 2o F in response to seasonal air temperatures (table 10).
At BC-4, a midslope recharge area (fig. 4) with downward vertical head gradients, water
temperatures recorded from July 2008 to January 2009 in upper regolith well BC-4S lagged seasonal
air temperatures by about 4 months and fluctuated 3o F (ranging from 54o to 57o F)(fig. 28; table 10).
Water temperatures in bedrock well BC-4D remained stable at 55o F throughout the measurement
period, and temperature data were not collected in lower regolith well BC-4I.
Storm-induced temperature fluctuations at BC-4 suggest a close hydraulic connection between
shallow groundwater and rainfall recharge. For example, rainfall on August 26 (1.5 inches in one hour)
and on August 27, 2008 (4 inches in one hour), a time during which air and rainfall temperatures are
seasonally warmer than groundwater temperatures, produced a temporary temperature increase in
upper regolith well BC-4S (fig. 29) and lagged the rainfall event by about 10 hours. This was typical of
the temperature lag time associated with other rainfall events at this location.

4 months
Figure 28. Seasonal temperature fluctuations in upper regolith well BC-4S and bedrock well BC-4D
(recharge area), in relation to air temperatures, July 2008 to January 2009, Bent Creek Research
Station, North Carolina.

50
51
52
53
54
55
56
57
58
8/26/08
12:00 AM
8/26/08
4:48 AM
8/26/08
9:36 AM
8/26/08
2:24 PM
8/26/08
7:12 PM
8/27/08
12:00 AM
8/27/08
4:48 AM
8/27/08
9:36 AM
8/27/08
2:24 PM
8/27/08
7:12 PM
8/28/08
12:00 AM
Te
m
p
e
r
a
t
u
r
e
,

F
0.0
0.5
1.0
1.5
2.0
Ho
u
r
l
y

R
a
i
n
f
a
l
l
,

i
n
c
h
e
s
BC-4S (upper regolith)
BC-4D (bedrock)
Hourly rain, inches,
NOAA raingage 0246C
BC-4D
BC-4S
Figure 29. Storm-induced temperature fluctuations in upper regolith well BC-4S and bedrock well BC-4D
(recharge area), in response to summer storm event, August 26 to 28, 2008, Bent Creek Research
Station, North Carolina.

Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
50

Portion of flow system
Daily Temperature
Fluctuation Seasonal Temperature Fluctuation
Rainfall-Induced Temperature
Fluctuation Vertical potentiometric head gradient
Recharge areas
Site BC-4 Upper regolith, BC-4S none 54 to 57o F (3o) with 4 month lag 2o with 10 hour lag Downward gradient to lower regolith
Lower regolith, BC-4I not measured not measured not measured Downward gradient to bedrock
Bedrock, BC-4D none no seasonal fluctuation (stable 55o F)no rainfall-induced fluctuation ----
Site BC-7 Upper regolith, BC-7S none 55 to 58o F (3o) with 4 month lag no rainfall-induced fluctuation Downward gradient to lower regolith
Lower regolith, BC-7I not measured not measured not measured Downward gradient to bedrock
Bedrock, BC-7D not measured not measured not measured ----
Discharge areas
Site P8 Upper regolith, P8S none 54 to 56o F (2o) with 3 month lag no rainfall-induced fluctuation ----
Lower regolith, P8I none 54 to 58 Fo (4o) with 3 month lag no rainfall-induced fluctuation Upward gradient to upper regolith
Boyd Branch stream water 4 to 9o 32 to 72o F (40o) with 2 hour lag
Bank sediment pore water 1 to 2o 35 to 69o F (34o) with 7 hour lag
Site BC-1 Upper regolith, BC-1S none 49 to 63o F (14o) with 1 month lag 1 to 2o with 1 to 2 day lag
Lower regolith, BC-1I none 54 to 55o F (1o) with 5 month lag no rainfall-induced fluctuation
Upward gradient to upper regolith, and
slight downward gradient to bedrock
Bedrock, BC-1D none no seasonal fluctuation (stable 55o F)no rainfall-induced fluctuation
Site BC-2 Upper regolith, BC-2S none 52 to 58o F (6o) with 3 month lag 1 to 2o with 5 to 10 hour lag ----
Lower regolith, BC-2I none 53 to 54o F (1o) with 7 month lag no rainfall-induced fluctuation
Neutral / slight upward gradient to upper
regolith
Bedrock, BC-2D none no seasonal fluctuation (stable 55o F)no rainfall-induced fluctuation Strong upward gradient to regolith
Other
Air not applicable 10 to 35o 3 to 89o F (86o)not applicable not applicable
Table 10. Temperature fluctuations in different portions of flow system in response to seasonal climate and short-term rainfall events, Bent Creek
Research Station, North Carolina.

4
54 to 57o F (3o)
54 to 56o F (2o) Slight
a
a Site P8 is an area of very slight discharge.
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
52
At BC-7, another midslope recharge area (fig. 4), water temperatures recorded from February
2009 to June 2009 in upper regolith well BC-7S lagged seasonal air temperatures by about 4 months
and fluctuated 3o F (ranging from 55 to at least 58o F)(fig. 30).

4 months
Figure 30. Seasonal temperature fluctuations in upper regolith well BC-7S (recharge area), in relation
to air temperatures, May 2008 to June 2009, Bent Creek Research Station, North Carolina.

At P8, located midslope and about 30 ft from Boyd Branch (fig. 4), water temperatures recorded
from August 2008 to February 2009 in upper regolith piezometer P8S and lower regolith piezometer P8I
lagged seasonal air temperatures by about 3 and 4 months, respectively, and fluctuated about 3o
(ranging from 54o to 57o F) and 2o (ranging from 54o to 56o F), respectively (fig. 31C). P8 was defined
as an area of transition from recharge to discharge, based on very slight to near neutral vertical head
gradients in clustered regolith piezometers measured approximately monthly. About 15 feet away,
water temperatures in Boyd Branch fluctuated daily (typically about 4o to 9o F) and seasonally (ranging
from 32o to 72o F), and lagged air temperatures (ranging from 3o to 89o F) by about 2 hours (figs. 31A
and B). Streambank sediment pore water temperatures fluctuated daily (typically about 1o to 2o F) and
seasonally (ranging from 35o to 69o F), and lagged stream temperatures by about 5 hours (figs. 31A
and B). As expected, in summer months temperatures in stream water and in streambed sediment
pores were warmer than shallow groundwater (fig. 31B). The reverse was true in winter months. Daily
temperature fluctuations were not observed in regolith piezometers P8S and P8I (fig. 31A).

Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
53

Figure 31. (A) Diurnal temperature fluctuations in upper regolith piezometer P8S, lower regolith piezometer
P8I, Boyd Branch stream water, and streambank sediment pore water, August 10 to 14, 2008, and (B) and (C)
seasonal temperature fluctuations in P8S, P8I, Boyd Branch stream water, and streambank sediment pore
water, August 2008 to February 2009, in relation to air temperatures, Bent Creek Research Station, North
Carolina.
48
53
58
63
68
73
78
8/10/2008 8/11/2008 8/12/2008 8/13/2008 8/14/2008
Te
m
p
e
r
a
t
u
r
e
,

F
P8S (upper regolith)
P8i (lower regolith)
BB at P8 (stream)
Left Bank of BB at P8
Air Temp, C
A
P8I
P8S
Boyd Branch
bank sediment
pore water
air
Boyd
Branch
stream
water
10
20
30
40
50
60
70
80
07/21/08 08/20/08 09/19/08 10/19/08 11/18/08 12/18/08 01/17/09 02/16/09
Te
m
p
e
r
a
t
u
r
e
,

F
P8S (upper regolith)
Air, F
BB at P8 (stream)
Left Bank of BB at P8
B
P8S
Boyd Branch bank
sediment pore water
air
Boyd Branch
stream water
10
20
30
40
50
60
70
80
90
Au
g
-
0
8
Se
p
-
0
8
Oc
t
-
0
8
No
v
-
0
8
De
c
-
0
8
Ja
n
-
0
9
Fe
b
-
0
9
Ai
r

T
e
m
p
e
r
a
t
u
r
e
,

F
53.5
54
54.5
55
55.5
56
56.5
57
57.5
58
Gr
o
u
n
d

W
a
t
e
r

T
e
m
p
e
r
a
t
u
r
e
,

F
Air Temp, C
P8S (upper regolith)
P8I (lower regolith)P8S
P8I
air
C
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
54
At BC-1, a flat, low-lying highly permeable discharge area adjacent to Bent Creek (fig. 4),
groundwater in the upper regolith system is inferred to be a mixture of vertically up-welled water from
the lower regolith system and water from the Bent Creek hyporheic zone. This observation is based on
a slight upward vertical head gradient that was observed between upper and lower regolith wells BC-1S
and -1I (typically about 0.3 ft, fig. 25E) and hourly water temperature measurements of the different flow
zones (figs. 32 A and B).
Rainfall on August 26 (1.5 inches in one hour) and on August 27, 2008 (4 inches in one hour), a
time during which air and rainfall temperatures are warmer than groundwater temperatures, produced a
temporary, slight decrease in water temperature in upper regolith well BC-1S (fig. 32). This suggests
that cooler water from the lower regolith (and possibly bedrock system) temporarily up-welled during the
summer precipitation event, an occurrence documented at a groundwater research station in the
Piedmont (Lake Wheeler Road Research Station, Chapman and others, 2005). Conversely, rainfall on
December 10 (2.1 inches in one hour) and December 11, 2008 (1.5 inches in one hour), a time during
which air and rainfall temperatures were mostly cooler than groundwater temperatures, produced a
temporary, slight increase in temperature in groundwater in BC-1S (fig. 32B). This suggests that
warmer water from the lower regolith (and possibly bedrock system) temporarily up-welled during the
winter precipitation event. These observations are consistent with conditions in a discharge area with
upward head gradients such as those at this site. In general, the storm-induced temperature
fluctuations in BC-1S resulting from up-welled water lagged a typical rainfall event by about 1 to 2 days.
The water temperature in lower regolith well BC-1I and bedrock well BC-1D remained stable (about 55o
F) during all storm events (figs. 32A and B).
Seasonal groundwater temperature fluctuations recorded from August 2008 to February 2009 in
upper regolith well BC-1S lagged seasonal air temperatures by about 1 month and fluctuated about 14o
F (ranging from 49o to 63o F)(fig. 33). This is a more rapid climate response than at midslope recharge
areas BC-4 and BC-7 (where shallow water temperatures fluctuated only 3o F and lagged air
temperature by about 4 months) and appears to be due to a shallower water table, a more permeable
aquifer matrix (alluvium, colluviums, and transition zone material), and a closer proximity to and greater
mixing with stream water at BC-1 than at the two midslope locations10. Temperatures recorded in the
lower regolith well BC-1I fluctuated only slightly (ranging from 54o to 55o F), and lagged seasonal air
temperatures by about 5 months (fig. 33). Temperatures in bedrock well BC-1D (about 55o F) remained
stable (fig. 33) throughout the period of measurement.

10 BC-1S is within about 60 ft of Bent Creek (fig. 34). Well BC-1S is screened across highly permeable alluvium in
a discharge area (k = 30 ft/day), whereas BC-4S and BC-7S are screened across less permeable saprolite in
recharge areas (k = 2 and 0.2 ft/day, respectively). Because of its location in an alluvial fan at the base of the
watershed and adjacent to Bent Creek, BC-1S is expected to be in close hydraulic communication with surface
water in Bent Creek.
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
55

46
51
56
61
66
71
76
8/26/08
12:00 AM
8/26/08
4:48 AM
8/26/08
9:36 AM
8/26/08
2:24 PM
8/26/08
7:12 PM
8/27/08
12:00 AM
8/27/08
4:48 AM
8/27/08
9:36 AM
8/27/08
2:24 PM
8/27/08
7:12 PM
8/28/08
12:00 AM
Gr
o
u
n
d

W
a
t
e
r

T
e
m
p
e
r
a
t
u
r
e
,

F
0.0
0.5
1.0
1.5
2.0
Ho
u
r
l
y

R
a
i
n
f
a
l
l
,

i
n
c
h
e
s
BC-1S (upper regolith)
BC-1I (lower regolith)
BC-1D (bedrock)
Air, F
Hourly rain, inches,
NOAA raingage 0246C
B
A
B

36
41
46
51
56
61
12/8/08
12:00 AM
12/9/08
12:00 AM
12/10/08
12:00 AM
12/11/08
12:00 AM
12/12/08
12:00 AM
12/13/08
12:00 AM
12/14/08
12:00 AM
12/15/08
12:00 AM
12/16/08
12:00 AM
12/17/08
12:00 AM
Gr
o
u
n
d

W
a
t
e
r

T
e
m
p
e
r
a
t
u
r
e
,

F
0.0
0.5
1.0
1.5
2.0
Ho
u
r
l
y

R
a
i
n
f
a
l
l
,

i
n
c
h
e
s
BC-1S (upper regolith)
BC-1I (lower regolith)
BC-1D (bedrock)
Air, F
Hourly rain, inches,
NOAA raingage 0246CBC-1S
BC-1D
air
BC-1I
BC-1S
BC-1D
air
BC-1I
Figure 32. Storm-induced temperature fluctuations in upper and lower regolith wells BC-1S and -1I and
bedrock well BC-1D (discharge area), in response to (A) hourly rainfall and air temperatures during a
summer storm event, August 26, 2008, and (B) hourly rainfall and air temperatures during an early winter
storm event, December 8 to 17, 2008, Bent Creek Research Station, North Carolina.
B

Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
56
1 month
5 months
Figure 33. Seasonal temperature fluctuations in upper regolith well BC-1S (discharge area), in relation
to air temperatures, July 2008 to January 2009, Bent Creek Research Station, North Carolina.

Figure 34. Monitor wells and ground water ages at site BC-1
located in low-lying discharge area, Bent Creek Research Station,
North Carolina.

BC-1D
BC-1I
BC-1S
Be
n
t

C
r
e
e
k
N 60 ft
Direction from
high land
elevation to low
land elevation
Direction from
high land
elevation to low
land elevation
.
..
Low angle (16o),
W dipping primary
fracture at depth
of 140 ft
GW age:
Modern to
~ 20 yrs
GW age:
Modern to
~ 30 yrs
GW age:
~20 to 40 yrs
BC-1D
BC-1I
BC-1S
Be
n
t

C
r
e
e
k
N 60 ft
Direction from
high land
elevation to low
land elevation
Direction from
high land
elevation to low
land elevation
.
..
Low angle (16o),
W dipping primary
fracture at depth
of 140 ft
GW age:
Modern to
~ 20 yrs
GW age:
Modern to
~ 30 yrs
GW age:
~20 to 40 yrs
25
yrs 25
yrs
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
57
At BC-2, despite the upward vertical head gradients characteristic of a low-lying discharge area,
water temperature data suggested a close hydraulic connection between groundwater in the upper
regolith system and recharge events. Temperature in upper regolith well BC-2S fluctuated about 1 to 2o
F in response to storm events (figs. 35A and 36A). As such, recharge was inferred to occur to the
shallowest part of the groundwater system, even though the net flux of groundwater at the site is
upward from bedrock to regolith. Temperatures in lower regolith well BC-2I and bedrock well BC-2D
remained stable at about 54o and 55.5o F respectively (figs. 35B and 36B). For example, a storm event
on September 27, 2004 (3.1 inches in 12 hours), a time during which air and rainfall temperatures are
warmer than groundwater temperatures, produced a temporary increase in temperature of about 0.8o F
in shallow groundwater in BC-2S (fig. 35A). A storm event on January 13, 2005 (1.2 inches in 12 hours,
a time during which air and rainfall temperatures are cooler than groundwater temperatures, produced a
temporary decrease in temperature of about 1.5o F in shallow groundwater in BC-2S (fig. 36A). These
upper regolith temperature trends were counter to those in BC-1S in the more permeable discharge
area adjacent to Bent Creek, which was associated with upwelling and mixing with stream water. In
general, the rainfall-induced temperature fluctuations in BC-2S lagged the start of a typical rainfall event
by about 5 to 10 hours.
Temperatures recorded in upper regolith well BC-2S lagged seasonal air temperatures by about
3 months and fluctuated about 6o F (ranging from 52o to 58o F)(fig. 37). This fluctuation is far less than
the 14o observed at BC-1S in the discharge area adjacent to Bent Creek and appears to be due to the
fact that BC-2S is less affected by stream interactions and is underlain by less a less permeable aquifer
matrix than BC-1S11. Water temperatures in lower regolith well BC-2I fluctuated only slightly (ranging
from 53o to 54o F) and lagged seasonal air temperatures by about 7 months (fig. 37). Temperature in
bedrock well BC-2D remained stable at 55o F.

11 BC-2S is located 150 feet from the 2nd order Boyd Branch and is screened across only moderately permeable
saprolite material (k = 3 ft/day), while BC-1S is only 60 feet from the 3rd order Bent Creek and is screened across
mostly alluvium, colluviums, and transition zone material (k = 30 ft/day).
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
58

Figure 35. (A) Storm-induced temperature fluctuations in upper and lower regolith wells BC-2S and -2I
(discharge area), and (B) storm-induced water level fluctuations at BC-2S and BC-2I, in response to
late summer storm event, September 26 to October 3, 2004, Bent Creek Research Station, North
Carolina.
52
53
54
55
56
9/
2
6
/
0
4
9/
2
7
/
0
4
9/
2
8
/
0
4
9/
2
9
/
0
4
9/
3
0
/
0
4
10
/
1
/
0
4
10
/
2
/
0
4
10
/
3
/
0
4
Gr
o
u
n
d

W
a
t
e
r

T
e
m
p
e
r
a
t
u
r
e
,

F
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Ho
u
r
l
y

R
a
i
n
f
a
l
l
,

i
n
c
h
e
s
BC-2S, F
BC-2I, F
Hourly rain,
inches, NOAA
raingage
0246C
2180
2181
2182
2183
2184
2185
2186
2187
2188
9/
2
6
/
0
4
9/
2
7
/
0
4
9/
2
8
/
0
4
9/
2
9
/
0
4
9/
3
0
/
0
4
10
/
1
/
0
4
10
/
2
/
0
4
10
/
3
/
0
4
Wa
t
e
r

L
e
v
e
l
,

f
e
e
t

a
b
o
v
e

s
e
a

l
e
v
e
l
0.0
0.2
0.4
0.6
0.8
1.0
Ho
u
r
l
y

R
a
i
n
f
a
l
l
,

i
n
c
h
e
s
BC-2S WL
BC-2I WL
Hourly rain,
inches,
NOAA
raingage
0246C
A
B
BC-2S temp
BC-2I temp
BC-2S WL
BC-2I WL
hourly rain
hourly rain
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
59

Figure 36. Storm-induced temperature fluctuations in upper and lower regolith wells BC-2S and -2I
(discharge area), and (B) storm-induced water level fluctuations at BC-2S and BC-2I, in response to
winter storm event, January 12 to January 22, 2005, Bent Creek Research Station, North Carolina.
53
54
55
56
1/
1
2
/
0
5
1/
1
3
/
0
5
1/
1
4
/
0
5
1/
1
5
/
0
5
1/
1
6
/
0
5
1/
1
7
/
0
5
1/
1
8
/
0
5
1/
1
9
/
0
5
1/
2
0
/
0
5
1/
2
1
/
0
5
1/
2
2
/
0
5
Gr
o
u
n
d

W
a
t
e
r

T
e
m
p
e
r
a
t
u
r
e
,

F
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Ho
u
r
l
y

R
a
i
n
f
a
l
l
,

i
n
c
h
e
s
BC-2S, F
BC-2I, F
BC-2D, F
Hourly rain,
inches, NOAA
raingage
0246C
2183
2184
2185
2186
1/
1
2
/
0
5
1/
1
3
/
0
5
1/
1
4
/
0
5
1/
1
5
/
0
5
1/
1
6
/
0
5
1/
1
7
/
0
5
1/
1
8
/
0
5
1/
1
9
/
0
5
1/
2
0
/
0
5
1/
2
1
/
0
5
1/
2
2
/
0
5
Wa
t
e
r

L
e
v
e
l
,

f
e
e
t

a
b
o
v
e

s
e
a

l
e
v
e
l
0.0
0.2
0.4
0.6
0.8
1.0
Ho
u
r
l
y

R
a
i
n
f
a
l
l
,

i
n
c
h
e
s
BC-2S WL
BC-2I WL
BC-2D WL
Hourly rain,
inches,
NOAA
raingage
0246C
A
B
BC-2I temp
BC-2S temp
BC-2S WL
BC-2I WL
BC-2D temp
hourly rain
hourly rain
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
60
3 months
7 months
Figure 37. Seasonal temperature fluctuations in upper and lower regolith wells BC-2S and -2I and
bedrock well BC-2D (discharge area), in relation to air temperatures, August 2004 to December 2005,
Bent Creek Research Station, North Carolina.

Aquifer Response to 72-Hour Pumping in Regolith

A 72-hour aquifer pumping test was conducted in lower regolith well BC-4I to evaluate the (1)
the degree of interconnectivity between the upper regolith (saprolite), lower regolith (transition zone),
and bedrock flow systems; (2) the degree of heterogeneity and preferential flow paths within the
system; and (3) the lateral extent of influence due to pumping. Because of the complexity of the three-
part, dual porosity, non-homogeneous, fractured, and anisotropic flow system, a presentation of
analytical modeling and quantitative determinations of aquifer properties was beyond the scope of this
report.
The aquifer system was conceptualized as being under unconfined hydraulic conditions as
evidenced by the very small vertical head gradients in clustered wells (or, where a vertical gradient
exists, the head in the shallow regolith well is greater than the head in the deeper transition zone well)
and the absence of a noted confining layer within the regolith itself. The bottom boundary of the regolith
flow system was fractured bedrock. This boundary was considered to be slightly leaky, as evidenced
by the connectivity between the pumping well BC-4I and the adjacent deep bedrock well BC-4D (total
water level drawdown in BC-4D during pumping was about 1.2 ft), though the quantity of water available
as storage in the fractures themselves was assumed to be minor when compared to the quantity of
water in storage in the regolith. Rainfall recharge to the unconfined aquifer system flows through the
upper and lower regolith and discharges to parts of Boyd Branch or Bent Creek. The remainder
infiltrates to underlying bedrock fractures and moves downgradient in response to hydraulic head
gradients. The upper regolith consists of a variably thick unit of loamy sand, silt, and clay saprolite that
is porous, heterogeneous, and anisotropic and extends to a depth of 20 to 30 ft bls. The lower regolith
is 10 to 20 ft thick and consists of a much more transmissive unit of partially weathered and
unweathered parent rock that also is relatively porous, heterogeneous, and variably thick. Both the
upper and lower regolith are expected to be vertically stratified (due to differential weathering of parent
rock) and anisotropic (due to relict foliations, fractures, and veins). Neither the pumping well nor the
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
61
observation wells fully penetrated the entire saturated regolith thickness so vertically stratified flow was
expected in the early portion of the test. Fractured bedrock occurs at a depth of about 40 to 45 ft bls.
Pumping occurred in well BC-4I, which was 41 feet in depth, 4 inches in diameter, and screened
from 26 to 41 feet bls. The top 3 feet of screen (26 to 29 feet bls) was open to saprolite, and the
remaining 12 feet of screen (from 29 to 41 ft bls) was open to transition zone material. The pump was
set at a depth of 41 ft bls and was operated at a relatively constant flow rate of 20 gpm for 72 hours
from October 11 to 14, 2004. Water levels were measured throughout the test in ten well/piezometer
clusters12 (19 observation wells in all) and at two staff gages in Boyd Branch (fig. 38A). Clustered wells
screened in the upper regolith and lower regolith (BC-4S and I, P1S and I, P5S and I, P6S and I, P7S
and I, P8S and I, P2S and I, and BC-5S and I, fig. 38A) were used to evaluate the degree of
interconnectivity between different parts of the regolith flow system. Bedrock well BC-4D was used to
observe connectivity between the regolith and fractured bedrock portions of the groundwater system.
Wells positioned in a radial pattern or along perpendicular transects were used to observe anisotropy
and the shape and extent of the zone of pumping influence. Assuming that the weathered parent rock
retains distinct foliations and therefore preferred pathways, then both the saprolite and transition zone
could be anisotropic, both horizontally and vertically.
Three transects were used to analyze drawdown: A-A‟, oriented in the approximate direction of
presumed foliation strike: BC-4I, P6S and I, P8S and I, and Boyd Branch; B-B‟: BC-4I, BC-4S, P1S and
I, and P7S and I; and C-C‟: BC-4I, P5S and I, P2S and I, and P3S. Geologic cross sections of the
regolith were constructed for each transect using information obtained during coring and drilling (figs.
38A-D). Wells positioned very close to the pumping well (BC-4S, P1S and I, P6S and I, and P5S and I)
were used to observe delay in the time-drawdown curve and to analyze aquifer characteristics in the
vicinity of the pumping well. Distant wells (P8S and I, P7S and I, BC-5S and I, P2S and I, P3S, and
P4S and I) were used to observe the distance that the cone of depression expanded during the test.
Well locations and screen intervals are shown in table 11.
Determinations of depth to saprolite, transition zone, and bedrock were based on observed drill
cuttings, auger penetration rates, and interpretations of rock core at BC-4. Interpolations were made
between drilled locations. Some subjectivity arose when determining depths because the degree of
weathering often varied over relatively short distances.
Aquifer characteristics in the test area were as follows13:
The regolith was not of uniform thickness in the test area (figs. 38a to 38d). The saturated
thickness of the regolith gradually thickened in the predominant groundwater flow direction from
NW to SE, from about 8 feet at P7 to about 35 feet at BC-4I to about 55 feet at P2 (figs. 38A to
38D).
Potential horizontal hydraulic boundaries in the test area were: (1) Boyd Branch, located about
240 feet northeast of pumping well BC-4I (figs. 38A and 38B), (2) a suspected schist-
metagraywacke contact that is less than 400 feet west-northwest of BC-4I (figs. 38A and 38C),
and (3) the thinned regolith flow system that nearly pinches out upslope in the vicinity of P7S
(less than 250 feet west-northwest of BC-4I) (fig. 38C).
A contact between mica schist and metagraywacke is projected to occur somewhere between
the pumping well and P7 located 370 feet to the northwest, as mapped by Merschat and Carter
(2002, appendix 1A) and drill cuttings.
The location of the pumping test is near the base of a secondary synform fold as mapped by
Merschat and Carter (2002, appendix 1A).

12 Observation wells in the upper and lower regolith included 2-inch diameter piezometers with 5-ft screen lengths
and 4-inch diameter wells with 15-ft screen lengths. Observation wells in the bedrock were open hole with
diameters of 6.25 inches.
13 Conditions that deviated from uniform and homogeneous may result in a variety of effects that must be
considered during analysis. For example, distinct changes in rock type could, in some localized areas, result in the
upper regolith (saprolite) being only sluggishly connected to the lower regolith (transition zone). Partially
penetrating well effects can produce time drawdown data that resemble that produced by a recharge boundary, an
aquifer of variable thickness, or an aquifer experiencing downward leakage.

Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
62
Four outcrops within 1000 feet show vertical or near vertical foliation with a NE strike (average
azimuth of about 36 degrees). Farther up slope, moderately dipping foliation also strikes NE
(average azimuth of about 55 degrees). The foliation strike to the northeast appears to be a
preferred pathway for groundwater flow in the regolith (which, though weathered, retains the
fabric of its parent bedrock) and contributes to anisotropic conditions in the aquifer test area.
Vertical gradients were slightly downward, minimal, or absent at clustered wells in the aquifer
test area, suggestive of unconfined aquifer conditions. An exception to this was a very slight
upward vertical gradient from the lower regolith well to the upper regolith well in the vicinity of
P7.
Total drawdown in pumping well BC-4I was 5.2 feet, and total drawdown in the nearest
saprolite, transition zone, and deep bedrock well was 3.6 feet (BC-4S), 3.1 feet (P1I), and 1.2 feet (BC-
4D), respectively (table 12). Of all observation wells surrounding the pumping well, P6I and P6S to the
northeast responded first to pumping (fig. 39), even though four wells were in closer proximity to the
pumping well (table 11; fig. 38A). Four outcrops within 1000 feet of the test area exhibit a NE foliation
strike (perpendicular to Boyd Branch) with near vertical dip angles (appendix 1a). Drawdown in wells
parallel to foliation strike generally are larger and occur more quickly (for example, in P6S and P6I) than
in other nearby wells (fig. 39). The next wells to respond to pumping were the nearby transition zone
wells P5I and P1I, followed by the nearest to farthest shallow wells BC-4S, P5S, and BC-1S (fig. 39).
Drawdown at P1S and P1I and P5S and P5I (positioned nearly perpendicular to foliation strike, fig. 38A)
occurred more slowly and were less pronounced than at P6S and P6I (fig. 39; table 12) supporting the
concept that preferential flow occurs parallel to foliation strike.
Drawdowns in clustered wells were compared at various times during the pumping test to
evaluate how freely connected the upper regolith (saprolite) was with the lower regolith (transition
zone). At P1, drawdown was 0.2 to 0.4 feet greater in the transition zone well (P1I) than in the saprolite
well (P1S) at each observed time (table 12 and fig. 40). This suggests vertical stratification between the
two open intervals. This effect also was observed at P5, where drawdown was 0.2 to 0.5 feet greater in
the transition zone well P5I than in the saprolite well P5S for the first 10 hours of the test. Nearby
transition zone wells generally responded to pumping effects from BC-4I more quickly than did the
saprolite wells. The approximate lateral extent of influence due to pumping well BC-4I at 20 gpm over
the 72-hour test is shown in fig. 41.

Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
63

Staff gage
Staff gage
Figure 13a. Plan view of observation wells used in BBSub aquifer test
Pumping
well
Figure 31a. Plan view of selected observation wells used in aquifer pumping test,
Bent Creek Research Station, North Carolina.
Approximate direction
of foliation strike
Figure 38A. Plan view of selected observation wells measured during 72-hour aquifer pumping test, Bent Creek
Research Station, North Carolina.

Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
64

Figure 31b. Cross section A-A‟ showing selected wells screened in regolith flow system.

Figure 38B. Cross section A-A‟ showing selected wells screened in regolith, Bent Creek Research Station, North Carolina.
BC-4D
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
65

Figure 31c. Cross-section B‟-B showing selected wells screened in regolith flow system,

Figure 38C. Cross-section B‟-B showing selected wells screened in regolith, Bent Creek Research Station, North Carolina.

Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
66

Figure 31d. Cross section C-C‟ showing selected wells screened in regolith flow system.
Figure 38D. Cross section C-C‟ showing selected wells screened in regolith, Bent Creek Research Station, North Carolina.

Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
67

Table 11. Position, depth, and screen information for observation wells measured during aquifer pumping test conducted at lower
regolith well BC-4I, Bent Creek Research Station, North Carolina.
Observation well
General orientation
from pumping well
Distance from
pumping well,
in feet
Depth below land
surface to top of
casing, in feet
Screen length, in
feet
Well
diameter,
in inches
Open to upper or
lower portion of
regolith
BC-4I (pumping well)not applicable 0 26 15 4 lower
BC-4S NW 18 7 15 4 upper
BC-4D SE 18 open hole to 500 6.25 not applicable
P1S NW 54 12 5 2 upper
P1I NW 45 25 5 2 lower
P5S E 32 12 5 2 upper
P5I E 38 24 5 2 lower
P6S NE 45 17 5 2 upper
P6I NE 48 29 5 2 lower
P8S NE 222 15 5 2 upper
P8I NE 222 29 5 2 lower
P7S NW 369 8.5 5 2 upper
P7I NW 411 20 5 2 lower
P2S SE 220 17 5 2 upper
P2I SE 226 54 5 2 lower
P3S SE 247 17 5 2 upper
P4S SE 708 12 5 2 upper
P4I SE 714 25 5 2 lower
BC-5S NW ~1000 9 15 4 upper
BC-5I NW ~1000 32 15 4 lower
N-NW
N-NW
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
68

Table 12. Drawdown in relation to time in observation wells measured
during aquifer pumping test conducted at lower regolith well BC-4I, Bent
Creek Research Station, October 11 to 14, 2004.
Well 2 min 10 min 30 min 1 hr 10 hrs 72 hrs
BC-4I 1.8 2.6 3.1 3.4 4.2 5.2
BC-4S nc 0.4 1.1 1.5 2.5 3.6
BC-4D nc nc nc 0.1 0.8 1.2
P1S nc 0.2 0.6 0.9 1.6 2.8
P1I nc 0.4 0.8 1.1 2 3.1
P5S nc 0.3 0.8 1.2 2.1 3.4
P5I 0.1 0.7 1.2 1.5 2.1 2.9
P6S 0.3 1.2 1.7 2.1 2.9 4.0
P6I 0.4 1.2 1.8 2.1 2.9 3.9
P8S nc nc 0.1 0.2 0.5 1.0
P8I nc nc nc nc 0.3 0.7
P2S nc nc nc nc 0.1 0.3
P2I nc nc nc nc 0.2 0.4
P3S nc nc nc nc 0.1 0.2
P7S nc nc nc nc nc 0.1
P7I nc nc nc nc nc 0.2
P4S nc nc nc nc nc nc
P4I nc nc nc nc nc nc
BC-5S nc nc nc nc nc nc
BC-5I nc nc nc nc nc nc
BC-5D nc nc nc nc nc nc
Background
wells 2 min 10 min 30 min 1 hr 10 hrs 72 hrs
BC-3S nc nc nc nc nc nc
BC-3I nc nc nc nc nc nc
BC-3D nc nc nc nc nc nc
BC-7S nc nc nc nc nc nc
BC-7I nc nc nc nc nc nc
BC-7D nc nc nc nc nc nc
nc, no change
Drawdown, in feet
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
69

0 50 100 150 200 250 300 350 400
PZ-1S
PZ-1I
PZ-6I
PZ-6S
PZ-5I
PZ-5S
BC-4S
.
Figure. 39. Time to initial drawdown in observation wells in vicinity of pumping well BC-4I during
aquifer test of October 11 to 14, 2004, Bent Creek Research Station, North Carolina.
Time since start of pumping to initial water level drawdown, in minutes
0 0.8 1.7 2.5 3.3 4.2 5 5.8 6.7
Increasing
distance
from
pumping
well BC-4I
P
P
P
P
P
P
Pump on
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
70

0
1
2
3
4
5
0.0 0.0 0.2 0.5 1.0 10
.
0
72
.
0
72
.
2
72
.
5
73
.
0
82
.
0
12
9
.
7
Dr
a
w
d
o
w
n
,

i
n

f
t

.
BC-4I
BC-4S
BC-4D
P1S
P1I
P5S
P5I
P6S
P6I
1 hr 0 min 2 min 10 min 30 min 1 hr 10 hrs 72 hrs 10 min 30 min 10 hrs 58 hrs
Pump off,
recovery period
Figure 40. Drawdown and recovery in observation wells during 72-hour aquifer pumping test conducted at BC-4I, October 11 to 17, 2004, Bent
Creek Research Station, North Carolina.
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
71

Aquifer Response to 45-Hour Pumping in Bedrock

A second aquifer test was conducted in bedrock well BC-4D from June 20 to 22, 2006, to
evaluate the hydraulic connection between the deeper, fractured rock flow system and the shallow
regolith flow system. BC-4D, an open-hole bedrock well, was pumped at a relatively constant flow rate
of about 4 gpm for 45 hours (fig. 42), and water levels were measured in wells and piezometers in the
upper regolith, lower regolith, and bedrock in the vicinity. Total drawdown in BC-4D after just over 45
hours was 45.6 ft. The pump was placed at a depth slightly below two primary, water producing
fractures observed on OTV logs for BC-4D at 105 feet bls.

0
5
10
15
20
25
30
35
40
45
50
0 10 20 30 40 50
Dr
a
w
d
o
w
n
,

f
e
e
t
Elapsed time of pumping, hours
Drawdown in Pumping Well BC4-D

Figure 42. Drawdown in bedrock pumping well BC-4D during 45-hour aquifer test.

Only a minor, diffuse hydraulic connection was observed between the pumped portion of the
flow system (fractured bedrock) and the overlying regolith. Drawdowns in regolith wells generally were
Figure 41. Influence due to pumping well BC-4I at 20 gallons per minute for 72 hours, October 11
to 14, 2004, Bent Creek Research Station, North Carolina.
total drawdown between 0.5 and 0.1 ft
total drawdown between 1.0 and 0.5 ft
total drawdown between 5.2 and 1.0 ft
Aquifer pumping test drawdown after 72 hours:
N ~ 150 ft
total drawdown less than 0.1 ft (no influence)
total drawdown between 0.5 and 0.1 ft
total drawdown between 1.0 and 0.5 ft
total drawdown between 5.2 and 1.0 ft
Aquifer pumping test drawdown after 72 hours:
N ~ 150 ft
total drawdown less than 0.1 ft (no influence)
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
72
modest and sluggish. The pumped bedrock well BC-4D was not hydraulically connected to bedrock
wells located about 1000 feet upgradient (BC-5D) and 1500 feet downgradient (BC-3D). Observation
wells with the greatest total drawdown were P5I (0.55 ft), P2I (0.49 ft), P5S (0.43 ft), BC-4I (0.31 ft),
P2S (0.3 ft), P6I (0.3 ft), P6S (0.28 ft), BC-4S (0.24 ft), P1I (0.23 ft), P1S (0.21 ft), P3S (0.19 ft), P8I
(0.12 ft), and P8S (0.12 ft). All other wells had drawdowns of less than 0.1 feet over the 45 hours, and
these values were interpreted as a minor seasonal water level trend and not due to pumping at BC-4D.
The fractures that produce the 4 gpm yield in BC-4D (inferred to be at 105 ft) may be better
connected to regolith beyond the area of observation or may be hydraulically connected to Boyd Branch
located about 200 feet to the east. In other words, the well may draw water from a source not
measured during the aquifer test (no attempt was made to measure flow rates in Boyd Branch during
the aquifer test). The fractures at 105 feet contained 20 to 40 year old groundwater (estimated by age
dating analysis) during a sampling event using an inflatable packer sampling system. As results from
other REP investigations in the Piedmont-Mountains make clear, hydraulic connectivity between
bedrock and regolith and between neighboring bedrock wells varies from site to site, depending on a
host of local hydrogeologic factors (Pippin and others, 2008; Chapman and others, 2005; oral
communications, S. Wang, J. Pippin, and L. Skidmore, NCDENR, April 21, 2008).

Age

Groundwater samples were collected and analyzed for CFCs (and tritium-helium in selected
samples) in clustered regolith and bedrock wells at BC-1, BC-2, BC-3, BC-4, BC-5, and BC-7 at various
dates in 2004 and 2005. For quality control, halon had been mixed with compressed air during the
original air-rotary well drilling activities to serve as a tracer that would indicate whether or not modern air
containing CFC‟s was introduced to the groundwater during well installation. If halon was not present in
the groundwater then the sample generally was considered representative of the natural flow system
and an age was estimated. Sampling and analysis procedures are described at the U.S. Geological
Survey‟s Reston Chlorofluorocarbon Laboratory in Reston, Virginia (USGS website
http://water.usgs.gov/lab/cfc/). Both the mean (piston-flow) CFC ages and tritium/helium ages (used to
determine the youngest component of groundwater) were estimated where possible.
Groundwater collected from a well was comprised of waters that traveled along different flow
paths and, in some cases, from different source areas. This was especially true in the case of bedrock
wells, which often tap and draw water from numerous fractures that may be connected to different
source areas located at varying distances from the well. Because of the complex nature of the three-
dimensional flow system, wells inherently drew water that is a composite of various source waters.
Thus, a range of ages were expected in a given sample of groundwater drawn from bedrock wells.
The apparent CFC-derived age of groundwater in bedrock wells ranged from about 20 to 50
years (written communication, M. Chapman, October 2, 2008)(fig. 43 and table 13). One exception to
this was at well BC-7D in which 40-year old water was mixed with younger, modern water, suggesting
that at least some local recharge (modern water) occurred near the well head. The apparent CFC-
derived age of groundwater in upper regolith saprolite wells (screened across the water table at depths
of 5 to 30 feet bls) ranged from 1 to 26 years, and the tritium/helium-derived age ranged from 2 to 3
years. The apparent CFC-derived age of groundwater in lower regolith transition zone wells (screened
at depths of 8 to 53 feet bls) ranged from about 1 to 30 years, and the single tritium/helium-derived age
was 14 years. The youngest component of groundwater derived from tritium/helium analyses generally
was unavailable for bedrock wells.
Groundwater age is dependent on flow path length and properties and, in the case of bedrock
wells, the connectedness of the producing fracture to the overlying regolith storage. To illustrate,
groundwater in bedrock wells in downslope, discharge areas (BC-1D and BC-2D, with median age
about 42 years)(fig. 4) tended to be older than groundwater in bedrock wells in mid- or upslope,
recharge areas with expected shorter median flow paths (BC-4D and BC-7D, with median age about 26
years)(table 13). At each sampled well cluster location, groundwater was older in the bedrock well than
in its corresponding overlying regolith well.
Two-decade old groundwater was identified in relatively shallow discharge-area regolith wells
BC-1S, -1I, and -2I, reflecting a longer flow path component of water and potentially upward mixing of
older water from deeper bedrock fractures. The oldest estimated age of water in BC-1S, -1I, and -2I is
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
73
25, 25, and 19 years, respectively (table 13 and fig. 43). The youngest estimated component of water
using tritium analysis in BC-1S is 2 years, reflecting a mixture of very young local recharge, potentially
from surface-water exchange, near the well head. This finding corroborates the results of groundwater
temperature recordings (section titled „Seasonal and Storm-Induced Groundwater Temperature
Fluctuations) that suggest lateral stream mixing (young water) and upwelling (somewhat older water) at
this location. Tritium analyses were not available for BC-1I and -2I.
Nearly two-decade old groundwater also was identified in the relatively shallow, mid-slope
regolith well BC-7S (recharge area), again reflecting a longer flow path component of groundwater.
While the well had a relatively shallow sample interval (screened opening from 10 to 25 feet bls), the
sediments comprising the regolith flow system at this location were particularly tight (slug test K of 0.2
ft/day) which would help account for slow moving recharge. The youngest component age date from
tritium analysis is 1 year, reflecting a mixture of very young local recharge.

Figure 43. Mean age dates (chlorofluorocarbon piston-flow range) for stream and ground water at
selected wells in the Bent Creek Research Station, North Carolina. Written communication, M.
Chapman, October 2, 2008.
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
74

Table 13. Apparent age at time of sampling in 2005 of stream and ground water at selected locations, Bent Creek Research
Station, North Carolina. Written communication, M. Chapman, October 2, 2008.
Older component: Youngest component:
Sample locationa hydologic setting Flow zone
Apparent ageb
using CFCs
Apparent agec using
tritium-helium, +/- 1 yr
Estimated range of
age at time of
sampling, in yrs; older
(and younger)
Basin stream surface water 1988 to 2005 1 to 17
BC-1S low lying, discharge area upper regolith, alluvium & TZ 1980 to 1990 2003 15 to 25 (and 2)
BC-1I low lying, discharge area lower regolith, TZ 1976 to 2005 1 to 29
BC-1I low lying, discharge area lower regolith, TZ 1986 to 1989 16 to 19
BC-1D (128-145 ft bls)d low lying, discharge area bedrock 1962 to 1973 32 to 43
BC-1D (55-83 ft bls)d low lying, discharge area bedrock 1962 to 1984 21 to 43
BC-1D low lying, discharge area bedrock 1968 to 1979 26 to 37
BC-2I low lying, discharge area lower regolith, TZ 1986 to 1989 16 to 19
BC-2D low lying, discharge area bedrock 1954 to 1966 39 to 51
BC-4S midslope, recharge area upper regolith, saprolite 1988 to 2005 1 to 17
BC-4S midslope, recharge area upper regolith, saprolite 1988 to 2001 4 to 17
BC-4I midslope, recharge area lower regolith, TZ 1988 to 2005 1992 1 to 17 (and 13)
BC-4D (99-111 ft bls)d midslope, recharge area bedrock 1968-1985 1986 20 to 37 (and 19)
BC-4D midslope, recharge area bedrock 1970-1986 19 to 35
BC-7S midslope, recharge area upper regolith, saprolite 1986 to 1994 2004 11 to 19 (and 1)
BC-7D midslope, recharge area bedrock 1977 to 1983 22 to 28
BC-7D midslope, recharge area bedrock 1965 to 2005 1 to 40
a Age date samples collected in January 2005 and November 2005
b CFC-derived age estimated using piston flow model
c Helium/tritium-derived age of youngest ground water present in sample
d Sample obtained from specified interval using downhole inflatable packers
yr, year; ft bls, feet below land surface; TZ, transition zone
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
75
WATER QUALITY

Water quality samples were collected periodically between 2003 and 2007, and continuously (at
well cluster BC-2) between February and June 2005, at the BCRS to characterize groundwater
chemistry in a schist-gneiss rock type. The purpose of sampling was to (1) determine the groundwater
“type” or geochemistry associated with the schist-gneiss rock type and compare this to the chemistry of
regional groundwater; (2) evaluate changes in water quality with season; (3) compare the chemical
signature from three depths of the groundwater flow system (upper regolith (saprolite), lower regolith
(transition zone), and fractured bedrock); (4) evaluate potential differences in water quality in recharge
and discharge areas; (5) evaluate changes in water quality in response to short-term rainfall events; (6)
determine the presence or absence of naturally occurring and anthropogenic contaminants in a pristine
mountain watershed; and (7) compare groundwater chemistry to the chemistry of the bedrock through
which the groundwater flows.
Water quality samples were obtained from 4 streams, 6 well clusters (19 wells in all), and 2
water supply wells in a nearby sub-basin (Arbor-3 and Arbor-4) within the Bent Creek watershed (fig.
44). Samples were collected during three to seven sampling events that corresponded to a variety of
seasonal and hydrologic conditions, including extreme low water levels during drought conditions (for
example, June 12, 2006 and August 20, 2007) and somewhat higher water levels after the heavy
rainfall during September 2004 (for example, November 1, 2004). Analyses included major ions,
metals, nutrients, volatile and semi-volatile organics, physical and biological properties, and selected
naturally occurring radionuclides. In addition, continuous hourly measurements of pH, DO,
temperature, and SC were made during the period between August 2004 and December 2005 at wells
BC-2S, -2I, and -2D, and at the stream gage in Boyd Branch, adjacent to well cluster BC-2 (B. Huffman,
written communication, September 2, 2008).
Unless otherwise noted, evaluations and discussion are based on a selected representative
value (typically the median value) at a given well or stream site as determined from all values measured
at that location over multiple sample dates. Previously published (Huffman and others, 2006) Piper
trilinear diagrams (Piper, 1953) displaying dominant cations and anions developed from well and stream
water chemistry data are presented for comparative purposes. Stiff diagrams (Stiff, 1951) are
presented for the current dataset14. The data also were compared to published regional datasets,
including the study by Trapp (1970) with 16 bedrock wells in Buncombe County, the data analyzed by
the North Carolina Public Health Lab (NCDHHS, 2008, webpage
http://slph.ncpublichealth.com/EnvironmentalSciences/inorganic/, accessed December 22, 2008) with
56 to 532 (depending on the constituent ) bedrock wells in Buncombe County, and the regional analysis
by Briel (1997) with 322 to 558 (depending on the constituent) bedrock wells across the Blue Ridge
Province.
The quality of water was related in large part to the concentration and types of dissolved ions
that it contained and was thus dependent upon the host rock through which it flowed. Water entering
the groundwater system as recharge contains dissolved gasses from the atmosphere, soil, and organic
matter (oxygen and carbon dioxide, for example) and, because this recharge water is typically acidic it
acts as a mild solvent. As the acidic water moves through the groundwater system, it dissolves
minerals such as the cations calcium, magnesium, sodium, and potassium in the soil and rock and, in
the process, carbon dioxide is converted to the bicarbonate anion.
Older, deeper groundwater that traveled greater distances was usually more mineralized than
younger, shallower groundwater (or surface water) that traveled shorter distances. Dissolved solids in
groundwater with a relatively short residence time were, in some cases, too low for the overall
groundwater composition to reflect effects of the dissolution of the rock formation. The groundwater
composition also varied somewhat due to the heterogeneity of the rock formations. Water quality data
analyzed in this study were consistent with this conceptual model.
Ground and surface water in the Bent Creek study area are of high quality. Concentrations of
dissolved inorganic constituents typically are very low, as are SC, total organic carbon (TOC), fecal
bacteria, nutrients, pesticides, herbicides, volatile and semi-volatile organic compounds, and naturally

14 To generate the Stiff diagrams, values reported as below method detection limits were assigned a value of one
half the detection limit.
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
76
occurring uranium, radium-226, and arsenic. These findings are consistent with regional findings in
many pristine settings (Briel, 1997; Trapp, 1970).

Groundwater Type and Comparison to Regional Water Quality

Based on major anion (bicarbonate, chloride, and sulfate) and cation (calcium, sodium,
magnesium, and potassium) concentrations observed in sample water from seven bedrock wells,
groundwater in the bedrock flow system in the study area is considered a calcium bicarbonate type. A
Stiff diagram (fig. 45) shows the major ion milliequivalent results from a June 2006 sampling event, as
analyzed by the NCDWQ laboratory. A Piper diagram (fig. 46) shows major ion results from an April
2003 sampling event, as analyzed by the USGS National Water Quality laboratory in Denver and
published by Huffman and others (2005). These findings are consistent with those published by Trapp
(1970) in which a Type II groundwater (calcium-bicarbonate) was identified for wells in similar rocks in
Buncombe County. Trapp noted that the Type II water was associated with garnet-muscovite schist
and biotite gneiss, both of which are predominant rock types in the study area. Tables 14 and 15 show
data used to evaluate water quality in the study area. Table 16 shows whole-rock chemistry oxide
analyses for rock samples obtained from core holes BC-1, BC-3, and BC-7.
Figure 44. Location of ground and surface water quality sample locations at Bent Creek
Research Station, North Carolina.

N
N
Surface water sample location
Ground water sample location
Arbor-4
Bent
C
r
e
e
k
Bent
C
r
e
e
k
Lake
Powhatan
Arbor-3
BC-4S, I, D
BC-5S, I, D
BC-6 D
BC-7S, I, D
BC-3S, I, D
BC-2S, I, D
BC-1S, I, D
Bent Creek
BB at
BC2
BB trib
at BC3
BB trib
at BC7
N
N
Surface water sample location
Ground water sample location
Arbor-4
Bent
C
r
e
e
k
Bent
C
r
e
e
k
Lake
Powhatan
Arbor-3
BC-4S, I, D
BC-5S, I, D
BC-6 D
BC-7S, I, D
BC-3S, I, D
BC-2S, I, D
BC-1S, I, D
Bent Creek
BB at
BC2
BB trib
at BC3
BB trib
at BC7
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
77
Bicarbonate was the principal anion in the study area bedrock well water (fig. 45 and 46), with
concentrations ranging from 15 to 57 mg/L (median = 44 mg/L) (tables 14 and 15). Dissolved
bicarbonate is produced by the conversion of carbon dioxide during the dissolution of cations during the
rock weathering process. Sulfate concentrations were the second most dominant anion, with
concentrations ranging from less than 4 to 26 mg/L (median = 10 mg/L). Chloride concentrations
ranged from less than 1 to 3.7 mg/L (median = less than 1 mg/L), with most samples near or below the
detection limit of 1 mg/L. By comparison, Trapp (1970) observed median concentrations of
bicarbonate, sulfate, and chloride of 54, 11, and 2.2 mg/L, respectively, in 16 bedrock wells in
Buncombe County (table 14).
Calcium was the principal cation in study area bedrock well water (figs. 45 and 46), with
concentrations ranging from 3.1 to 23 mg/L (median = 18 mg/L) (table 14 and 15). The source of
dissolved calcium is dissolution of rock that contains calcium-bearing minerals such as plagioclase
feldspar, amphibole, and garnet. For example, migmatite and biotite gneiss/metagraywacke rocks are
very common within the study area and both contain significant amounts of plagioclase feldspar (32
percent and 32 to 42 percent, respectively (Merschat and Carter, 2002)). Amphibolite, though less
common, did appear within some intervals of rock core, and contained plagioclase feldspar at amounts
from trace to 38 percent (Merschat and Carter, 2002). Garnet is a very common accessory mineral in
the study area in both the schist and gneiss.
Sodium generally was the second most dominant cation, with concentrations ranging from 2.8
to 20 mg/L (median = 3.7 mg/L) (table 14). Possible sodium sources include potassium feldspar (one
rock sample showed 28 percent) and plagioclase feldspars. Magnesium concentrations ranged from
0.4 to 3.3 mg/L (median = 1.3 mg/L), and potassium concentrations ranged from 0.8 to 4.3 mg/L
(median = 2 mg/L). Generally, the migmatite and biotite gneiss/metagraywacke rocks in the study area
contain significantly higher percentages of calcium oxide and sodium oxide than schist rocks (Merschat
and Carter, 2002; table 16), and the schist rocks contain significantly higher iron oxide content
(Merschat and Carter, 2002) and the iron sulfide mineral pyrrhotite (A. Merschat, written
communication, August 1, 2003) than the migmatite and metagraywacke. Pyrrhotite and other iron and
sulfur rich minerals weather to produce dissolved iron and sulfate. The iron can then oxidize and
precipitate out of solution to produce orange stains on rocks and fracture openings. The solution from
pyrrhotite is acidic, which allows for further dissolution of calcium and other ions from the host rock. So,
depending on the geochemical conditions of the groundwater and the lithology of the host rock through
which groundwater passes, higher calcium, sodium, and sulfate concentrations are possible, and some
mixture of the three is the result.
To determine how groundwater quality in the study area compares with regional water quality,
summary well data were compiled for study wells, Buncombe County (16 well samples from Trapp,
1970; 56 to 532 well samples, depending on constituent, from NCDHHS, 2008, webpage
http://slph.ncpublichealth.com/EnvironmentalSciences/inorganic/ accessed December 22, 2008), and
the Blue Ridge Province (322 to 558 well samples, depending on constituent, from Briel, 1997) (table
14). Data summaries also were compiled for three adjacent counties (Henderson, Madison, and
Transylvania) (Trapp, 1970) (table 14).
Levels of constituents in study area well water generally were comparable to levels reported for
Buncombe County and Blue Ridge Province wells (table 14). An exception to this was elevated pH,
and, to a lesser extent, calcium in 4 of 7 study wells (table 15). These highly buffered wells - BC-2D, -
3D, -4D, and -7D - had a pH range of 8.6 to 7.9 (median = 8.1) and a calcium range of 23 to 18 mg/L
(median = 21 mg/L) (table 15). In contrast, the 68th and 95th percentiles for pH in the Buncombe County
NCDHHS dataset (n = 524) were 7.6 and 8.2 respectively, and for the Buncombe County Trapp dataset
(n = 16) were 7.6 and 8.6, respectively. For calcium, the 68th and 95th percentiles in the NCDHHS
dataset (n = 531) were 27 mg/L and 39 mg/L, respectively, and the 68th and 95th percentiles in the
Trapp dataset (n = 16) were 22 mg/L and 32 mg/L, respectively. Comparing the datasets of bedrock
well chemistry for the study area, Buncombe County (NCDHHS, 2008), Buncombe County (Trapp,
1970), and Blue Ridge Province (Briel, 1997), median values of pH were 7.9, 7, 6.8, and 6.6,
respectively; SC were 122, not determined, 129, and 103 uS/cm; bicarbonate were 48, not determined,
54, and 32 mg/L; chloride were <1, <5, 2.2, and 2.1 mg/L; sulfate were 10, 7, 5.4, and 5 mg/L; calcium
were 18, 11.2, 11, and 8.3 mg/L; magnesium were 1.3, 2.5, 3, and 2.5 mg/L; potassium were 2, not
determined, 2.2, and 1.2 mg/L; sodium were 3.7, 6, 5.3, and 5 mg/L; and silica were 17, not determined,
23, and 16 mg/L (table 14).
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
78
Only three study area well locations, BC-1D (0.84 mg/L), BC-3D (0.39 mg/L), and BC-5D (0.69
mg/L), had median dissolved iron concentrations above the detection limit of 0.05 mg/L (table 15).
Dissolved manganese concentrations were low in all locations and ranged from less than the detection
limit of 0.01 mg/L to 0.058 mg/L. Dissolved lead and cyanide were below their respective detection
limits of 0.01 and 0.02 mg/L in all sampled locations. Similarly, all other measured dissolved minor
metals were at relatively low concentrations, typically at or near their respective detection limits (table
15).

Changes in Groundwater Quality with Season

In general, water quality at a given well location did not vary significantly across different
sampling dates and seasons. A typical example of this is provided in fig. 47, which shows relatively
consistent concentrations of selected water quality parameters in bedrock well BC-4D measured during
several sampling events. These findings suggest that water quality in this setting was not significantly
dependent upon seasonal changes. Typically, this was also the case at other well locations. Where
they exist, differences in water quality between events may be due to short-term rainfall events (flushing
influences), seasonal hydrologic variation, laboratory precision, and (or) slight changes in pump depth
during sampling.

Changes in Groundwater Quality with Depth and Topographic Setting

To compare water chemistry at different depths of the groundwater flow system, water samples
were collected from six well clusters drilled in the upper regolith (saprolite, water table wells), lower
regolith (transition zone), and open-hole bedrock; four streams also were sampled (fig. 44). In general,
groundwater in the upper and lower regolith was slightly acidic due to humic acids in soil and to
chemical and mechanical weathering of iron-sulfide bearing minerals in parent rock. The shallow
groundwater thus acted as a weak acid that tended to dissolve metals in the soil and rock as it moved
into and through the deeper, fractured bedrock flow system. The result was an increase in dissolved
ions, SC, and pH, and a decrease in DO in bedrock wells. This is typical of other groundwater research
stations in the Piedmont-Mountains (Pippin and others, 2008, Chapman and others, 2005) and
suggests that most of the DO contained in rain water that seeps into the soil is consumed in the
microbially mediated oxidation of organic matter as the water moves downward through the regolith.
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
79

Figure 45. Stiff diagrams showing major ion milliequivalents in water samples collected from (A) streams (0 to 0.2 scale), (B) upper
regolith wells (0 to 0.1 scale), (C) lower regolith wells (0 to 0.1 scale), and (D) open-hole bedrock wells (0 to 2 scale), June 2006,
Bent Creek Research Station, North Carolina.

A. Streams B. Upper regolith C. Lower regolith D. Open hole bedrock
Bent Creek
HCO3
HCO3
HCO3
HCO3
HCO3
HCO3
HCO3
Cl
Cl
Cl
Cl
Cl
Cl
Cl
SO4
SO4
SO4
SO4
SO4
SO4
SO4
HCO3
HCO3
HCO3
HCO3
HCO3
HCO3
Cl
Cl
Cl
Cl
Cl
Cl
SO4
SO4
SO4
SO4
SO4
SO4
HCO3
HCO3
HCO3
HCO3
HCO3
Cl
Cl
Cl
Cl
Cl
SO4
SO4
SO4
SO4
SO4
HCO3
HCO3
HCO3
HCO3
Cl
Cl
Cl
Cl
SO4
SO4
SO4
SO4
Ca
Na
Mg
BC-1D
Ca
Na
Mg
BC-2D
Ca
Na
Mg
BC-3D
Ca
Na
Mg
BC-4D
Ca
Na
Mg
BC-5D
Ca
Na
Mg
BC-6D
Ca
Na
Mg
BC-7D
Ca
Na
Mg
BC-1I
Na
BC-2I
Ca
Na
Mg
BC-3I
Ca
Na
Mg
BC-4I
Ca
Na
Mg
BC-5I
Ca
Na
Mg
BC-7I
Ca
Na
Mg
BC-1S
Ca
Na
Mg
BC-3S
Ca
Na
Mg
BC-4S
Ca
Na
Mg
BC-5S
Ca
Na
Mg
BC-7S
Ca
Na
Mg
Ca
Na
Mg
Ca
Na
Mg
Ca
Na
Mg
Ca
Mg
Milliequivalents per liter Milliequivalents per liter Milliequivalents per liter Milliequivalents per liter 220.10.10.10.10.20.2
Boyd
Branch
at BC2
trib to
Boyd
Branch
at BC3
trib to
Boyd
Branch
at BC7
Bent Creek
HCO3
HCO3
HCO3
HCO3
HCO3
HCO3
HCO3
Cl
Cl
Cl
Cl
Cl
Cl
Cl
SO4
SO4
SO4
SO4
SO4
SO4
SO4
HCO3
HCO3
HCO3
HCO3
HCO3
HCO3
Cl
Cl
Cl
Cl
Cl
Cl
SO4
SO4
SO4
SO4
SO4
SO4
HCO3
HCO3
HCO3
HCO3
HCO3
Cl
Cl
Cl
Cl
Cl
SO4
SO4
SO4
SO4
SO4
HCO3
HCO3
HCO3
HCO3
Cl
Cl
Cl
Cl
SO4
SO4
SO4
SO4
Ca
Na
Mg
BC-1D
Ca
Na
Mg
BC-2D
Ca
Na
Mg
BC-3D
Ca
Na
Mg
BC-4D
Ca
Na
Mg
BC-5D
Ca
Na
Mg
BC-6D
Ca
Na
Mg
BC-7D
Ca
Na
Mg
BC-1I
Na
BC-2I
Ca
Na
Mg
BC-3I
Ca
Na
Mg
BC-4I
Ca
Na
Mg
BC-5I
Ca
Na
Mg
BC-7I
Ca
Na
Mg
BC-1S
Ca
Na
Mg
BC-3S
Ca
Na
Mg
BC-4S
Ca
Na
Mg
BC-5S
Ca
Na
Mg
BC-7S
Ca
Na
Mg
Ca
Na
Mg
Ca
Na
Mg
Ca
Na
Mg
Ca
Mg
Milliequivalents per liter Milliequivalents per liter Milliequivalents per liter Milliequivalents per liter 220.10.10.10.10.20.2
Boyd
Branch
at BC2
trib to
Boyd
Branch
at BC3
trib to
Boyd
Branch
at BC7
Discharge
areas
Recharge
Areas
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
80

Figure 46. Trilinear Piper diagrams showing water chemistry data from April 2003, at (A) regolith wells
and stream sites, and (B) open hole bedrock wells, Bent Creek Research Station, North Carolina, from
Huffman and others (2006).
A
B
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
81

Table 14. Statistical summaries of water quality results for regional studies and sample locations at the Bent Creek Research Station, North Carolina.
BCRS sample locations and regional datasets pH SC DO TDS HCO3 Cl SO4 Ca Mg K Na Si
uS/cm mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L
Saprolite wellsa
minimum 5.1 14 3.4 15 4 <1 <2 0.6 0.5 0.6 0.8 6.9
maximum 5.5 24 8.2 24 7 1 <2 1 0.7 1.2 1.9 11.5
average 5.4 18 6.1 19 5 <1 <2 0.8 0.6 0.8 1.3 8.9
median 5.4 17 6 19 5 <1 <2 0.7 0.7 0.8 1.2 8.5
Transition zone wells
minimum 5.3 12 5.4 14 3 <1 <2 0.5 0.5 0.3 0.7 7
maximum 5.7 19 9.0 26 7 1.1 <2 1.1 1.5 0.8 2.0 12.0
average 5.5 16 6.9 21 5 <1 <2 0.8 0.7 0.5 1.2 8.8
median 5.5 16 7.0 22 6 <1 <2 0.8 0.6 0.6 1.2 8.9
Bedrock wells
minimum 6.2 46 0.1 38 15 <1 <4 3.1 0.4 0.8 2.8 10
maximum 8.6 214 7 140 57 3.7 26 23 3.3 4.3 20 23
average 7.5 114 1.6 81 37 1.2 10.6 13.9 1.5 2.2 6.7 17.4
median 7.9 122 0.2 82 48 <1 10 18 1.3 2 3.7 17
Surface water
minimum 6.6 15 7.8 23 5 <1 <2 0.9 0.6 0.6 1.1 8
maximum 6.9 22 9.4 28 7 2.5 <2 1.6 0.9 0.8 1.8 13
average 6.8 18 8.6 26 6 1 <2 1.2 0.7 0.7 1.5 10.6
median 6.8 18 8.7 26 6 <1 <2 1.2 0.7 0.7 1.5 10.7
Regional data
NCDHHS, 2008b (n = 56 to 532)7 <5 7 11.2 2.5 6
Trapp, 1970c (4 counties, n = 62)6.8 129 54 2.2 5.4 11 3 2.2 5.3 23
Buncombe Co, n = 16 6.8 129 54 2.2 5.4 11 3 2.2 5.3 23
Henderson Co, n = 22 6.3 45 26 1.2 0.2 2.8 1.2 0.9 3.7 19
Transylvania Co, n = 12 6.4 18 21 1.1 0.4 3 0.6 0.9 4 18
Madison Co, n = 12 6.3 125 53 2 3 10 4.9 2.3 4.9 32
Briel, 1997d (n = 322 to 558)6.6 103 6.2 73 32 2.1 5 8.3 2.5 1.2 5 16
a Saprolite well BC-2S was not included in statistical analyses due to potential grout contamination; b Data obtained from bedrock drinking wells in Buncombe Co, with n = 56 to 532,
depending on constituent (n = 14 for DO); c Data obtained from drilled bedrock drinking wells in Buncombe, Henderson, Transylvania, and Madison Counties, with n = 62; d Data
obtained from bedrock drinking wells in the Blue Ridge Province, with n = 322 to 558, depending on constituent. Abbreviations: SC, specific conductance; DO, dissolved oxygen;
Alk, alkalinity; TDS, total dissolved solids; HCO3, bicarbonate; Cl, chloride; SO4, sulfate; Ca, calcium; mg, magnesium; K, potassium; Na, sodium; Si, silica; mg/L, milligrams per
liter; BCRS, Bent Creek Research Station.
6.9
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
82

Table 15. Median water quality results for locations within the Bent Creek Research Station, North Carolina, 2003 to 2007, and data
compiled from regional studies.
BCRS sample locations and regional datasets pH SC DO TDS HCO3 Cl SO4 Ca Mg K Na Si
uS/cm mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L
BC-1S 5.3 16 5.4 22 6 <1 <2 0.8 0.5 0.3 1.2 10
BC-1i 5.5 14 7 20 5 <1 <2 0.8 0.5 0.3 1.1 8.9
BC-1D 6.8 71 0.2 64 21 1.8 8.1 7.5 1.3 0.8 5.6 23
BC-2Sa 6 200 380 78 3.7 18 25 8.3 1.8 2.1 6
BC-2i 5.5 19 6 24 6 1 <2 0.7 1.5 0.4 1.2 7
BC-2D 8.6 214 0.1 140 48 <7.4 26 18 0.4 1.1 20 22
BC-3S 5.1 15 6.3 24 5 <1 <2 0.8 0.7 0.6 1.4 10
BC-3i 5.7 17 7.8 24 6 <1 <2 1.1 0.6 0.8 1.3 9.2
BC-3D 8.3 136 0.2 100 57 0.8 11 23 1.7 2.5 3.2 17
BC-4S 5.5 14 8.2 15 4 <1 <2 0.6 0.5 0.7 0.8 7
BC-4i 5.6 13 9 16 4 <1 <2 0.7 0.5 0.8 0.9 7.6
BC-4D 7.9 122 0.1 82 46 <1 10 18 1.1 3.7 3.7 14
BC-5S 5.4 24 3.4 22 7 0.8 <2 1 0.7 0.8 1.9 11.5
BC-5i 5.5 18 5.8 26 7 <1 <2 0.8 0.6 0.6 2 12
BC-5D 6.2 50 3.5 44 22 <1 2.6 4.8 1.4 2 3 16
BC-6D 6.7 46 6.9 38 15 <1 <3.5 3.1 1 1.1 2.8 20
BC-7S 5.4 18 6.4 16 4 1 <2 0.7 0.7 1.2 1 6.9
BC-7i 5.6 12 7.5 14 3 1.1 <2 0.5 0.7 0.6 0.7 7
BC-7D 7.9 160 0.3 100 56 <1 15 23 3.3 4.3 8.7 10
Boyd Branch at BC2 6.8 18 8.6 25 7 <1 <2 1.2 0.7 0.8 1.6 12
trib to Boyd Branch at BC3 6.8 17 9.4 26 <5 <2 1.1 0.6 0.7 1.1 9.4
trib to Boyd Branch at BC7 6.9 22 7.8 28 6 <1 <2 1.6 0.9 0.7 1.8 13
Bent Creek 6.6 15 8.7 23 5 <1 <2 0.9 0.6 0.6 1.4 8
Arbor-3 (~ 2 mi from BCRS)6.7 73 70 18 <1 12 0.6 2 2.1 4.7
Arbor-4 (~ 2 mi from BCRS)6.7 89 82 32 1.8 9.1 1.1 2.2 1.1 6.5
Regional data
NCDHHS, 2008b (n = 56 to 532)7 47 <5 7 11.2 2.5 6
Trapp, 1970c (4 counties, n = 62)6.8 129 54 2.2 5.4 11 3 2.2 5.3 23
Briel, 1997d (Blue Ridge Province, n = 322 to 558)6.6 103 6.2 73 32 2.1 5 8.3 2.5 1.2 5 16
a Saprolite well BC-2S was not included in statistical analyses due to potential grout contamination; b Data obtained from bedrock wells in Buncombe Co (n = 14 to 532,
depending on constituent); c Data obtained from drilled bedrock drinking wells in Buncombe, Henderson, Transylvania, and Madison Counties (n = 62); d Data obtained
from bedrock drinking wells in the Blue Ridge Province (n = 322 to 558, depending on constituent). Abbreviations: SC, specific conductance; DO, dissolved oxygen; Alk,
alkalinity; TDS, total dissolved solids; HCO3, bicarbonate; Cl, chloride; SO4, sulfate; Ca, calcium; mg, magnesium; K, potassium; Na, sodium; Si, silica; mg/L,
milligrams per liter; BB, Boyd Branch; BB trib, un-named tributary to Boyd Branch; BCRS, Bent Creek Research Station.
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
83

Table 16. Whole-rock chemistry oxide analyses for rock samples obtained from core holes BC-1, BC-3, and
BC-7, Bent Creek Research Station, North Carolina

Sample
depth*SiO2 Al2O3 Fe2O3 MgO CaO Na2O K2O TiO2 P2O5 MnO
ft bls %%%%%%%%%%
Core Hole at BC-1 31 54.1 22.8 9.5 2.7 0.5 0.6 4.0 1.1 0.2 0.2
Core Hole at BC-1 60 47.8 24.9 10.4 3.4 0.9 0.9 5.1 1.3 0.2 0.3
Core Hole at BC-1 124 52.7 22.1 10.8 2.9 1.7 1.8 3.1 1.0 0.3 0.2
Core Hole at BC-1 127 49.5 23.7 11.0 3.2 1.0 0.8 4.3 1.4 0.2 0.2
Core Hole at BC-3 60 67.8 14.0 6.6 2.1 1.4 2.6 2.6 1.0 0.2 0.1
Core Hole at BC-3 109 70.5 12.6 5.7 2.0 1.5 1.8 3.1 1.1 0.2 0.1
Core Hole at BC-3 125 48.7 13.4 6.1 5.9 11.6 2.0 2.8 0.4 0.5 0.2
Core Hole at BC-7 47 48.5 25.5 11.5 3.3 0.6 1.4 3.6 1.2 0.2 0.3
Core Hole at BC-7 104 43.3 27.0 13.1 4.0 1.6 1.5 3.8 1.5 0.3 0.2
Core Hole at BC-7 143 62.5 15.3 7.3 2.5 1.6 2.2 4.6 1.2 0.3 0.1
Rock samples from Bent Creek watershed (Merschat and Carter, 2002)
Chl-mus schist 47.2 24.9 11.9 0.5 0.1 0.6 4.5 1.6 0.1 0.0
Chl-gar-mus schist 36.5 27.6 17.7 5.1 1.0 0.7 4.4 1.6 0.1 0.5
Chl-gar-mus schist 47.8 26.1 10.7 3.1 0.5 1.0 4.1 1.2 0.3 0.2
Gar-mica schist 45.4 25.8 12.4 2.5 0.1 0.7 5.1 1.1 0.1 0.2
Metagraywacke/bg 73.9 11.8 4.5 1.3 3.4 2.8 1.6 0.7 0.2 0.2
Metagraywacke/bg 66.6 13.9 6.0 1.8 1.5 2.8 3.6 1.0 0.1 0.1
Mig with metagraywacke/bg 67.0 17.2 3.5 1.1 2.6 5.4 2.6 0.6 0.0 0.0
Ft bls, feet below land surface; chl, chlorite; mus, muscovite; gar, garnet; mig, migmatite; bg, biotite gneiss
Rock samples from BCRS

0
20
40
60
80
100
120
140
pH SC,
uS/cm
TDS,
mg/L
HCO3,
mg/L
SO4,
mg/L
Ca,
mg/L
Mg,
mg/L
K, mg/L Na,
mg/L
Silica,
mg/L
22-Apr-03
27-Apr-04
2-Nov-04
27-Jun-05
13-Jun-06
21-Aug-07
Open hole bedrock well BC-4D
Figure. 47. Water quality results for selected parameters measured in open-hole bedrock well BC-4D
at various sample dates from 2003 to 2007, Bent Creek Research Station, North Carolina.

Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
84
Streams and upper regolith wells generally are characterized as a sodium/magnesium-
bicarbonate type (figs. 45 and 46) and contain low SC (median = 18 and 17 uS/cm, respectively), low
dissolved solids (median = 26 and 19 mg/L, respectively), and relatively high DO (median= 8.7 and 6
mg/L, respectively) (tables 14 and15 and fig. 48). Streams had a near neutral pH (median = 6.8), and
the upper regolith and lower regolith wells were slightly acidic (median = 5.4 and 5.5, respectively).
Trends were not noted in groundwater chemistry from recharge to discharge areas in the upper regolith
flow system.
Water from six lower regolith wells was similar to that of upper regolith wells, suggesting
generally well-mixed or very similar geologic conditions within the regolith flow system. The lower
regolith wells, like those of the upper regolith, generally are characterized as a sodium/magnesium-
bicarbonate type (figs. 45 and 46), and are associated with low SC (median = 16 uS/cm), low dissolved
solids (median = 22 mg/L), and relatively high DO (median = 7 mg/L) (fig. 47) (tables 14 and 15 and fig.
48). Trends were not noted in groundwater chemistry from recharge to discharge areas in the lower
regolith (transition zone) flow system.
Water from seven open-hole bedrock wells generally is characterized as a calcium-bicarbonate
type , with the exception of BC-2D, which is a calcium/sodium-bicarbonate type (figs. 45 and 46).
Groundwater residence times are longer for bedrock wells than for regolith wells, and, in general, the
longer the water is in contact with soil and rock, the greater the mineral dissolution. As a result,
bedrock wells contained significantly higher dissolved ions than regolith wells. The pH and SC in
bedrock wells are significantly higher (median = 7.9 and 122 uS/cm, respectively) and the DO
significantly lower (median = 0.2 mg/L), than the values observed in the regolith wells (tables 14 and 15
and fig. 48). Major ions, including calcium, magnesium, sodium, bicarbonate, chloride, and sulfate, are
significantly higher in bedrock wells than in regolith wells (figs. 49 and 50). Total dissolved solids
generally were about 1.3 times the SC in regolith wells and about 0.7 times the SC in bedrock wells.
Trends were not noted in groundwater chemistry from recharge to discharge areas in the fractured
bedrock flow system.
Relatively high levels of pH, calcium, bicarbonate, and sulfate and low DO were observed in 4
of the 7 bedrock wells located in both discharge (BC-2D) and recharge (BC-3D, -4D, and -7D) areas
(table 15). Likely, these results are due to a combination of factors including (1) the lithologic
composition and degree of weathering of host rock through which the sampled groundwater flowed, (2)
the length of time the water spent in contact with the host rock, and (3) potential grout contamination
during drilling and well completion activities. Calcium-rich feldspars such as plagioclase are found in
varying concentrations throughout the study area rocks, particularly the metagraywackes (plagioclase
content from 32 to 42 percent) and the less commonly occurring amphibolites (plagioclase content from
trace to 38 percent) (Merschat and Carter, 2002, appendix 1a). Moreover, calcium-rich amphibole
outcroppings are present in the general vicinity or upgradient of these four areas (Merschat and Carter,
2002). Another source of calcium common in the study area is the silicate accessory mineral garnet.
Rock samples analyzed for calcium oxide (CaO) varied from 1.4 to 11.6 percent at 3 depths in core hole
BC-3, and 0.6 to 1.6 percent at the 3 depths in core hole BC-7 (table 16). The finding of 11.6 percent
calcium oxide at one depth in corehole BC-3 may be associated with an interlayered sequence of
calcium-bearing amphibole, which occurs locally in some parts of the Bent Creek study area, or a
dominant layer of migmatite.
Silica concentrations ranged from 6.9 to 23 mg/L and generally were higher in bedrock wells
(associated with longer residence times) than in regolith wells (fig. 51). A principal source of dissolved
silica is the weathering and decomposition of mineral silicates present in the parent rock, leaving behind
clay minerals in the weathered portions of rock.

Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
85

Figure 48. Range and median values for (A) pH, (B) specific conductance, and (C) dissolved oxygen
in wells and streams during periodic sampling events between 2003 and 2007, Bent Creek Research
Station, North Carolina.

n upper regolith = 5
n lower regolith = 6
n bedrock = 7
n streams = 4
(A) pH
(B) Specific
Conductance
(C) Dissolved
Oxygen
0
50
100
150
200
250
upper regolith lower regolith bedrock streams
Sp
e
c
i
f
i
c

C
o
n
d
u
c
t
a
n
c
e
,

u
S
/
c
m
0
1
2
3
4
5
6
7
8
9
10
upper regolith lower regolith bedrock streams
Di
s
s
o
l
v
e
d

O
x
y
g
e
n
,

m
g
/
L
5
5.5
6
6.5
7
7.5
8
8.5
9
upper regolith lower regolith bedrock streams
pH
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
86

.

Figure 49. Range and median values for (A) calcium, (B) magnesium, and (C) sodium in wells and
streams during periodic sampling events between 2003 and 2007, Bent Creek Research Station,
North Carolina.
n upper regolith = 5
n lower regolith = 6
n bedrock = 7
n streams = 4
(A) Calcium, mg/L
(B) Magnesium, mg/L
(C) Sodium, mg/L
0
5
10
15
20
25
upper regolith lower regolith bedrock streams
Ca
l
c
i
u
m
,

m
g
/
L
0
0.5
1
1.5
2
2.5
3
3.5
upper regolith lower regolith bedrock streams
Ma
g
n
e
s
i
u
m
,

m
g
/
L
0
5
10
15
20
25
upper regolith lower regolith bedrock streams
So
d
i
u
m
,

m
g
/
L
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
87

n upper regolith = 5
n lower regolith = 6
n bedrock = 7
n streams = 4
(A) Bicarbonate, mg/L
(B) Chloride, mg/L
(C) Sulfate, mg/L
Figure 50. Range and median values for (A) bicarbonate, (B) chloride, and (C) sulfate in wells and
streams during periodic sampling events between 2003 and 2007, Bent Creek Research Station,
North Carolina.
0
10
20
30
40
50
60
upper regolith lower regolith bedrock streams
Bi
c
a
r
b
o
n
a
t
e
,

m
g
/
L
0
0.5
1
1.5
2
2.5
3
3.5
4
upper regolith lower regolith bedrock streams
Ch
l
o
r
i
d
e
,

m
g
/
L
0
5
10
15
20
25
30
upper regolith lower regolith bedrock streams
Su
l
f
a
t
e
,

m
g
/
L
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
88

Changes in Groundwater Quality with Short-Term Rainfall Events

Storm-induced upwelling of water from the bedrock system to the regolith system was
documented at well cluster BC-2 (discharge area with artesian conditions) using hourly water quality
data. Hourly recordings of water level, SC, and DO in lower regolith well BC-2I from February 2005 to
March 2005 were logged and plotted (fig. 52A). As storm-induced water levels increased rapidly, SC
temporarily increased, and DO temporarily decreased in lower regolith well BC-2I. Water in the bedrock
system at BC-2 contains significantly higher SC and significantly lower DO than water in the regolith
system (median SC in bedrock well BC-2D is 214 uS/cm and in BC-2I is 19 uS/cm; DO in bedrock well
BC-2D is 0.1 mg/L and in BC-2I is 6 mg/L). As shown in fig. 52, one inch of rain occurred in a 24-hour
period between February 20 and 21, 2005, and within hours resulted in a distinct and near
simultaneous increase in SC (from 10 to 38 uS/cm) and water levels (fig. 52A) and a decrease in DO
(from 8 to 5 mg/L)(fig. 52B). All but about 30 percent of the rise/fall in levels dissipated over the next 7
days, consistent with patterns observed during other rainfall events during the 14-month measurement
period. Site 2 is in a discharge area with a strong upward vertical head gradient from the deeper
bedrock system to the regolith system (fig. 25F). Rainfall recharge to the upper regolith may not
penetrate downward into the lower regolith system in this location, and may instead move laterally and
discharge to Boyd Branch. Vertical head gradients within the regolith itself (between the upper regolith
and lower regolith) are mostly neutral, and temperature in lower regolith well BC-2I remained stable
during individual rainfall events (figs. 35A, 36A) while temperature in upper regolith well BC-2S
fluctuated (figs. 35A and 36A). This is consistent with conditions in a discharge area and has been
documented at a NCDWQ-USGS Piedmont groundwater research station near Raleigh, North Carolina
(Chapman and others, 2005).

0
5
10
15
20
25
BC
-
1
S
BC
-
1
i
BC
-
1
D
BC
-
2
S
BC
-
2
i
BC
-
2
D
BC
-
3
S
BC
-
3
i
BC
-
3
D
BC
-
4
S
BC
-
4
i
BC
-
4
D
BC
-
5
S
BC
-
5
i
BC
-
5
D
BC
-
6
D
BC
-
7
S
BC
-
7
i
BC
-
7
D
BB

a
t

B
C
2
BB

t
r
i
b

a
t

B
C
3
BB

t
r
i
b

a
t

B
C
7
Be
n
t

C
r
e
e
k
mg
/
L
Silica, mg/L
Figure. 51. Median silica concentrations observed in stream and well samples during periodic
sampling events between 2003 and 2007, Bent Creek Research Station, North Carolina.
Median values obtained using data
from multiple sample dates
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
89

0.0
0.1
0.2
0.3
0.4
0.5
15
-
F
e
b
-
0
5
16
-
F
e
b
-
0
5
17
-
F
e
b
-
0
5
18
-
F
e
b
-
0
5
19
-
F
e
b
-
0
5
20
-
F
e
b
-
0
5
21
-
F
e
b
-
0
5
22
-
F
e
b
-
0
5
23
-
F
e
b
-
0
5
24
-
F
e
b
-
0
5
25
-
F
e
b
-
0
5
26
-
F
e
b
-
0
5
27
-
F
e
b
-
0
5
28
-
F
e
b
-
0
5
1-
M
a
r
-
0
5
2-
M
a
r
-
0
5
3-
M
a
r
-
0
5
4-
M
a
r
-
0
5
5-
M
a
r
-
0
5
Ho
u
r
l
y

R
a
i
n
f
a
l
l
,

i
n
c
h
e
s
52
52.5
53
53.5
54
54.5
55
Gr
o
u
n
d

W
a
t
e
r

T
e
m
p
e
r
a
t
u
r
e
,

F
Hourly rain,
inches, NOAA
raingage 0246CGround Water
Temperature, F
6
6.2
6.4
6.6
6.8
7
7.2
7.4
7.6
7.8
15
-
F
e
b
-
0
5
16
-
F
e
b
-
0
5
17
-
F
e
b
-
0
5
18
-
F
e
b
-
0
5
19
-
F
e
b
-
0
5
20
-
F
e
b
-
0
5
21
-
F
e
b
-
0
5
22
-
F
e
b
-
0
5
23
-
F
e
b
-
0
5
24
-
F
e
b
-
0
5
25
-
F
e
b
-
0
5
26
-
F
e
b
-
0
5
27
-
F
e
b
-
0
5
28
-
F
e
b
-
0
5
1-
M
a
r
-
0
5
2-
M
a
r
-
0
5
3-
M
a
r
-
0
5
4-
M
a
r
-
0
5
5-
M
a
r
-
0
5
Wa
t
e
r

L
e
v
e
l
,

f
e
e
t

b
e
l
o
w

l
a
n
d

s
u
r
f
a
c
e
0
10
20
30
40
50
60
Sp
e
c
i
f
i
c

C
o
n
d
u
c
t
a
n
c
e
,

u
S
/
c
m
WL, ft bls
SC, uS/cm
6
6.2
6.4
6.6
6.8
7
7.2
7.4
7.6
7.8
15
-
F
e
b
-
0
5
16
-
F
e
b
-
0
5
17
-
F
e
b
-
0
5
18
-
F
e
b
-
0
5
19
-
F
e
b
-
0
5
20
-
F
e
b
-
0
5
21
-
F
e
b
-
0
5
22
-
F
e
b
-
0
5
23
-
F
e
b
-
0
5
24
-
F
e
b
-
0
5
25
-
F
e
b
-
0
5
26
-
F
e
b
-
0
5
27
-
F
e
b
-
0
5
28
-
F
e
b
-
0
5
1-
M
a
r
-
0
5
2-
M
a
r
-
0
5
3-
M
a
r
-
0
5
4-
M
a
r
-
0
5
5-
M
a
r
-
0
5
Wa
t
e
r

L
e
v
e
l
,

f
e
e
t

b
e
l
o
w

l
a
n
d

s
u
r
f
a
c
e
3
3.5
4
4.5
5
5.5
6
6.5
7
7.5
8
Di
s
s
o
l
v
e
d

O
x
y
g
e
n
,

m
g
/
L
WL, ft bls
DO, mg/L
Figure. 52. Hourly recordings in lower regolith well BC-2I for (A) specific
conductance, (B) dissolved oxygen and water levels, and (C) ground-water
temperature, in response to storm events from February 15 to March 5, 2005, Bent
Creek Research Station, North Carolina.
C
B
A
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
90

Nutrients and Bacteria

Several nutrients were analyzed at study area wells and stream locations, including TKN, nitrate
plus nitrite as nitrogen (N), ammonia as N, nitrate as N, nitrite as N, total phosphorous as P, and
phosphate. Nutrient concentrations were very low or not detected (below 0.4 mg/L) in all wells and
stream samples, with the exception of a nitrate value of 2.7 mg/L at BC-4S during one sample event in
2003 (all four other sample events at BC-4S showed nitrate levels below the detection limit). Most
samples were below the detection limit for all nutrients analyzed. Detections were somewhat more
common in stream samples than well samples, and detections were slightly more common in regolith
wells than in bedrock wells. Locations that are vulnerable to surface runoff of fertilizers and animal
wastes (such as streams and shallow wells) are more likely to have nutrient detections than properly
constructed bedrock wells. No samples exceeded state or federal nutrient standards.
Fecal and total coliform values generally were near or below 1 colony per mL for all wells.
Positive detections of total and fecal coliform occurred in all sampled surface waters, with a total
coliform range of 130 to 900 colonies per mL and a fecal coliform range of 34 to 76 colonies per mL.
The highest observations occurred in the tributary to Boyd Branch, adjacent to BC-7.

Volatile Organics, Semi-Volatile Organics, Pesticides, Herbicides, and Other
Constituents and Measures

As expected in the pristine setting of the study area, pesticides, herbicides, semivolatile
organics, and volatile organic compounds essentially were absent from all sampled wells and streams
in this pristine study area. Minor exceptions were attributed to artifacts of the sampling and analytical
process.
Total organic carbon was below the detection limit of between 2 and 5 mg/L for all well and
stream samples. Similarly, biological oxygen demand and chemical oxygen demand were below their
respective detection limits of 2 mg/L and 20 mg/L for all well and stream samples. Total suspended
solids (TSS) values were generally at or near the detection limit of between 2.5 and 6.2 mg/L. An
exception was at BC-7S where the TSS concentration was about 30 mg/L.

Naturally Occurring Radionuclides and Arsenic

The primary naturally occurring compounds measured during the study included total uranium,
radium-226, radon-222 (radon), and arsenic. Gross beta activity also was measured and occurs both
naturally and as a man-made product from the nuclear industry. Total uranium levels were very low or
were below the detection limit of 0.02 μg/L, with 22 of 23 locations below 1.6 μg/L. The exception was
a value of 5 μg/L at BC-4D. The EPA MCL for uranium is 30 μg/L. Radium-226 values also were very
low with 22 of 23 locations below 0.09 picocuries per liter (pCi/L) and one location (BC-1D) at 0.38
pCi/L. The EPA MCL for combined radium-226 + radium-228 is 5 pCi/L.
Dissolved radon was detected above 100 pCi/L at 16 of 17 wells and ranged from 90 to 5310
pCi/L (median for all wells = 1300 pCi/L) (fig. 53). Average radon values were higher at well cluster BC-
1 (average = 4040 pCi/L) and well cluster BC-5 (average = 2250 pCi/L) than well clusters BC-2, -3, -4,
and -7, which averaged 1520, 1150, 1400, and 550 pCi/L, respectively. Concentrations in regolith wells
were not significantly different than concentration in bedrock wells. Because radon volatilizes rapidly, it
was not measured in streams. The EPA has proposed an MCL of 300 pCi/L for radon in water and an
alternate standard of 4000 pCi/L for water suppliers with a radon mitigation and outreach program (U.S.
EPA, 2003), but, to date, neither standard has been promulgated. Radon is a natural decay product of
uranium-238 (its immediate parent is radium-226) and is common in uranium rich granitic rocks and, to
a lesser degree, in other rocks in the Piedmont-Mountains of North Carolina, including schists like those
in the study area. The lifetime health risk associated with ingesting radon in water at the median value
of 1300 pCi/L is about 1 in 13,000 (NRC, 1999). Arsenic was below the detection limit of between 5
and 10 μg/L at all sampled wells and streams.

Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
91

Comparison of Groundwater Chemistry to Rock Chemistry

Whole-rock analyses from core samples (three depths at each of three locations, BC-1, BC-3,
and BC-7) were compared to groundwater chemistry from the seven bedrock wells (table 16). Rock
core recovered from these locations consisted of interlayered schist, biotite gneiss/metagraywacke, and
migmatite. Using the average oxide percentage from these samples (all depths and locations), the
dominant whole rock ion species were silica (51%), aluminum (20%), iron (9%), potassium (3.6%),
magnesium (3.2%), calcium (2.4%), and sodium (1.4%). One sample (BC-3, at a depth of 125 ft) had a
relatively high calcium oxide content of 11.6%. From the samples analyzed by Merschat and Carter
(2002) across the Bent Creek watershed, the dominant whole rock ion species for schist,
metagraywacke, and migmatite samples were silica (44%, 70%, and 67%, respectively), aluminum
(26%, 13%, and 17%, respectively), iron (13.2%, 5.3%, and 3.5%, respectively), potassium (4.5%,
2.6%, and 2.6%, respectively), magnesium (2.8%, 1.6%, and 1.1%, respectively), sodium (0.8%, 2.8%,
and 5.4%, respectively), and calcium (0.4%, 2.5%, and 2.6%, respectively). In comparison, median ion
values from all bedrock well groundwater results were calcium (18 mg/L), silica (17 mg/L), sodium (3.7
mg/L), potassium (2 mg/L), magnesium (1.3 mg/L), iron (0.2 mg/L), and aluminum (<0.05 mg/L).

SUMMARY

The BCRS is one of ten hydrogeologic research stations in the Piedmont-Mountains installed to
evaluate groundwater availability, flow, and quality as part of the cooperative NCDENR-USGS REP.
The BCRS represents a well-foliated mica schist geologic type area, with lesser amounts of
interlayered, poorly foliated biotite gneiss and migmatite.
Groundwater in the study area occurs within a three-part flow system: upper regolith, lower
regolith, and fractured bedrock. The upper regolith system has a high porosity and variable
permeability and consisted of sandy, clayey, silty residuum-saprolite. The lower regolith system has a
variable porosity and high permeability and consisted of partially weathered transition zone material
with both primary porosity (pore spaces) and secondary porosity (rock partings and veins). The
0
1000
2000
3000
4000
5000
6000
BC
-
1
S
BC
-
1
i
BC
-
1
D
BC
-
2
i
BC
-
2
i
BC
-
3
S
BC
-
3
i
BC
-
3
D
BC
-
4
S
BC
-
4
i
BC
-
4
D
BC
-
5
S
BC
-
5
i
BC
-
5
D
BC
-
7
S
BC
-
7
i
BC
-
7
D
pC
i
/
L
Dissolved radon-222, picocuries per liter, April 23, 2003
Figure. 53. Median dissolved radon-222 concentrations observed in well samples during periodic
sampling events between 2003 and 2007, Bent Creek Research Station, North Carolina.
Median values obtained using data
from multiple sample dates
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
92
thickness of the regolith system measured in the study area ranged from 33 to 60 feet (median = 48 ft).
The upper regolith ranged from 2 to 38 feet thick (median = 23 ft), and the lower regolith ranged from 5
to 46 feet thick (median = 19 ft). Because of differential weathering, the thickness, composition, and
permeability of the upper and lower flow systems varied with location, as did the depth to bedrock, and
interlayering of saprolite, transition zone material, and un-weathered bedrock occurred in 3 of 7 core
holes. This finding has implications for the proper design and construction of contamination
assessment and water supply wells at similar settings within the Piedmont and Mountain regions.
Depth to competent bedrock ranged from 33 to 60 feet below land surface. Bedrock foliations
strike predominantly to the NE and dip to the NW and SE at angles of 25 to 85 degrees. Bedrock
fractures generally cross cut foliation and strike predominantly NW, dipping to the NE or SW at angles
of 50 to 70 degrees. Less often, fractures are parallel to foliation or are of a stress relief type.
Lower regolith (transition zone) wells transmitted water much more readily than did upper
regolith (saprolite) wells, with a median yield and hydraulic conductivity of 5 gallons per minute and 8
feet per day, respectively, as compared to 1.5 gpm and 2 feet per day for saprolite wells.
Bedrock was characterized by low porosity and permeability. Bedrock well yields ranged from
less than 0.5 to 40 gpm (median = 1 gpm and average = 7 gpm). These yields are far lower than the
average yield of 16 to 22 gpm reported in similar hydrogeologic units elsewhere in the Piedmont-
Mountains of North Carolina (Daniel, 1989). This may be due to the combined effect of lithology
(predominantly schist, whose laminated cleavage planes can readily fold and deform without fracturing),
structure (a relative lack of low angle water bearing fractures and foliations), and the fact that the wells
were not drilled in locations or to depths that were intended to maximize yields. Rock coring, heat pulse
flow metering, and optical televiewer imaging indicated relatively few primary (open and significant)
fractures at seven well locations. Results from a 45-hour aquifer pumping test conducted in a bedrock
well support this assessment and demonstrated a lack of connectivity between the pumped well and
two bedrock wells located within 1500 feet. The response in overlying regolith wells within 250 feet
generally was modest and sluggish, with total drawdowns of about a half foot or less.
Groundwater flow at BCRS generally is from topographic highs to lows, consistent with
conceptual models of flow in the Blue Ridge and Piedmont physiographic provinces. Vertical head
gradients generally are downward in upslope and midslope recharge areas and upward in downslope
discharge areas. Storm events produced a rapid, temporary increase of about 0.5 to 3 feet in water
levels in regolith and bedrock wells, which dissipated after 4 to 8 days. Water temperature and vertical
head gradients in clustered wells were used to demonstrate upwelling from the deeper to the shallower
groundwater system during rain events in a low-lying discharge area. Data indicate that at least some
recharge infiltrates to the water table within 0.5 to 2 days of rainfall events in two low-lying discharge
areas. Age estimates of groundwater in four bedrock wells ranged from about 20 to 50 years, and,
within a given well, a mixture of ages was common (including water less than 2 years old in at least one
upslope bedrock well). Observing a mixture of older and younger waters within bedrock wells is
consistent with the regional conceptual model in which multiple potential flow paths (and source waters)
contribute to the total volume of water in a bedrock well.
Bent Creek, a 3rd order stream draining the larger watershed, is a major control on groundwater
flow direction in the Boyd Branch sub-basin. Locally, groundwater also discharges to Boyd Branch, a
small, weakly incised 2nd order tributary to Bent Creek. Groundwater under-flowed Boyd Branch in one
measured reach. This is significant when considering issues of contaminant transport and groundwater
flow direction in mountain watersheds, and demonstrates the importance of understanding several
factors that may affect groundwater flow direction and potential underflow of mountain streams,
including: depth to water, depth of stream incision, underlying geology and thickness of regolith, size
and shape of recharge boundaries, and relative topographic differences in both the sub-basin and the
larger watershed of which it is part.
The water table typically occurred at a depth of about 3 to 32 feet below land surface and varied
seasonally and with topographic setting. Artesian flow conditions occur at one downslope location.
Water levels tended to be lowest in late summer when evapotranspiration effects tend to be greater
than recharge, and highest in late winter. Groundwater temperatures lagged seasonal air temperatures
by 1 to 4 months in the upper regolith and by 3 to 7 months in the lower regolith. Temperatures in
bedrock wells remained stable across seasons and storm events.
Annual mean streamflow in Bent Creek is estimated to be about 15 cfs (6700 gpm) based on
hydrograph separation computations using stream flow records and the USGS PART program
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
93
(Rutledge, 1998). The program calculated baseflow in the pristine, largely undisturbed forest to be
about 80 percent of this amount, or 12 cfs (5400 gpm). Hourly mean stream flow in Boyd Branch
ranged from 0.5 (220 gpm) to 56 cfs (25,000 gpm) for the period between March 2004 and November
2005, with a median of 1.5 cfs (670 gpm) at USGS gage 0344789265 in Boyd Branch at Bent Creek
Road near Lake Powhatan. Baseflow was not estimated due to the limited period of record.
The dominant water type for bedrock wells is calcium bicarbonate, and for streams and regolith
wells is sodium/magnesium-bicarbonate. Ground and surface water are relatively soft, of high quality,
and suitable for most domestic and industrial-use purposes. Values of pH, calcium, magnesium, and
sulfate were higher than typical regional levels in some of the BCRS bedrock wells, due in part to the
rock type and local geochemistry. Deeper groundwater contained higher amounts of dissolved ions
than shallow groundwater due to increased residence time for the dissolution of minerals. Iron and
manganese staining was common in the regolith at core locations (primarily within the transition zone),
indicative of oxidation-reduction reactions in the weathered, aerobic portion of the aquifer. Staining
occurred less frequently in the underlying bedrock fractures, particularly in anaerobic settings such as
discharge areas. Fecal bacteria, nutrients, pesticides, herbicides, volatile and semi-volatile organic
compounds, and naturally occurring arsenic generally were not detected. Naturally occurring uranium
and radium-226 levels generally were near or below the detection limit, and radon-222 was found above
the proposed EPA standard of 300 picocuries per liter in 3 deep wells, with a high of 3030 picocuries
per liter.

SELECTED REFERENCES

Bear, J., 1972, Dynamics of Fluids in Porous Media, Hardcover, Dover Publications.

Bouwer, H., and Rice, R.C., 1976, A slug test for determining hydraulic conductivity of unconfined
aquifers with completely or partially penetrating wells: Water Resources Research v. 12, no. 3, p. 423-
428.

Briel, L.I., 1997, Water quality in the Appalachian Valley and Ridge, the Blue Ridge, and the Piedmont
Physiographic Provinces, Eastern United States: U.S. Geological Survey Professional Paper 1422-D,
115 p.

Campbell, T.R., 2006, Radon-222 and other naturally-occurring radionuclides in private drinking water
wells and radon in indoor air in Buncombe, Henderson, and Transylvania Counties, North Carolina,
2005: North Carolina Department of Environment and Natural Resources, Division of Water Quality,
Ground Water Circular Number 20, 81 p.

Campbell, T.R., 2006b, Radon-222 and other naturally-occurring radionuclides in private drinking water
wells and radon in indoor air in selected counties in western North Carolina, 2006: North Carolina
Department of Environment and Natural Resources, Division of Water Quality, Ground Water Circular
Number 2006–01, 27 p.

Campbell, T.R., 2008, Radon-222 and other naturally-occurring radionuclides in private drinking water
wells and radon in indoor air in selected counties in western North Carolina, 2007: North Carolina
Department of Environment and Natural Resources, Division of Water Quality, Ground Water Circular
Number 2008–01, 37 p.
Chapman, M.J., Bolich, R.E., and Huffman, B.A., 2005, Hydrogeologic setting, ground-water-
Flow, and ground-water quality at the Lake Wheeler Road Research Station, 2001-03, North
Carolina Piedmont and Mountains Resource Evaluation Program: U.S. Geological Survey
Scientific Investigations Report 2005-5166, 85 p.
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
94
Daniel, C.C. III, and Dahlen, P.R., 2002, Preliminary hydrogeologic assessment and study plan for a
regional ground water resource investigation of the Blue Ridge and Piedmont Provinces of North
Carolina: U.S. Geological Survey Scientific Investigations Report 02-4105, 60 p.
Daniel, C.C., III, and Harned, D.A., 1998, Ground-water recharge to and storage in the regolith-
fractured crystalline rock aquifer system, Guilford County, North Carolina: U.S. Geological
Survey Water-Resources Investigations Report 97-4140, 65 p.
Daniel, C.C. III, Smith, D.G., and Eimers, J.I., 1997, Hydrogeology and simulation of ground-water flow
in the thick regolith-fractured crystalline rock aquifer system of Indian Creek basin, North Carolina: U.S.
Geological Survey Water-Supply Paper 2341-C, 137 p.
Daniel, C.C., III, 1996, Ground-water recharge to the regolith-fractured crystalline rock aquifer system,
Orange County, North Carolina: U.S. Geological Survey Water-Resources Investigations Report 96-
4220, 59 p.
Daniel, C.C. III, 1992, Correlation of well yield to well depth and diameter in fractured crystalline rocks,
North Carolina, in Daniel, C.C., III, White, R.K., and Stone, P.A., eds., Ground water in the Piedmont:
Clemson, South Carolina, Clemson University, Conference on Ground Water in the Piedmont of the
Eastern United States, Proceedings, p. 638-653.

Daniel, C.C. III, and Payne, R.A., 1990, Hydrogeologic Unit Map of the Piedmont and Blue Ridge
Provinces of North Carolina: US Geological Survey Water Resources Investigations Report 90-4035, 1
map sheet, scale 1:500,000.

Daniel, C.C. III, 1989, Statistical analysis relating well yield to construction practices and siting of wells
in the Piedmont and Blue Ridge Provinces of North Carolina: US Geological Survey Water-Supply
Paper 2341-A, 27 p.
Daniel, C.C., III, and Sharpless, N.B., 1983, Ground-water supply potential and procedures for well-site
selection in the upper Cape Fear River Basin, North Carolina: North Carolina Department of Natural
Resources and Community Development, 73 p.
Driscoll, F.G., 1986, Groundwater and Wells, Hardcover, Publisher: Reynolds Guyar Designs.
Fenneman, N.M., and Johnson, D.W., 1946: Map of physiographic divisions of the conterminous United
States, U.S. Geological Survey.

Fetter, C.W., 2001, Applied Hydrogeology, 4th Edition Hardcover, Publisher: Prentice Hall.

Freeze, R. A. and J. A. Cherry, 1979, Groundwater, Publisher: Prentice Hall, Inc.
Giese and Mason, Low-flow characteristics of streams in North Carolina: US Geological Survey Water-
Supply Paper 2403, 1993, 31 p.

Harned, D.A., and Daniel, C.C. III, 1992, The transition zone between bedrock and saprolite - conduit
for contamination?, in Daniel, C.C., White, R.K., and Stone, P.A., eds., Ground Water in the Piedmont:
Clemson, SC, Clemson University, p. 336-348.

Harned, D.A. and Daniel C.C., III, 1987, Ground-water component of Piedmont streams; implications
for ground water supply systems and land-use planning: Geological Society of America Abstracts with
diagrams, v. 19, no. 2, 89 p.

Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
95
Hatcher, R.D., Jr., 1971, Geology of Rabun and Habersham Counties, Georgia: A reconnaissance
study: Georgia Geological Survey Bulletin 83, 48p.
Heath, R.C., 1980, Basic elements of ground-water hydrology with reference to conditions in North
Carolina: U.S. Geological Survey Water-Resources Investigations Open-File Report 80-44, 86 p.
Heath, R.C., 1983, Basic ground water hydrology: U.S. Geological Survey Water Supply Paper 2220,
84 p.

Heath, R.C., 1984, Ground water regions of the United States: U.S. Geological Survey Water Supply
Paper 2242, 78 p.

Heath, R., 1994, Ground water recharge in North Carolina, NC Department of Environment and Natural
Resources and NC State University, Ground Water Report, 52 p.

Huffman, B.A., Pfeifle, C.A., Chapman, M.J., Bolich, R.E., Campbell, T.R., Geddes, D.J., and Pippin,
C.G., 2006, Compilation of water-resources data and hydrogeologic setting for four research stations in
the Piedmont and Blue Ridge Physiographic Provinces of North Carolina, 2000-2004: U.S. Geological
Survey Open-File Report 2006-1168, 102 p.

LeGrand, H.E., 1967, Ground water of the Piedmont and Blue Ridge Provinces in the Southeastern
States: Geological Survey Circular 538, 11 p.

LeGrand, H.E., and Nelson, P.F., 2004, A master conceptual model for hydrogeologic site
characterization in the Piedmont and Mountain region of North Carolina, A guidance manual: Raleigh,
North Carolina Department of Environment and Natural Resources, Division of Water Quality,
Groundwater Section Ground Water Report, 50 p.
McSwain, McSwain, K.B., Bolich, R.E., Chapman, M.J., and Huffman, B.A., 2009, Water-
resources data and hydrogeologic setting at the Raleigh hydrogeologic research station, Wake
County, North Carolina, 2005-2007: U.S. Geological Survey Open-File Report 2008-1377, 48 p.
(only online at http://pubs.water.usgs.gov/ofr2008-1377).
Merschat, Carl E., and Carter, Mark W., 2002, Bedrock geologic map of the Bent Creek Research and
Demonstration Forest, Southern Research Station, USDA Forest Service, including the North Carolina
Arboretum and a portion of the Blue Ridge Parkway: North Carolina Geological Survey, Geological
Survey Map Series 9, scale 1:12,000.

Mesko, T.O., Swain, L.A., and Hollyday, E.F., 1999, Hydrogeology and hydrogeologic terranes of the
Blue Ridge and Piedmont Physiographic Provinces in the Eastern United States: US Geological Survey
Hydrologic Investigations Atlas HA-732-B, US.

Mew, H.E., Hirth, D.K., Lewis, D.K., Daniels, R.B., and Keyworth, A.J., 2002, Methodology for compiling
ground water recharge maps in the Piedmont and Coastal Plain provinces of North Carolina: North
Carolina Department of Environment and Natural Resources, Ground Water Bulletin Number 25, 68 p.

National Research Council, 1999, Risk assessment of radon in drinking water. National Academy
Press.

North Carolina Division of Water Quality, Aquifer Protection Section, April 2008, Standard operating
procedures for conducting hydrogeologic field investigations at groundwater research stations, North
Carolina Piedmont and Mountains, Groundwater Resource Evaluation Program, North Carolina
Department of Environment and Natural Resources, Division of Water Quality, Ground Water Report,
89 p., accessed at http://h2o.enr.state.nc.us/gwp/documents/GWREP_SOP-Draft_Dec_08.pdf.

Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
96
North Carolina Geological Survey, 1985, Geologic map of North Carolina: Raleigh, North Carolina
Geological Survey, 1 sheet, scale 1:500,000.

Piper, A.M., 1953, A graphic procedure in the geochemical interpretation of water analyses: U.S.
Geological Survey Ground-Water Chemistry Notes, no. 12, 14 p.

Pippin, C.G., Chapman, M.J., Huffman, B.A., Heller, M.J., and Schelgel, M.E., 2008, Hydrogeologic
setting, ground-water flow, and ground-water quality at the Langtree Peninsula Research Station, Iredell
County, North Carolina, 2000-2005: U.S. Geological Survey Scientific Investigations Report 2008-5055,
54 p.

Pomeroy, J.S.,1991, Map showing late 1977 debris avalanches southwest of Asheville, western North
Carolina: U.S. Geological Survey Open File Report 91-334, scale 1:24,000.
Reid, Jeffrey C., 1993a, A geochemical atlas of North Carolina, U.S.A., in F.W. Dickson and L.C. Hsu
(Editors), Geochemical Exploration 1991, J. Geochemical Exploration, v. 47, p. 11-27.
Reid, Jeffrey C., 1993b, A hydrochemical atlas of North Carolina. Bulletin 94. North Carolina Geological
Survey: Raleigh.
Roscoe Moss Company, 1990, Handbook of ground water development, Publisher: John Wiley & Sons.
Rutledge, A.T., 1998, Computer programs for describing the recession of ground-water
discharge and for estimating mean ground-water recharge and discharge from streamflow data
– update: U.S. Geological Survey Water-Resources Investigations Report 98-4148, 43 p.
Rutledge, A. T. and Mesko, T. O., 1996, Estimated hydrologic characteristics of shallow aquifer systems
in the Valley and Ridge, the Blue Ridge, and the Piedmont physiographic provinces based on analysis
of streamflow recession and base flow: U.S. Geological Survey Professional Paper 1422-B, 58 p.

Seaton, W.J., and Burbey, T.J., 2005, Influence of ancient thrust faults on the hydrogeology of the Blue
Ridge Province, Journal of Ground Water, Volume 43, No. 3, May-June, p. 301-313.

Stewart, J.W., 1962, Water-yielding potential of weathered crystalline rocks at the Georgia Nuclear
Laboratory: U.S. Geological Survey Professional Paper 450-B, 2 p.

Stiff, H.A., Jr., 1951, The interpretation of chemical water analyses by means of patterns: Journal of
Petroleum Technology, v. 3, p. 15-17.

Swain, L.A., Mesko, T.O., and Hollyday, E.F., 2004, Summary of the hydrogeology of the Valley and
Ridge, Blue Ridge, and Piedmont Physiographic Provinces in the Eastern United States: U.S.
Geological Survey Professional Paper 1422-A, 23 p.

Trapp, H., 1970, Geology and ground-water resources of the Asheville Area North Carolina, U.S.
Geological Survey Ground Water Bulletin Number 16, 127 p.

U.S. Environmental Protection Agency, 2003. EPA assessment of risks from radon in homes. EPA 402-
R-03-003.

U.S. Geological Survey, 2005, Water use data tables, accessed from the internet
(http://nc.water.usgs.gov/wateruse/data/Data_Tables_2005.html) on December 10, 2008.

Webb, Stephen, 2005, A survey of ground water resources in the Cashiers, North Carolina Vicinity, NC
Department of Environment and Natural Resources, Division of Water Resources, Ground Water
Report 44 p.
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
97

White, B.A., and Burbey, T.J., 2007, Evidence for structurally controlled recharge in the Blue Ridge
Province, Virginia, USA, Hydrogeology Journal, 15: p. 929-943.

Winner, M.D., 1977, Ground water resources along the Blue Ridge Parkway, North Carolina: U.S.
Geological Survey Water Resources Investigations 77-65, 170 p.

Witt, A.C., 2005, Using a GIS to model slope instability and debris flow hazards in the French Broad
River Watershed, North Carolina, Master‟s Thesis, Marine, Earth, and Atmospheric Sciences
Department, North Carolina State University, 165 p.

Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
98

1 inch ~ 900 ft N
1
2
3
7
6
5
4
Approximate
boundary of
Boyd Branch
sub-basin
6
N ~ 900 feet
APPENDIX 1A. Geology and structural features in the vicinity of Boyd Branch, Bent Creek Experimental Forest
watershed, North Carolina, as modified from 1:12,000 scale map by Merschat and Carter, 2002.
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
99

1200 ft
N
1200 ft
N
1200 ft
N
1200 ft1200 ft
N

APPENDIX 1B. Geology and structural features in the vicinity of Boyd Branch, Bent Creek Experimental
Forest watershed, North Carolina, written communication, B. Burton, July 25, 2008.
Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
100

APPENDIX 2. Geophysical and optical televiewer logs for bedrock wells BC-1D at Bent Creek Research Station, North Carolina.

Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
102

APPENDIX 2 Continued. Geophysical and optical televiewer logs for bedrock wells BC-2D at Bent Creek Research
Station, North Carolina.

Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
103

APPENDIX 2, continued. Geophysical and optical televiewer logs for bedrock wells BC-3D at Bent Creek Research
Station, North Carolina.

Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
104

APPENDIX 2 Continued. Geophysical and optical televiewer logs for bedrock wells BC-4D at Bent Creek Research
Station, North Carolina.

Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
105

APPENDIX 2 Continued. Geophysical and optical televiewer logs for bedrock wells BC-5D at Bent Creek Research
Station, North Carolina.

Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
106

APPENDIX 2 Continued. Geophysical and optical televiewer logs for bedrock wells BC-7D at Bent Creek Research
Station, North Carolina.

Hydrogeology & Ground & Surface Water Quality at Bent Creek Research Station, 2002 - 2008
107

APPENDIX 3. Well-construction details for wells drilled at Bent Creek Research Station, North Carolina.

Well name Ground water
zone Latitude Longitude Date installed
Measuring
point
elevation
Land surface
elevation
Total
depth
Cased or
open
interval
screen/open
hole diameter
Well yield
(approximate)
ft msl ft msl ft bls ft bls in gpm
BC-1S transition zone 35.477591667 82.636153333 17-Oct-2002 2203.94 2201.0 22 8 - 22 4 5
BC-1I transition zone 35.477806667 82.636200000 16-Oct-2002 2205.29 2202.5 53 38 - 53 4 15
BC-1D bedrock 35.478015000 82.636575000 8-Oct-2002 2204.27 2201.8 221 54.7 - 221 6.25 OH 40
BC-2S saprolite 35.481786667 82.634983333 30-Oct-2002 2193.65 2190.7 20 5 - 20 4 1
BC-2I transition zone 35.481701667 82.634858333 24-Oct-2002 2194.61 2191.7 36 21 - 36 4 5
BC-2D bedrock 35.481588333 82.635120000 23-Oct-2002 2194.74 2190.2 300 53 - 300 6.25 OH 0.5
BC-3S saprolite 35.482388333 82.636666667 3-Oct-2002 2212.94 2210.1 30 15 - 30 4 0.5
BC-3I transition zone 35.482500000 82.636611667 3-Oct-2002 2212.51 2209.5 50 35 - 50 4 2
BC-3D bedrock 35.482361667 82.636695000 17-Sep-2002 2211.81 2209.1 300 61 - 300 6.25 OH <0.5
BC-4S saprolite 35.485611667 82.640750000 12-Sep-2002 2262.78 2259.7 22 7 - 22 4 2
BC-4I transition zone 35.485583333 82.640583333 11-Sep-2002 2261.82 2258.8 41 26 - 41 4 20
BC-4D bedrock 35.485528333 82.640528333 29-Aug-2002 2261.55 2258.5 501 61 - 501 6.25 OH 10
BC-5S saprolite 35.486221667 82.642916667 8-Aug-2002 2303 2300.0 24 9 - 24 4 3
BC-5I transition zone 35.486250000 82.643111667 22-Aug-2002 2304.89 2302.2 47 32 - 47 4 nt
BC-5D bedrock 35.486111667 82.643028333 12-Aug-2002 2307.42 2304.8 300 62 - 300 6.25 OH 1
BC-6D bedrock 35.488600000 82.644170000 4-Nov-2005 2363 2360.0 190 55-190 6.25 OH nt
BC-7S saprolite 35.490861667 82.644195000 17-Jul-2002 2371.19 2368.2 25 10 - 25 4 1
BC-7I transition zone 35.490888333 82.644028333 16-Jul-2002 2371.95 2369.0 45 30 - 45 4 1
BC-7D bedrock 35.490916667 82.644111667 10-Jul-2002 2372.81 2369.9 285 62 - 285 6.25 OH <0.5
P1S saprolite 16-Jun-2004 2263.21 2260.4 17 12 - 17 2 nt
P1I transition zone 15-Jun-2004 2262.83 2260.0 30 25 - 30 2 nt
P2S saprolite 22-Jun-2004 2255.01 2252.1 22 17 - 22 2 nt
P2I transition zone 23-Jun-2004 2254.95 2251.9 64 54 - 59 2 nt
P3S saprolite 23-Jun-2004 2254.28 2251.4 22 17 - 22 2 nt
P4S saprolite 24-Jun-2004 2235.27 2231.9 17 12 - 17 2 nt
P4I transition zone 24-Jun-2004 2234.78 2232.2 30 25 - 30 2 nt
P5S saprolite 30-Jun-2004 2259.86 2257.1 17 12 - 17 2 nt
P5I transition zone 30-Jun-2004 2259.25 2256.6 29 24 - 29 2 nt
P6S saprolite 13-Jul-2004 2260.96 2257.9 22 17 - 22 2 nt
P6I transition zone 1-Jul-2004 2260.59 2257.7 34 29 - 34 2 nt
P7S saprolite 14-Jul-2004 2274.26 2271.3 13.5 8.5 - 13.5 2 nt
P7I transition zone 20-Jul-2004 2276.06 2273.3 25 20 - 25 2 nt
P8S saprolite 14-Jul-2004 2259.69 2256.7 20 15 - 20 2 nt
P8I transition zone 14-Jul-2004 2259.69 2256.7 34 29 - 34 2 nt
ft, feet; msl, mean sea level; bls, below land surface; OH, open hole; in, inches; gpm, gallons per minute; nt, not tested
see map on figure 10
see map on figure 10
see map on figure 10
see map on figure 10
see map on figure 10
see map on figure 10
see map on figure 10
see map on figure 10
see map on figure 10
see map on figure 10
see map on figure 10
see map on figure 10
see map on figure 10
see map on figure 10
see map on figure 10
23