Laurie K. Loftin. WATER CHEMISTRY OF THE POCOSINS OF THE CROATAN
NATIONAL FOREST, CARTERET, CRAVEN AND JONES COUNTIES, NORTH CAkOLINA.
(Under the direction ot Dr. Lee J. Otte) Department of Geology, October
1985.
ABSTRACT
The Croatan National Forest, located on the central portion of the
lower North Carolina Coastal Plain, contains one of the largest and least
disturbed peat-bearing wetlands in the state. Within the Croatan system,
three different water regimes have been chemically distinquished : (1)
standing or flowing surface waters, (2) peat pore waters, and (3) mineral
pore waters. Upon analysis for 27 ions, these water regimes were
chemically differentiated, based on concentrations of the following
dissolved ions: A1, Ca, K, Mg, Mn, Na, NH¿j-N, NO3-N, Si02~Si, total
dissolved PO¿^-P, total organic C, Ti, and V.
Optimum indicators of the three water regimes were total organic C,
Si, and Ca. Mineral pore waters contained the highest amounts of
dissolved Si (ave •• 3.55 mg/1), intermediate levels of Ca (ave * 1.51
mg/1), and the lowest concentrations of total organic C (ave « 42.7
mg/1). Peat pore waters contained the highest concentrations of total
organic C (ave » 212.5 mg/1) and Ca (ave - 2.09 mg/1), and intermediate
Si concentrations (ave - 0.89 mg/1) of the three regimes. Standing
surface waters contained the lowest concentrations of dissolved Si (ave =
0.45 mg/1) and Ca (ave - 0.20 mg/1) and had intermediate concentrations
of total organic C (ave • 51.0 mg/1).
The seasonality of outflow waters was as follows: highest total
organic Cj Ca, and Si concentrations corresponded to periods of Lowest
water table, when peat and mineral pore waters were draining. During the
winter and spring, increased surface runoff and the subsequent addition
of its chemical components caused total organic C, Ca, and Si
concentrations in the drainage canals to decline. Potassium (K),
although not an indicator species for any particular water regime,
exhibited a seasonality in outflow waters. Concentrations of K in
drainage waters increased during periods of low flow (summer and fall)
and declined during periods of high flow (winter and spring). Other ions
present in the outflow waters, as well as in the well waters, were not
indicative of a specific water regime, nor did they exhibit a strong
degree of seasonality in outflow.
WATER CHEMISTRY OF THE POCOSINS
OF THE CROATAN NATIONAL FOREST
CARTERET, CRAVEN, AND JONES COUNTIES,
NORTH CAROLINA
A Thesis
Presented to
the Faculty of the Department of Geology
East Carolina University
In Partial Fulfillment
of the Requirements for the Degree
Master of Science in Geology
by
Laurie K. Loftin
October, 1985
WATER CHEMISTRY OF THE POCOSINS
OF THE CROATAN NATIONAL FOREST
CARTERET, CRAVEN, AND JONES COUNTIES,
NORTH CAROLINA
by
Laurie K. Loftin
APPROVED BY:
DIRECTOR OF THESIS
THESIS COMMITTEE
Mark M. Brinson, Ph.D.
DEAN OF THE GRADUATE SCHOOL
ACKNOWLEDGEMENTS
The completion of this project has been achieved through the
contributions of several people. The rangers of the Croatan National
Forest were helpful and encouraging during the initial stages of this
study, which was funded by the Water Resources Research Institute
(project no. 84-02-62002). I am grateful to Dr. Lee J. Otte, my thesis
chairman, for his inspiration and tireless efforts throughout this
project, but especially thankful for his helping me to find my niche in
geology. I would like to recognize my thesis committee for their
thorough review of this manuscript; Dr. John T. Bray, Dr. Charles Q.
Brown, and Dr. Mark M. Brinson, my outside advisor.
For the lab space provided by Dr. John Bray, Martha Jones, and the
folks at the Institute for Coastal and Marine Resources, I am sincerely
grateful. Special appreciation is extended to Martha Jones, Grace
Lexton, and Debbie Daniel for their guidance and patience while I learned
the necessary lab procedures. For the analyses performed for me, I thank
Frank Reilly (trace element analyses), the Clay Mineralogy class of 1984
(clay analyses), and Dr. Kevin O'Brien (statistical analyses).
Several friends should also be recognized for the generous giving of
their time during the initial field work of well emplacement: Garland
Loftin (my father), Mike Lyle, David Mallinson, Lee Otte, Chee Saunders,
and Jim Watson. Special acknowledgements are due to Garland Loftin and
Mike Lyle for seeing to it that I was not "out there alone" each month.
In addition, I wish to thank Brenda Smith for the inspiration and example
she's set for me, and Teri Moore for the support I've needed to sprint
the last 100 yards.
My parents deserve the final and most important "thank you" for their
encouragement through the years, their patience while I pursued this
degree, and the funding that made it possible.
TABLE OF CONTENTS
Page
LIST OF TABLES iv
LIST OF FIGURES v
LIST OF APPENDIX TABLES vii
INTRODUCTION 1
OBJECTIVES 3
GEOGRAPHIC SETTING 4
PREVIOUS INVESTIGATIONS 7
CLIMATE 10
REGIONAL GEOLOGY 13
HYDROLOGY 19
VEGETATION 23
PROCEDURES 27
Field Methods 27
Laboratory Methods 28
Statistical Methods 29
RESULTS AND DISCUSSION 31
2H 35
Indicator Species 37
Total Organic Carbon 37
Silica 41
Calcium 44
Non-Indicator Species 48
Nitrogen 48
Nitrate 48
Ammonium 50
Total Dissolved Phosphorus 52
Magnesium/Potas slum/Sodium 55
Aluminum/Manganese/Titanium/Vanadium 61
SUMMARY 65
REFERENCES CITED 68
APPENDIX 73
iii
LIST OF TABLES
Page
1. Comparison of water content of peat soils near drainage canals
and away from drainage canals in the Croatan National Forest... 22
2. Number of water samples from pocosin environments during the
twelve month period 33
3. Comparison of water chemistry in surface, peat, and mineral waters
of the Croatan with Pungo National Wildlife Refuge, Peat Methanol
Associates area, Minnesota perched bog streams, and West Brice
Creek 34
IV
LIST OF FIGURES
Page
1. Environments of peat accumulation on the North Carolina Coastal
Plain 2
2. Location map of the Croatan National Forest with peat isopachs. 5
3a. 35 year average monthly precipitation (cm) from Maysville 6 SW
station 11
3b. Monthly average precipitation (cm) from Maysville 6 SW station
during sample year (6/83 - 5/84) 11
4a. 31 year average monthly temperature (oC) for Maysville 6 SW
station 12
4b. Monthly average temperature (oC) for Maysville 6 SW station
during sample year (6/83 - 5/84) 12
5. Soil profile prepared from mineral sediments collected along
Transect  16
6. Soil profile prepared from mineral sediments collected along
Transect B 17
7. Location of study area and sampling sites, in association with
the natural and man-made drainage systems 25
8a. Monthly average TOC concentrations for the three water regimes. 38
8b. Average TOC concentrations for the peat waters compared to peat
thickness 38
9. Monthly average TOC concentrations for standing and flowing
surface waters 40
lOa. Monthly average Si concentrations for the three water regimes.. 42
10b. Monthly average Si concentrations for standing and flowing
surface waters 42
lla. Monthly average Ca concentrations for the three water regimes.. 46
llb. Monthly average Ca concentrations for standing and flowing
surface waters 46
12a. Monthly average NO^-N concentrations for the three water
regimes 49
V
LIST OF FIGUEES (continued)
Page
12b. Monthly average NH,-N concentrations for the three water
regimes.
13a. Monthly average TDP concentrations for the three water regimes. 53
13b. Monthly average Mg concentrations for the three water regimes.. 53
14a. Monthly average K concentrations for the three water regimes... 57
14b. Monthly average K concentrations for standing and flowing
surface waters 57
15. Monthly average Na concentrations for the three water regimes.. 59
vi
LIST OF APPENDIX TABLES
Page
1. Precision and accuracy data for water analyses 74
2. Anova tables comparing surface, peat, and mineral waters,
considering water type and transect as factors 75
3. Anova tables comparing peat and mineral waters, considering
water type, season, and peat thickness as factors 76
4. Pearson product-moment correlation coefficients 77
5. Raw element data 78
vii
INTRODUCTION
The North Carolina Coastal Plain supports thousands of square miles
of peat-forming wetland ecosystems (Figure 1). Excluding the coastal
brackish marsh systems, most of these wetlands are in serious danger of
being irreversibly altered by man. Tens of thousands of acres have
already been ditched, drained, and cleared for agriculture and
silvaculture (Otte and Loftin, 1983). Only two large freshwater wetland
tracts (greater than 4047 hectares) remain that can be considered
relatively undisturbed: the large open pocosin and freshwater marshes of
southeastern Dare County, and the pocosins of the Croatan National Forest
in Carteret, Craven, and Jones Counties.
Most research on coastal wetlands has focused on man's influence on
the various aspects of these environments (Gambrell, Gilliam, and Weed,
1974; Kuenzler, Mulholland, Ruley, and Sniffen, 1977; Kirby-Sraith and
Barber, 1979; Gilliam and Skaggs, 1981). Unfortunately, these works
cannot be put into perspective until the natural, undisturbed systems are
understood. In the past 30 years, man has concentrated on developing the
pocosin wetlands - the large, shrub-dominated systems underlain with
varying thicknesses of peats and peaty mineral soils. The major concern
over the development of these pocosins, however, does not appear to be
the alteration of the land itself, but rather, the effects of water
draining from these wetlands into the surrounding streams and estuaries.
To better understand the influence of these waters as they leave the
pocosin, a need exists to determine the exact source of the water within
the pocosin
2
Figure 1 Environments of peat accumulation on the North Carolina
Coastal Plain. (1 mile = 1.6 kilometers)
3
OBJECTIVES
This project is designed to qualify and quantify the chemical
parameters of the source waters in the pocosins of the Croatan National
Forest. Preliminary field and lab investigations for this project
indicated that three different water regimes exist in the pocosins of the
Croatan National Forest: a) standing or flowing surface water, b) pore
water in the peat soils, and c) pore water in the underlying and adjacent
mineral soils. Because the physical conditions of each regime are
distinct, one should also expect that different chemical characteristics
exist for each regime. Standing or flowing surface waters should largely
reflect the addition of rainwater, in that the concentrations of elements
are appreciably less than those of peat pore water or mineral pore water.
Water associated with the peat should be enriched in elements that either
concentrate in the peat environment or elements that are released by the
decay of organics. Water associated with the underlying and adjacent
mineral soils should be less enriched in those ions directly associated
with organic matter and richer in ions that might be derived from the
minerals in the soils.
With the seasonal fluctuation of the water table (See Hydrology
section), a typical occurrence in a pocosin, it is logical for a
particular water regime to dominate or contribute more to drainage waters
in the appropriate season. Therefore, in addition to determining the
chemistry of the pocosin waters and documenting source regimes, this
project is designed to detect the change in outflow source(s) on a
seasonal basis
4
GEOGRAPHIC SETTING
The Croatan National Forest contains approximately 60,704 ha of land
(Schumann, 1977), 14,164 ha of which are underlain with peat (Ingram and
Otte, 1981). The Croatan is located in the Tidewater Region of the North
Carolina Coastal Plain and is situated on the Talbot Terrace. According
to Mixon and Pilkey (1976), the Talbot Terrace slopes gently
southeastward, is of Pleistocene age, and ranges in elevation from 7.6 to
12.2 m above mean sea level.
The Croatan is located approximately 23.9 km south of New Bern and
12.7 km west of Havelock in Carteret, Craven, and Jones Counties (Figure
2). Primary access to the forest is along Catfish Lake Road (SSR 1100,
1105), but several Forest Service Roads branch from Catfish Lake Road and
provide additional access (Figure 2).
Five lakes, the origin of which has not been fully documented, are
found in the Croatan (Catfish, Ellis, Great, Little, and Long), and all
are found along the margins of peat deposits (Figure 2). Tietz (1981)
studied the sedimentation and evolution of Great Lake and suggested that
this lake formed from the damming of Hunters Creek (method unknown) and
the subsequent flooding of its upper tributaries. Based on ^^C dating,
Tietz interpreted this to have occurred within the last 4000 years, after
and unrelated to the beginning of peat deposition in the Croatan pocosins
(estimated at 11,000 years before present). (See Regional Geology
section for information regarding peat deposition).
Peat is found in five areas in the Croatan National Forest: the
Northwest Catfish Lake deposit, the Northwest Great-Long Lake deposit,
the South Great Lake deposit, the South Ellis Lake deposit, and the
5
NEU BERIÍ
Isopach map of Croatan Foreac peaCa.
2 ft iaopach Interval.
A-Narthw«.c C.cfl.h Lake d.po.lt.
B'Northu.aC Cre.t-Long Lake dcpoalt.
C'South Great Lake depoait.
XHSouth Ellla Lake depoait.
E-Hlllia Road deposit.
Figure 2. Location map of the Croatan National Forest witli peat isopachs.
(2 feet = 0.6 meter and 1 mile = 1.6 kilometers)
6
Millis Road deposit (Figure 2). Surface elevations of the peatlands in
the Croatan range from 12.2 m in the northeast to 9.1 m in the southeast
(Ingram and Otte, 1981). The surface of the major deposit rises as a
broad dome (elevation slightly greater than 12 m above mean sea level)
just northwest of Great Lake (Ingram and Otte, 1981).
The study area for this project includes land west of Great and Long
Lakes and east and north of Catfish Lake and Forest Service Road #152
(SSR IlUO) (See Figure 2). This tract consists of approximately 7080 ha
of peatland, with peat thicknesses ranging up to 2.5 m. The peat thins
to the edge of the deposit and grades laterally into mineral soils
(Ingram and Otte, 1981). In this tract, the only major, man-made
disturbance is along its western edge, where approximately 51.0 km of
canals and canal maintenance roads cut through about 15.5 km2 of
pocosin. These canals, dug 30 years ago, are small and shallow and now
act as natural drainages rather than maintained, man-made drainages.
While working in the area, it was noted that an area of vegetation in
close proximity to the canals is usually slightly higher in stature than
what would normally be expected. Minimal change in vegetation, however,
occurs further away from the canals (roughly 10 ra). Thus, the wetland
system seems to have reached equilibrium with the canals, and wells were
emplaced approximately 30 m away from them.
7
PREVIOUS INVESTIGATIONS
Hydrological studies of coastal North Carolina have thus far
concentrated on various aspects of fluvial systems and their related
estuaries. Several approaches have been taken and may be broadly
classified into groups as follows: (a) those studies concerned with
nutrient levels in estuaries (Hobbie and Smith, 1975) and the impacts of
coastal plain streams upon estuaries (Brinson, Noltemeier, and Jones,
1980); (b) analyses of agricultural effects on drainage waters and
associated estuaries (Gambrell, et al., 1974; Kirby-Smith and Barber,
1979; Gilliam and Skaggs, 1981); and (c) studies dealing with the effects
of channelization (Kuenzler et al., 1977). In another hydrological study,
several North Carolina lakes were sampled and nutrient levels were
determined between 1971 and 1975 (Weiss and Kuenzler, 1976). However,
none of the five natural lakes of the Croatan National Forest was
sampled.^
These cited studies were concerned mostly with fluvial systems, not
peatlands. This study's approach to understanding nutrient levels in a
peatland is different in that the 1^ situ waters of a pocosin ecosystem
are observed, not just the drainage or outflow waters. This study
attempts to document that the i^ situ waters are chemically different in
response to differing physical parameters.
Peatland water studies have been performed in northern areas such as
Minnesota and Canada (Heinselman, 1970; Verry, 1975; Richardson, Tilton,
Kadlec, Chamie, and Wentz, 1978). The climatic and geological
differences between North Carolina and these areas, however, are
significant enough that comparison of data must be made cautiously.
8
Hydrological studies of North Carolina peatlands are, unfortunately, few
in number. Peat Methanol Associates have monitored the drainage canals
in a proposed peat mining area in Washington and Tyrell Counties, North
Carolina in an attempt to assess the effects of peat mining on drainage
waters. In addition, they monitored wells that collect water from peat
environments and from areas dominated by mineral sediments. The
unpublished well data are available as PMA documents (PMA, 1983) and may
be compared to the subsurface water data from this project.
A study was conducted in the Pungo National Wildlife Refuge between
the years 1977 and 1979 (Daniel, 1981) that is similar to this project
for the Croatan pocosins. The Daniel work included a comparative
analysis of peat pore waters and mineral pore waters between cultivated
and natural areas in the refuge. A difference in water chemistries was
observed in the study area and thus may serve as a reference for Ca, K,
Mg, and Na data obtained in this project.
During the years 1978 and 1979, Noltemeier conducted a water quality
study of Coastal Plain streams, in which peat-draining streams were
compared to mineral soil-draining streams. The peat-draining streams in
her study have their headwaters originating in the peat deposits of the
Croatan National Forest. A peat-draining stream in Noltemeier's study
refers to one in which a mixture of peat, peaty mineral soil, and mineral
soil is present in the drainage basin. Of the peat-draining streams
studied by Noltemeier, West Brice Creek was sampled in both her study and
for this project. Noltemeier's sampling site was located downstream from
the one in this study, however, and thus probably represents a greater
degree of mixing between peat, mineral, and surface waters. Comparison
of data is possible, however, and is of interest.
The Croatan National Forest has served as a research area for
9
various other projects, including vegetational analyses (Snyder,
1977,1978; Wilbur and Christensen, 1982) and a peat inventory conducted
for the U.S. Department of Energy (Ingram and Otte, 1981). The research
and conclusions of the Ingram and Otte study (1981) have been heavily
relied upon in this study as it is the only existing report of its kind
for the Croatan. The sedimentation of the Great Lake area of the
Croatan was studied in 1981 by Tietz. Water chemistry was not included
in Tietz's work, but stratigraphic and sedimentological data from the
cores of Great Lake have been helpful in this study.
10
CLIMATE
The North Carolina Coastal Plain experiences a humid, warm temperate
climate, with cool winters and warm summers. On the average. North
Carolina receives 127 cm of precipitation per year, with July being the
wettest month (19.5 cm) and October the driest (8.3 cm). A histogram of
average monthly precipitation data for the past 35 years at the Maysville
6 SW station is presented in Figure 3a. This is the closest weather
station to the Croatan. A histogram of the rainfall data, by month, for
this project's duration, June of 1983 through May of 1984 (Figure 3b), is
also presented. Generally, the summer of 1983 was drier than the
average, with the winter being somewhat wetter.
Monthly average temperatures for the Maysville station are presented
in Figure 4 and are compared to those for the sampling year. For the
long-term average, July and August are the warmest months, with average
temperatures of 25.4°C and 25.1°C, respectively. January and Febuary
are the coolest months, having average temperatures of 6.4°C and 7.3°C ,
respectively. Temperatures for the summer of 1983 were slightly higher
than the 31 year average.
The sampling year for this project can be characterized as having a
drier than average summer with higher temperatures, and a wetter than
average winter with temperatures that closely approximate the long-range
mean
11
o
cvj“
JJASONDJFMAM
Figure 3a. 35 year average monthly precipitation (cm) from
Maysville 6 SW station.
O
cvj“
JJASOND JFMAM
Figure 3b. Monthly average precipitation (cm) for Maysville 6 SW
station sampling year (6/83 - 5/84).
11
Figure 4a. 31 year average temperature (°C) for Maysville 6 SW station.
Figure 4b. Monthly average temperature (°C) for Maysville 6 SW station
during sample year (6/83 - 5/84).
13
REGIONAL GEOLOGY
Four major wetland geomorphological types can be found on the lower
North Carolina Coastal Plain: coastal marshes, floodplains, Carolina
bays, and blocked drainage systems (to which the Croatan pocosins belong)
(Otte, 1981).
The development of coastal marshes and sediment-filled floodplains
occurred in direct response to flooding associated with the geologically
14
recent rise in sea level. Available C dates indicate that the initial
flooding responsible for the development of these systems occurred 50Ü0
to 6000 years before present (Otte, 1981). Sea level, aç this time, was
approximately 5 m below its present level.
The cause of formation and development of Carolina Bays is not
known. The development of blocked drainage systems occurred between
10,000 and 12,000 years before present, based on ^^C dates of peats, as
well as matching pollen profiles with known floral succession associated
with glacial retreat (Otte, 1981). In Tietz's (1981) work on the
Holocene evolution of Great Lake, two peat samples from the Croatan
pocosin adjacent to the lake were age dated. ^'^C dates suggest that the
start of organic build-up in the Croatan was between 10,000 and 11,000
years before present. This age correlates with other dates reported for
the Dismal Swamp of North Carolina and Virginia (Whitehead, 1972) and
Hofmann Forest in North Carolina (Daniels, Gamble, Wheeler, and Holzhey,
1977a).
At 11,000 years before present, sea level was approximately 25 m
below its present level and roughly 38 km east of the present shoreline.
Sea level at this time was so far removed, with respect to elevation and
14
distance, from these blocked drainage systems as well as Carolina Bays
that Otte (1981) believed sea level unlikely to have been a major control
in the formation of these wetlands. Had this type of wetland development
been related to a rise in sea level, these systems should be more
widespread over the entire coastal plain.
Otte (1981) suggested a model for the development of blocked
drainage systems in which thicker peats fill a stream channel whereas
thinner peats spread over the adjacent interstream divides. Downstream
of peat-filled channels, areas of sandy sediment that mark the site of
blockage are found. Free-flowing, open-water streams flow downstream
from the blockages, indicating that some outflow still follows the
original drainage pattern. The origin of these blockages is, as yet,
unknown (Otte, 1981).
An impermeable, basal layer of clay which could account for a
perched water table is found in most of the blocked channels. Perhaps
the combination of basal clays with the blockage of drainage created a
ponded environment in which peat accumulation could begin (Otte, 1981).
In any case, the water-holding capacity of the peat probably allowed the
peat deposits to maintain themselves with or without the aid of a water
impermeable layer beneath the entire pocosin (Otte, 1981).
During 1979-1980, Ingram and Otte (1981) mapped the peat deposits of
the Croatan National Forest. Peat was found in five, physically
separated areas: Northwest Catfish Lake deposit. Northwest Great-Long
Lake deposit. South Great Lake deposit. South Ellis Lake deposit, and
Millis Road deposit (See Figure 2 for peat thicknesses).
Basically, two peat types occur in the Croatan: (1) brown,
decomposed fibrous peat, and (2) black, fine-grained, highly decomposed
humic peat. In the deeper parts of the filled stream channels, brown
15
fibrous peat can be found with thicknesses ranging up to 1.4 m. It is
usually absent from the top 0.6 to 1.2 m of peat (Ingram and 0tte,1981).
Black humic peat dominates most of the top 1.2 to 1.5 ra of peat,
overlying sands where the basal fibrous peat is absent (Ingram and Otte,
1981).
The Plio-Pleistocene stratigraphy underlying the peat deposits of
the Croatan region can be generalized as follows: (1) the Lower Pliocene
Yorktown Formation comprises the lowermost of these stratigraphic units.
The Yorktown has been described by Snyder, Mauger, and Akers (1983) as
being an unconsolidated, very muddy, quartz sand that thickens to the
northwest and east of New Bern (Daniels, Gamble, Wheeler, and Holzhey,
1972). (2) The Plio-Pleistocene Croatan Formation overlies the Yorktown
(Richards, 1950; Fallaw and Wheeler, 1969; and Daniels, Gamble, Wheeler,
and Holzhey, 1977b). The Croatan Formation consists of fossiliferous,
unconsolidated marine sands and silty sands and is equivalent to the
James City Formation of Dubar and Solliday (1963) and to the Small
Sequence of Daniels et al. (1972). (3) Upper Pleistocene sediments
overlie the Croatan Formation, with the contact marked by the Horry Clay,
a gray to olive-brown massive clay imbedded with cypress stumps and
lacking marine fossils (Tietz, 1981). Overlying the Horry Clay in the
Croatan region is the upper Pleistocene Planner Beach Formation, which
Teitz (1981) equates with the Talbot morphostratigraphic unit. This late
Pleistocene deposit is composed of unconsolidated sand, clayey sand, and
clay with some highly fossilized zones near the base (Tietz, 1981). The
lithology of the Planner Beach Formation is extremely variable laterally
as well as vertically.
For this project, nine cores along two east-west transects were
collected from the sediment layers adjacent to and underlying the Croatan
16
peat deposits (See Methods section for explanation and Figures 5 and 6
for cross-sections). In general, these sediments are very fine to fine
quartz sands, silty sands, sandy clays, and clays with minor amounts of
fine-grained opaques. X-ray diffraction analyses of some of the clay
samples indicate that the clays are primarily composed of kaolinite,
illite, vermiculite, and smectite. Vertical and lateral variation is so
widespread that correlation between cores is difficult. Beyond large
scale bedding features, a lack of recognition of sedimentary structures
may be due to sampling techniques, but Tietz also noted a lack of
structures in his study of the Great Lake area. These sediments are
Pleistocene in age and correspond to those of Teitz's Talbot
morphostratigraphic unit (= Planner Beach Formation), which represents
deposition in an estuarine environment during a marine transgression.
Figure 5. Soil profile prepared from mineral sediments collected along Transect A (Figure 7). Samples
were collected at 1/2 foot intervals. (1 foot = 0.3 meter)
Figure 6. Soil profile prepared from mineral sediments collectedmeatleorn)g Transect B
(Figure 7). Samples
were collected at 1/2 foot intervals. (1 foot = 0.3
19
HYDROLOGY
The term "pocosin" refers to a type of wetland system dominated by a
mixture of sparse trees and shrubs, organic soils, high water tables, and
poor drainage (Otte, 1981). An Algonquian Indian term, "pocosin" means
"swamp-on-a-hi 11" (looker, 1899). Originally, it referred to wetlands
believed to have developed on broad interstream flats that were
topographically higher than the surrounding terrain. This interpretation
has been modified based on recent work which indicates that the
interstream flats are really peat domes that developed in blocked stream
valleys. These peat domes grew both vertically and laterally until the
valley filled, after which the domes spread over the adjacent
interstream divides (Otte, 1981). As a result of this doming effect,
water flows out of the pocosin, rather than into it, making precipitation
the major source of water to the system.
Water tables fluctuate on a seasonal basis in pocosins. During the
winter and early spring, the water table is at its highest and is near,
at, or above ground surface. At this time standing surface water can be
found in areas of deep peat, grading outward to the edges of the pocosin
where the mineral-dominated soils are typically only water-saturated
(Otte and Loftin, 1983). As temperatures increase in the spring and
summer (Figure 4b), the vegetation begins to transpire and pumps a
significant quantity of water out of the system. During this time, as
the water table drops, surface waters disappear. Next, the water table
descends through the underlying sediments, draining the surficial peats
and peaty mineral sediments. Water tables typically drop 1.0 to 1.5 m
below ground surface during the summer and fall months (Otte, 1981). In
20
the thinner peats and peaty mineral soils, the water table may descend
through these soils and into the underlying mineral sediments. In the
thicker peats, the underlying mineral sediments are rarely unsaturated.
During late fall and winter, after évapotranspiration reaches a minimum,
influx of water from precipitation raises the water table back to its
spring-time high (Otte, 1981).
Most of the outflowing surface drainage during the spring-time high
is by sheet flow. Extensive Sphagnum mats and dense shrub growth baffle
water flow and aid in spreading this water evenly over the ground surface
(Otte, 1981). Such features as these and the lack of steep elevational
gradients hinder the development of channelized flow in a pocosin (Otte
and Loftin, 1983).
Sheet flow in the Croatan pocosins eventually converges and forms
the headwaters of several small creeks along the edges of the peat
deposits. These creeks radiate to the west into White Oak River, to the
east into Neuse River estuary, to the north into Trent River, and to the
south into Bogue Sound (Ingram and Otte, 1981).
Five blackwater lakes are located in the Croatan National Forest.
These include Catfish, Ellis, Great, Little, and Long Lakes, all of
which occupy shallow depressions with maximum depths ranging from 1.5 to
3.0 m. Of these lakes, the maximum diameter is approximately 4.0 km for
Great Lake, and the minimum diameter is roughly 1.2 km for Little Lake.
Drainage into and out of Great Lake is predominantly by groundwater flow,
the major influx coming from a northwest, topographically higher region
in the pocosin. Discharge is to the south, eventually forming the
headwaters of Hunters Creek en route to the White Oak River (Tietz,
1981)
21
In this study, Catfish Lake served as a lake sampling site. The
maximum depth is approximately 1.5 m, and the maximum and minimum
diameters are roughly 2.8 km east to west and 0.8 km north to south,
respectively. The mineral sediments around the periphery and bottom of
the lake are dominated by quartz sands, sandy clays, and clays. The
reworking of sediments by wave action around the shoreline has led to the
concentration of quartz sands at the lake's edge. In the deeper, quieter
portions of the lake, a thin layer (up to 30 cm) of organic-rich
sediments can be found overlying the sands and sandy clays. Several
small creeks flow south and southeast into Catfish Lake, whereas one
small outflow creek flows southeastward toward Great Lake (Ingram and
Otte, 1981).
Fire-break canals, dug about 30 years ago, can also be found in the
Croatan. Water flow in these canals fluctuates with the natural
fluctuation in the water table. The pocosin soils further away
(approximately 0.4 km) from the canals are slightly wetter than those
immediately adjacent to the canals (Table 1) (Otte and Loftin, 1983).
This difference is due to increased surface and subsurface drainage near
the canals, which allows aeration and oxidation of the peat in the
immediate area. Associated with the increased oxidation of soils
adjacent to canals is the increased release of nutrients. The vegetation
directly along the canals responds to better drainage by developing
denser and, in some cases, taller growth. This creates an area of
locally increased évapotranspiration and indirectly aids in subsurface
drainage
22
PEAT THICKNESS - IN FEET
1 2 3^ 5 6
<a
75.3 77.3 76.3 79.7 80.5 80.8 0-1 3 uo
70.6 77.5 81.5 82.3 83.5 1-2 «4-1 V
78.0 81.1 83.2 83.5 2-3 3 V(n 4M
79.5 81.4 82.6 3-4 Xu T3 c
81.6 81.1 4-5 cx C01 3
85.8 5-6 ?o O
QC
water content of peat within 100 feet of canals
PEAT THICKNESS - IN FEET
1 2 3 4 5 6
a>
82.0 81.6 85.6 84.7 84.7 85.8 0-1 :> o
79.9 82.7 83.6 84.5 85.2 1-2 4J
80.5 81.1 83.3 84.5 2-3 3
76.5 82.3 84.2 3-4 j: T3 c
82.0 84.0 4-5 a C
85.9 5-6 TJ o
CO
water content of peat at least 1/4 mile from canals
PEAT THICKNESS - IN FEET
1 2 3 4 5 6
V
6.7 4.3 9.3 5.0 4.2 5.0 0-1 g o
1.3 5.2 2.1 2.2 1.7 1-2
2.5 0.0 0.1 1.0 2-3 3 a
-3.0 0.9 1.6 3-4 X c
0.4 2.9 4-5 o. c
« 3 1
0.1 5-6 •o O
CO
difference between water content of peats 1/4 mile irom
canals and those 100 feet from canals
Table 1. Comparison of water content of peat soils near drainage canals
and away from drainage canals, in the Croatan National Forest
of North Carolina. Numbers refer to % water content of total
peat sample - as collected in the field. Data based on analyses
of 440 samples. (1 foot =0.3 meter)
23
VEGETATION
Pocosins are tertiary mire systems, dominated by differing
concentrations of evergreen, semi-evergreen, and deciduous shrubs, as
well as pond pine (Pinus serótina) (Otte and Loftin, 1983). Three major
environmental controls have direct influence over the distribution of
vegetation in a pocosin: peat thickness, hydroperiod, and fire (Otte,
1981). In peat less than about 1.2 m in thickness, the water table can
drop below the mineral/peat contact during the dry season; and roots can
reach the more nutrient-rich underlying mineral sediments. In peat
greater than 1.2 m in thickness, the water table rarely drops to the
mineral/peat contact. The vegetation associated with these deeper peats
is more likely to be limited in growth by low nutrients. As a result,
the plants are more stunted.
In a peatland, most of the nutrients fixed in plants are removed
from circulation and stored in accumulating peat (Smith, 1980).
Decomposition is so slow that these nutrients are only slowly recycled.
Fire can be a very important controlling factor in determining plant
growth in the pocosins, as it can rapidly recycle stored nutrients. The
vegetational succession pattern following a fire is short-lived, however,
in that the vegetation of a burned area soon returns to that of the pre-
burn system (Wilbur and Christensen, 1982).
Within a pocosin ecosystem, four vegetation types are recognized:
(l)Pond Pine Forest, (2) Pond Pine Woodland,(3) Tall Pocosin, and (4)
Short Pocosin (Otte, 1981). Within the Pond Pine Forest, plants are
rooted in soil that grades between peaty sand to about 0.6 m of sandy
peat. The shrub layer is closed and ranges from 3 to 6 m in height.
24
Pond pines typically form a closed canopy which is less than 15 m in
height. The subcanopy is composed of the bay species: loblolly bay
(Gordonia lasianthus), red bay (Persea borbonia), and sweet bay (Magnolia
virginiana), plus additional species such as red maple (Acer rubrum) and
loblolly pine (Pinus taeda) (Otte, 1981). Site 2A of this project is
located in a Pond Pine Forest. (See Figure 7 for location of sample
sites.)
The trees of a Pond Pine Woodland may also be rooted in mineral
soil, depending on the depth of the peat which ranges from 0.3 to 0.6 m
in thickness. Shrub roots can reach the underlying mineral soil, and the
shrub layer is usually closed, ranging from 1.8 to 4.6 m in height. The
bay species are more restricted to this shrub layer, whereas in the Pond
Pine Forest, they compose the subcanopy. The pond pines are taller and
larger than those of the Pond Pine Forest. The canopy is typically less
than 50% closed (Otte, 1981). Sites 4A and 2B can be categorized as
Pond Pine Woodland (Figure 7).
Tall Pocosins are underlain by 0.6 to 1.2 m of peat thus plant roots
typically can not reach the underlying mineral soils. With less nutrient
availability, the pond pines and bay species are widely scattered, and
the pines are gnarled at the tops due to past fires. The shrubs are 1.2
to 2.4 m in height and form a closed layer except in old burn sites
(Otte, 1981). Sites 6A and 4B are located in a Tall Pocosin (Figure 7).
Within a Short Pocosin, peat thickness is usually greater than 1.2
m; thus, plant roots can rarely reach the underlying mineral sediment.
The shrubs range from 0.6 to 1.2 m in height and are usually rooted in
the top 0.3 m of peat. Pond pines are widely scattered and stunted
(approximately 3 m in height), implying low nutrient availability (Otte,
1981). In this study, sites 8A and 6B are typical areas that may be
25
Figure 7. Location of study area and sampling sites, in association with
the natural and man-made drainage systems in this section of
the Croatan National Forest. (2 feet = 0.6 meter and 1 mile
= 1.6 kilometers)
26
classified as a Short Pocosin (Figure 7).
Sites OA and OB (Figure 7) of this project are representative areas
where peat is absent and mineral soils dominate. The vegetation
community at site OA is a bay forest, which contains a mixture of pond
pine, Atlantic White cedar (Chamaecyparis thyoides), sweetgum
(Liquidambar styraciflua), and the three bay species. The soil at site OA
is dominated by clays and clayey sands. Site OB, dominated by sandy
soil, is located in a pine forest in which loblolly pine (Pinus taeda) is
the dominant canopy species.
During the development of a pocosin ecosystem, sites of peat
accumulation expand laterally as vertical deposition occurs.
Associated with this expansion, the four pocosin types shift with the
changing patterns of peat thickness, hydroperiod, and nutrient
availability. The Pond Pine Forest and Pond Pine Woodland, which occupy
the outer areas of the pocosin, are transient. They occupy an area only
as long as conditions are proper for their existence. The Tall Pocosin
can possibly exist for a longer period of time in one area, but it too is
categorized with the Pond Pine Forest and Pond Pine Woodland. The Short
Pocosin is the only true climax community of pocosin types. Once peat
accumulation is of sufficient thickness that plant roots no longer reach
the underlying mineral soils, the Short Pocosin will remain a permanent
community, if not subjected to a major fire (Otte, 1981).
27
PROCEDURES
Field Methods
Seventeen locations were established to collect water samples,
including West Brice Creek, five canal sites, two sites on Catfish Lake,
and nine well sites (Figure 7). At each well site, two wells were
emplaced, one to collect peat pore water and a second to collect pore
water from the underlying mineral soils. The well sites range from areas
of mineral soils to areas of deep peat up to 2.4 m in thickness and were
positioned to form two parallel transects, here named transects A and B.
Each site (along both transect A and transect B) was situated wherever a
0.6 m increase in peat thickness was observed (Figure 7). Standing
surface water was collected next to the well sites located in the deeper
peats (1.8 to 2.4 m in thickness). Canal sites are located adjacent to
each well site along both transects. A soil profile for each well site
was constructed (from samples collected with a soil auger) in
anticipation of relating change in water chemistry to change in soil
characteristics (Figures 5 and 6).
Data were collected for one year (June 1983 to May 1984). Well
water was collected during the summer months of June, July, and August,
1983 to establish the chemical character of the mineral and peat pore
waters. Wells were again sampled in December of 1983 and January of
1984 to determine if any measurable changes in chemical parameters had
occurred with change in season. Standing surface water at each site was
collected when available. No surface water was present during mid-summer
to late fall due to the seasonal drop in water table below ground
surface. The canals, lake, and stream were monitored every month to
28
identify any mixing of regimes and to recognize any flux of dissolved
ions that might occur on a seasonal basis.
All water samples were collected and stored in 250 ml Nalgene
bottles that were acid washed in the lab and rinsed with canal water in
the field. Three 250-ml bottles were collected for each sample. Samples
were kept on ice until they were returned to the lab, where one 250-ml
bottle of each sample was centrifuged for 30 min at 4080 g and then
filtered, using 0.45 urn filter paper on a Millipore filtering system.
The filtered portion and the two unfiltered portions for each sample were
kept at 40c until lab analyses were performed. Originally, pH
measurements were to be taken in the field; however, due to the
questionable accuracy of the field pH meter, laboratory pH readings were
made instead, within 24 h of collection time.
Laboratory Methods
In order to identify the water chemistry of each regime, a variety
of wet chemical and instrumental methods of analyses were conducted for
the determination of pH and the quantification of the following dissolved
ions: Ag, Al, As, B, Ca, Cd, Co, Cr, Cu, Fe, K, Li, Mg, Mn, Mo, Na, Ni,
NH^-N, NO3-N, Pb, Se, Si02“Si, total dissolved PO^-P (TOP), total
organic C (TOC), Ti, V, and Zn.
The analyses for NH^-N (Solorzano, 1969), NO3-N (Jones, 1984), SÍO2-
Si (Stainton, Capel, and Armstrong, 1974), and TOP (EPA, 1976) were
performed on filtered samples in the Central Environmental Laboratory of
the East Carolina University Biology Department, utilizing a Bausch and
Lomb Spectronic 88 spectrophotometer. Precision and accuracy data are
presented in Table 1 in the Appendix. Approximately 125 ml of filtered
sample were left after the analyses for these ions. This remaining
29
amount was then acidified with HNO^ (to facilitate solution of cations
and serve as a preservative) and stored at 4oC until analyses could be
conducted for Ca, K, and Mg. Prior to cation analyses, each preserved
sample was split into two aliquots. To suppress ionization
interferences, lanthanum oxide was added to the samples used for Ca and
Mg analyses, and lithium sulfate was added to those used for K analyses.
Quantification of Ca, K, and Mg concentrations was performed in the
geochemistry lab of the ECU Geology Department, using an atomic
absorption spectrophotometer (Instrumentation Laboratory 457).
Total organic carbon was determined, using unfiltered samples, on an
Oceanography International organic carbon analyzer in the Institute for
Coastal and Marine Resources. Testing for Ag, Al, As, B, Ca, Cd, Co, Cr,
Cu, Fe, K, Li, Mg, Mn, Mo, Na, Ni, P, Pb, Se, Si, Ti, V, and Zn was
performed on approximately 100 samples with an inductively coupled argon
plasma emission spectrometer (ICAPES) in the Elemental Analytical Lab of
the Shared Research Resources Laboratories, ECU School of Medicine.
Samples for ICAPES analysis were centrifuged for 30 min at 4080 g and
filtered through a 0.45 urn filter paper on a Millipore filtering system.
For each sample, 200 ml were then preserved with 15 ml of double-
distilled HNO-j , and 150 ml were concentrated to 5 ml. Of the 24
elements analyzed, only Al, Ca, Mg, Mn, Na, Ti, and V were found in
detectable amounts. This limited number of detectable elements is due
partly to contamination problems in the labs and partly to the naturally
very low concentrations in the samples.
Statistical Methods
To test for significant chemical differences between the three water
regimes (mineral, peat, and surface) and to determine the effect, if any.
30
of transect location, seasonal variation, and peat thickness, several
randomized, incomplete, block analyses of variance were computed (SAS) :
one for each element detected. Fisher's F-test was used for the
comparison of the three water regimes, with transect, season, and peat
thickness also considered as factors. Sidak t-tests were then applied
for pairwise comparisons between the three water regimes to find which
pairs of regimes differed significantly. The Scheffe method for
comparisons was utilized to contrast well water types (mineral and peat),
seasons, and peat thicknesses. When peat thickness was determined a
varying factor, further pairwise comparisons were made to indicate which
sites varied significantly. To determine the relationship between
monthly canal data and water regimes, correlations were calculated
between respective monthly averages by using the Pearson product-moment
correlation coefficient.
Statistical analyses were performed on the following detected
elements; Ca, K, Mg, NH¿^-N, NO^-N, TDP, Si02~Si, and TOC. These
analyses can be found in Tables 2, 3, and 4 in the Appendix. However,
because of the seasonal sampling of well waters and the lack of surface
waters at certain times of the year, the interpretation of the
statistical analyses must be approached with caution. The test
statistics should only be considered as "approximations" due to the
considerable imbalance and number of missing observations. For the other
detected elements, Al, Mn, Na, Ti, and V, no statistical methods were
applied due to an even greater amount of missing data. In the cases of
Al, Mn, Ti, and V, mineral, stream, and/or lake waters were the only
water regimes that contained detectable concentrations. Sodium (Na) was
found in all three regimes, but due to equipment problems in the lab, Na
data was missing for all but three months of the sampling year.
Therefore, no statistical methods were attempted for this element.
31
RESULTS AND DISCUSSION
Because of the unique nature of an ombrotrophic system, special
processes control the elemental concentrations in the ecosystem's waters.
The major input of nutrients to the system is via precipitation and the
obvious export is by surface runoff. During the interval between
arriving and exiting the system, constituents in the water are subjected
to several possible elemental alterations. The loss of nutrient ions
from rainwater by plant uptake often results in a long-term storage of
elements since organic decay is very slow. However, during times of low
water table, the rate of organic decay can increase, releasing nutrient
ions and affecting the composition of drainage waters. The weathering of
silicate minerals, which predominate in the sandy clays and sands
adjacent to and underlying the peat deposits, is another source of ions
for the waters of this system. Ion exchange, whether by the colloidal
action of the underlying clays, the peat, or by the surface vegetation
(specifically Sphagnum), is another process which affects the pocosin's
waters. In addition, the precipitation and dissolution of ions due to
changes in concentration, pH, and redox conditions will affect the waters
of an ecosystem. It is the interaction of all these processes, and
perhaps of some not presented here, that affects and controls the
composition of the hydrosystem. With respect to a particular element or
ion, some specific processes play greater roles than others; and when it
was possible to determine these factors they are described.
The data presented in this section denote ionic and elemental
concentrations expressed in mg/1. Yearly average concentrations for each
of the three regimes (mineral, peat, and standing surface) and monthly
32
averages for flowing surface waters are discussed. The raw data
collected from each site and for each element may be found in the
Appendix, as can the results of statistical analyses computed for the
appropriate elements. Table 2 presents a summary of the total number of
samples collected for each month from each regime. When possible,
comparisons between average concentrations determined here and those
reported by Verry (1975), Daniel (1981), Peat Methanol Associates (1983),
and Noltemeier (1984) are made (Table 3). The reader will note that
Peat Methanol Associates' (PMA) data are consistently higher in value
than those reported in the other studies. This is possibly because some
of the PMA wells were located in areas that had been fertilized. While
comparisons with the PMA data may not be entirely applicable, some of the
trends may be real and the comparisons are of interest.
The ions discussed under "Indicator Species" represent those in
which an apparent difference in concentration between regimes occurs. In
each case, the differences are supported by statistical analyses. The
indicator species for specific regimes are (1) total organic carbon (an
indicator for peat pore waters), (2) dissolved silica (an indicator for
mineral pore waters), and (3) dissolved calcium (an indicator for surface
waters). Those ions and elements classified as "Non-indicator Species"
have been so classified for two reasons. For the inorganic forms of N,
total dissolved PO^^-P, Mg, and K, there is neither an apparent nor a
statistically significant difference between the average concentrations
in the various regimes. For Na and the trace metals A1, Mn, Ti, and V,
there are insufficient data on which to perform valid statistical
analyses. For ease of discussion, the results of the major cations Mg,
K, and Na are presented together as are the results for the trace metals
A1, Mn, Ti, and V
Table 2. Number of water samples from pocosin environments during the twelve month period
June 1983 through May 1984.
ENVIRONMENT J J A S 0 N D J F M A M TOTAL
Mineral 8 9 9 - - - 9 9 - - - - 44
Peat 5 4 4 - - - 6 6 - - - - 25
Standing
Surface 6 - - - - - 4 4 6 6 6 4 36
Canal 5 5 5 4 - 5 5 5 5 5 5 5 54
Lake 2 2 2 2 2 2 2 2 2 2 2 2 24
Stream 1 1 1 1 1 1 1 1 1 1 1 1 12
TOTAL 27 21 21 7 3 8 27 27 14 14 14 12 195
U)
U)
Table 3. Comparison of water chemistry in surface, peat, and mineral waters of Croatan National Forest,
peat and mineral waters of Pungo National Wildlife Refuge (Daniel, 1981), peat and mineral
waters in Peat Methanol Associates area (PMA, 1983), Minnesota perched bog streams (Verry, 1975),
and West Brice Creek (Noltemeier, 1984). Averages of concentrations are expressed in mg/1.
SITES pH TOC SÍ02-SÍ Ca N03-N NH4-N TDP04-P Mg K Na Mn
This study
mineral ave. 3.8 42.7 3.55 1.51 0.03 0.15 0.03 0.61 0.64 3.72 0.00
peat ave. 3.8 212.5 0.89 2.09 0.12 0.34 0.06 0.76 0.43 2.99 0.00
surface ave. 3.7 51.0 0.45 0.20 0.02 0.02 0.02 0.51 0.12 2.14 0.00
West Brice Cr. 3.8 65.7 1.47 0.36 0.02 0.01 0.01 0.54 0.22 2.75 0.00
canal ave. 3.7 52.2 0.48 0.27 0.04 0.03 0.01 0.49 0.20 2.44 0.00
Daniel, 1981
mineral ave. 8.00 2.43 0.78 5.51
peat ave. 0.66 0.43 0.00 4.00
PMA. 1983
mineral ave. 6.5 41.8 34.6 49.1 0.02 0.15 8.93 1.19 10.89 0.27
peat ave. 5.1 233.5 10.4 24.2 0.02 0.39 10.95 7.68 13.25 0.23
Noltemeier, ' 1984
West Brice Cr. 4.4 2.14 0.12 0.03 0.02 0.47 0.27 2.78
Verry, 1975
perched bog
stream ave. 3.6 2.40 0.20 0.45 0.97 1.30 0.60
LO
35
pH
The pH data collected during this study were determined with a
laboratory meter nearly 24 hours after sampling. It should be mentioned
that although pH can be measured accurately in the lab, the obtained
results depend upon sampling conditions which can not always be
controlled, and that the pH of water samples may change during storage.
The carbon dioxide-bicarbonate-carbonate equilibrium and buffer system is
most easily disturbed by loss of carbon dioxide (CO2). It is impossible
to obtain a truly accurate lab pH from groundwater samples because some
CO2 will be lost before it is even possible to get a sample (Hem, 1959).
The CO2 content is lowered as temperature rises (Rainwater and Thatcher,
1960). In addition, the predominance of oxidized or reduced forms of
some ions depends upon the environment of the water. With the contact of
air, a reduced form may change to an oxidized state and pH can change as
a result (Hem, 1959).
These uncertainties involved in using the laboratory pH
determinations have caused some investigators to favor field
determinations made situ before some of the possible changes can
occur. However, laboratory data also have a place of importance (Hem,
1959).
Because of the questionable reliability of the field pH meter
available for this study, a laboratory pH meter was used and pH data was
obtained within 24 hours of collection time. Samples were kept in the
dark and refrigerated until time of analysis. Although these precautions
were taken, the lab pH data of this study is not used as an indicator for
any particular water regime and is only presented for completeness.
36
For all of the waters sampled, pH varied only slightly. Mineral and
peat pore waters both had an average pH of 3.8; while both standing
surface and flowing canal waters had an average pH of 3.7. Catfish Lake
had an average pH of 3.9, and West Brice Creek had an average pH of 3.8.
Noltemeier (1984) reported an average pH of 4.1 for precipitation
collected in the Croatan National Forest, and an average pH of 4.4 for
her sampling site on West Brice Creek. Noltemeier's sampling site on the
creek was downstream from the one in this study. Her pH value may have
been higher because of a greater input of mineral soil waters. Verry
(1975) found an average pH of 3.6, a comparable value to that determined
here, for a stream in Minnesota whose headwaters originate in a perched
bog. Peat Methanol Associates (1983) reported a pH average of 6.5 for
wells in their fine sand unit and an average pH of 5.1 for wells in their
peat unit. Both values are much higher than those determined for the
well waters of this study.
In most natural systems exposed to the atmosphere, pH is controlled
by reactions involving the carbonate system (carbonic acid and solvated
carbon dioxide) (Drever, 1982). In addition, several other factors
influence the pH of pocosin soils. Acidic substances are produced by
vegetation as by-products of oxidative metabolism (Moore and Bellamy,
1974). Living plants release carboxyl-associated hydrogen (H+) ions,
which are exchanged for nutrient cations contributed by rainfall or
groundwater. Thus, the acidity of soil waters increases (Wetzel, 1983).
In addition, organic acids (humic, fulvic, and tannic) are released with
decomposition (Moore and Bellamy, 1974). These compounds accumulate in
the system with the continued accretion of peat (Moore and Bellamy,
1974). Organic acids can be leached from peat as water exits the system
by runoff or base flow.
37
Noltemeier (1984) also noted that the low pH of surface waters In
the Croatan may be influenced by acid rainfall (ave = 4.1). Waters with
low dissolved solids, such as rain, tend to become increasingly acidic in
contact with peat (Gosselink and Turner, 1978). Therefore, rainfall may
be a contributing factor, along with the influence of vegetation, to the
acidity of pocosin waters.
Indicator Species
Total Organic Carbon
Total organic carbon (TOC) associated with peat pore waters (ave=
212.5 mg/1) was higher than that in the other environments (Figure 12a).
Standing surface waters and mineral pore waters contained average
concentrations of 51.0 and 42.7 mg/1, respectively. Catfish Lake, West
Brice Creek, and the canals had average concentrations of 42.9, 65.7,
and 52.2 mg/1 TOC for the sampled year.
The histogram (Figure 8a) comparing monthly average TOC
concentrations for each regime, supported by statistical analyses (Tables
2 and 3 in the Appendix), dramatizes the differences between water
types. Because of the agreement between statistical data and the obvious
differences between average concentrations for each regime, high
concentrations of TOC are interpreted to be an indicator of peat waters.
Data also indicate that, with increasing peat thickness, there is an
increase in TOC concentration of peat waters (Figure 8b). Overall, an
increase does occur. The increase in TOC may reflect a reduced flushing
rate, and thus greater accumulation in pore water with increasing
thickness (Noltemeier, 1984).
Figure 8a. Monthly average TOC concentrations (mg/1) for the three
water regimes. See the Appendix for explanation of missing
data.
t =? mineral peat standing surface
j" + standard deviation of the mean
Figure 8b. Average TOC concentrations (mg/1) for peat waters compared to
peat thicknesses (feet). (2 feet = 0.6 meter)
38
(TmOgC/I)
I213Q0ü5O0
T T I
0 2 e 10
Peat Thickness (ft.)
39
Positive correlations (ranging from .921 to 1.000) for TOC occurred
between peat and canal waters during the summer months of June to August
through December (Table 4 in the Appendix). By March, a negative
correlation (-.962) existed between standing surface and canal waters.
West Brice Creek also followed this trend of increasing TOC during the
dry summer and fall months as peat-water drainage increased. TOC
concentration of the creek declined in the winter months as increased
surface drainage diluted the creek waters. This trend is consistent with
Noltemeier's (1984) findings of highest TOC concentrations for peat-
draining streams occurring during low flows.
The TOC data of Catfish Lake also showed an increase in
concentration with the dry fall months. According to Gorham (1967),
organics and associated carbon may be concentrated with the lowering of
water level. As water level rises during the winter and spring months,
TOC concentration declines. (See Figure 9 for monthly trends of standing
and flowing surface waters.)
The high TOC concentrations of peat pore waters (ave = 212.5 mg/1)
are consistent with those reported by Peat Methanol Associates (1983)
(ave “ 233.5 mg/1) for peat waters. The pore water of the peat has
longer contact time with the carbon-rich organics than does any other
water regime. Therefore, the peat pore waters would be expected to have
greater concentrations of TOC. The underlying mineral pore waters
contained an average TOC concentration of 42.7 mg/1, which is also
similar to the 41.8 mg/1 average found by Peat Methanol Associates (1983)
for their fine sand unit. It is possible that carbon from the overlying
peat unit migrates downward (via both diffusion and mass transport) into
the mineral layers, thus accounting for the TOC concentrations of the
mineral sediments. These mineral sediments could contain some original
40
T(mOgC/I) 11205000
Figure 9. Monthly average TOC concentrations (mg/1) for standing and
flowing surface waters.
stream water canal water
— lake water standing surface water
41
colloidal organics, however, which would mean that the organic carbon is
"authigenic" to the mineral sediments.
Standing surface waters contained an average of 51.0 mg/1 TOC,
possibly the result of minimal contact time between these waters and
peat. With the lowering of the water table in summer and fall, flowing
surface waters became more concentrated in organic carbon, as can be
seen in Figure 9. With the rise of the water table during the winter and
spring months, TOC concentration declined from the summer high and
fluctuated very little.
Silica
Dissolved silica (Si) was most concentrated in the mineral pore
waters (ave = 3.55 mg/1). Peat pore waters contained a yearly average of
0.89 mg/1 Si, approximately three-fourths less than that of the
underlying and adjacent mineral soils. Catfish Lake, standing surface
waters, and water draining by way of the canals contained the least
amount of Si, averaging yearly concentrations of 0.41, 0.45, and 0.48
mg/1, res^jectively. West Brice Creek averaged 1.47 mg/1 Si, but ranged
in Si content from 0.61 to 2.50 mg/1 during the sampling year.
Noltemeier (1984) did not determine Si concentrations in her study, so no
comparisons are possible for stream data.
A histogram of monthly average Si concentrations for each regime
(mineral, peat, and surface waters) exemplifies the contrast in
concentrations (Figure 10a), and is supported by statistical analyses
(Tables 2 and 3 in the Appendix). Therefore, Si is considered to be an
indicator element for mineral pore waters.
Figure 10a. Monthly average Si concentrations (mg/1) for the three water
regimes. See the Appendix for explanation of missing data.
mineral peat = standing surface
y + standard deviation of the mean
Figure 10b. Monthly average Si concentrations (mg/1) for standing and
flowing surface waters.
stream water — canal water
lake water standing surface water
42
S(rngi/l)
213456.,.0.000
43
Positive relationships, indicated by correlation coefficients of
1.00, 1.00, and .885, between peat and canal waters occurred during the
months of July, August, and December (Table 4 in the Appendix). For
January, a positive correlation between both peat and mineral waters with
the canal waters occurred (.889, .724, respectively). During February
and March, positive correlations existed between surface and canal waters
(.984 and .958, respectively). This information is interpreted to mean,
and supports the hypothesis, that peat water is predominant in the
drainage waters during times of low water table. With the rise of the
water table, drainage from the areas of no peat that are dominated by
mineral soil increases. This mixing is represented by both peat and
waters being positively correlated with canal waters. Again, these
statistical interpretations support the observed behavoir of the rise and
fall of the water table, draining the respective environments. As the
water table rises above ground surface, surface waters dominate the
drainage waters, indicated by the dilution of Si in the canals. This is
supported by the positive correlation between surface and canal waters
for February and March. (See Figure 10b for monthly trends of canal and
surface waters and Table 4 in the Appendix for correlation data.)
The Si content in West Brice Creek fluctuated in much the same way
as in the canals. Si concentration increased during the dry summer and
fall months, followed by a decline in the winter months (See Figure 10b).
West Brice Creek flows through an area of mineral-dominated soils
upstream from the sampling site, thus accounting for the high
concentrations in Si of the summer and fall. With increased surface
runoff during the winter and spring. Si concentrations in the creek were
diluted. There was a sharp decrease in Si concentration in December
44
(Figure 10b) between the summer /fall high and the winter/spring decline,
which is unexplainable at this time.
Catfish Lake is directly underlain by quartz sands and sandy clays.
With the decline of water level during the summer and fall, one would
expect an increase in Si content due to a concentration of Si ions
already present (Gorham, 1967). However, Si concentration took a sharp
decline during the fall months. One possible explanation for this
decline could be uptake by a diatom bloom (Schelske, Stoermer, Conley,
Robbins, and Glover, 1983). (See Figure 10b for monthly trends of lake
water.) However, chlorophyll concentrations and other necessary data
were not obtained and thus, insufficient supportive evidence is available
to document a bloom.
The Si found in all the systems is derived from chemical weathering
of silicate minerals (Stumm and Morgan, 1981), which dominate the mineral
component of the non-peat soils and the non-peat fraction of the peat
soils. Peat Methanol Associates (1983) found Si concentrations in the
peat and in the fine sand units to be 10.4 mg/1 and 34.6 mg/1,
respectively. Although their Si values are much higher than those
reported here, their data exhibit the notable difference in Si content
between mineral and peat waters.
Calcium
Standing and flowing surface waters contained the least amounts of
dissolved calcium (Ca). West Brice Creek, Catfish Lake, and the canals
had yearly averages of 0.36 mg/1, 0.66 mg/1, and 0.27 mg/1 Ca, whereas
standing surface waters averaged 0.20 mg/1 Ca. Peat pore waters
45
contained the highest levels of Ca, averaging 2.09 mg/l, and mineral pore
waters the second highest levels, averaging 1.51 mg/l dissolved Ca.
A histogram of monthly average Ca concentrations (Figure 11a) for
each regime (mineral, peat, and surface waters) contrasts the difference
in Ca content between standing surface waters and both types of well
waters. The significance of this difference is confirmed by statistical
analyses (Tables 2 and 3 in the Appendix) which compare the three water
regimes with respect to transect. Low concentrations of Ca therefore,
are considered an indicator for surface waters. Positive correlation
coefficients (ranging from 1.00 to .742) between surface waters and
drainage canal waters for December through May support the observation
that surface waters were draining from the pocosin during the time when
the water table was at its highest (Table 4 in the Appendix).
This seasonality in Ca content for outflowing waters is shown in
Figure 11b. Both the creek and canals increased in Ca concentration
during the driest months of July to September, reflecting the
concentration of Ca ions with the lowering of water level in the drainage
channels (Gorham, 1967). With Increased surface runoff during the winter
and spring months, Ca content of the canals and creek declined. This is
suggested by the positive correlation between surface and canal data.
Noltemeier (1984) reported a higher average concentration of 2.14 mg/l Ca
for West Brice Creek, compared to the 0.36 rag/1 average reported here.
However, in general her peat draining streams exhibited higher Ca
concentrations during periods of low flow, the same trend observed here.
Catfish Lake exhibited a similar trend in which Ca concentration
increased during the driest months with the lowering of water level and
subsequent concentration of Ca ions.
Figure lia. Monthly average Ca concentrations (mg/1) for the three water
regimes. See the Appendix for explanation of missing data.
mineral peat = standing surface
T + standard deviation of the mean
Figure 11b. Monthly average Ca concentrations (mg/1) for standing and
flowing surface waters.
stream water canal water
--lake water standing surface water
46
C(mga/I) 53421...0000
47
Since precipitation is the major ionic contributor to surface
waters, these waters are naturally low in ionic concentrations.
Noltemeier (1984) reported an average Ca concentration of 1.26 mg/1 for
precipitation in the Croatan. Standing surface waters collected in this
study from the low pocosin areas contained the lowest level of Ca (ave =
0.2Ü mg/1) compared to other waters in the system. In these low pocosin
areas, extensive Sphagnum mats are found. Sphagnum mats behave as cation
exchangers because of their high concentration of polymerized
unesterlfied uronic acids (Wetzel, 1983). Cations (Ca and Mg) are
absorbed by the moss, resulting in the release of hydrogen (H+) ions and
organic acids (Clymo, 1963). The Ca ions in surface waters not adsorbed
by Sphagnum serve as plant nutrients for the pocosin vegetation
(Richardson et al, 1978) or are lost in runoff (Hemond, 1980). Thus,
surface waters contain low levels of Ca due to a combination of factors.
Daniel (1981) reported Ca concentrations for organic soil waters and
underlying mineral soil waters of 0.66 mg/1 and 8.00 mg/1, respectively.
Peat Methanol Associates (1983) reported even higher concentrations of
24.2 mg/1 and 49.1 mg/1 Ca for wells in the peat unit and fine sand unit.
The trend observed from both PMA and Daniel's data is that mineral waters
contain higher levels of dissolved Ca than do peat waters. This trend
contradicts the one detected here, where peat waters contain higher
levels of Ca than mineral waters. As of yet, I do not have a good
explanation for this contradiction. It may be that mineral sediments in
the PMA and Daniel studies contained higher concentrations of CaC03 shell
debris, in which case, mineral pore waters would probably contain higher
concentrations of Ca than peat pore waters. In explanation of the trend
observed here, it may be possible that some Ca, associated with the
organic peat, is slowly being released by decomposition, allowing the Ca
48
concentration for peat waters to be greater than that of mineral waters.
According to Damman (1978), an increase in Ca concentration resulting
from organic matter breakdown can result without any additions from other
sources.
Non-Indicator Species
Nitrogen
Nitrate. Standing surface waters and mineral pore waters contained
less dissolved nitrogen in the form of nitrate (NO^-N) (ave = 0.02 and
0.03 mg/1) than peat pore waters (ave = 0.12 mg/1). Catfish Lake , West
Brice Creek, and canal waters had yearly averages of 0.08 mg/1, 0.02
mg/1, and 0.04 mg/1, respectively.
A histogram (Figure 12a) contrasting monthly averages for each
regime (mineral, peat, and standing surface waters) illustrates that peat
waters appear to have much greater NO^-N concentrations than either
mineral or surface waters. Statistical analyses (Tables 2 and 3 in the
Appendix), however, do not support the apparent differences noted in
Figure 12a. There is a wide range in N02~N concentrations at any one
time for peat waters which affects the statistical analyses to a greater
degree than what the histogram depicts. The statistical data actually
indicate that there is a significant difference in NO^-N concentrations
between mineral and surface waters. However, since neither mineral nor
surface waters are significantly different (according to statistical
analyses) from peat waters, NO^-N can not be considered a good indicator
species for any specific water regime.
Figure 12a. Monthly average N03-N concentrations (mg/1) for the three
water regimes. See the Appendix for explanation of missing
data.
1 mineral peat = standing surface
J + standard deviation of the mean
Figure 12b. Monthly average NH4-N concentrations (mg/1) for the three
water regimes. See the Appendix for explanation of missing
data.
1 mineral peat = standing surface
"j" + standard deviation of the mean
49
(NmOg^N-H/N.l-N) J JASON D J FMAM(mg/I)
50
Standing surface waters were lowest in NO^-N concentrations. The
average NO3-N concentration for West Brice Creek, in this study (ave =
0.02 mg/1) is less than that found by Noltemeier (1984) (ave = 0.12
mg/1). Noltemeier's sampling site is a few miles downstream of the one
in this study and is influenced by a larger area and diversity of soils.
Noltemeier also used a different analytical method for NO3 determination,
which may also account for the difference in NO^ concentration.
Peat pore waters were highest in NO^-N concentrations of the three
regimes. With the lowering of the water table during the summer months,
allowing aeration of shallow peats, aerobic nitrifiera can convert
ammonium (NH¿^) to NO3 (van Cleemput, Patrick, and Mcllhenng, 1975).
Thus, peat waters range in NO3-N concentrations from relatively low (0.02
mg/1) levels in the deeper peats to higher (0.28 mg/1) levels in the
shallow, periodically aerated peats. This range in concentrations has
led to a higher overall average (ave = 0.12 mg/1) for NO3-N
concentrations in peat waters. Peat Methanol Associates (1983) reported
a lower average concentration of 0.02 mg/1.
Mineral pore waters were consistently lower in NO3-N concentration
than peat, pore waters. A continously anaerobic environment is reflected
by these low concentrations, as a significant conversion of NH^ to NO3
does not occur (van Cleemput et al., 1975). Peat Methanol Associates
(1983) reported an average NO3-N concentration of 0.02 mg/1 for wells in
the fine sand unit. This concentration is consistent with the NO3-N
average (0.03 mg/1) found in this study.
Ammonium Nitrogen. Except for the peat pore waters (ave = 0.34
mg/1), all the environments contained relatively low levels of NH^-N:
51
mineral pore waters 0.15 rag/1, standing surface waters 0.02 mg/1, canal
waters 0.03 mg/1, lake waters 0.02 mg/l, and stream water 0.01 mg/1.
The histogram of monthly NH^ averages for mineral, peat, and standing
surface waters (Figure 12b) exemplifies the higher NH^-N concentration of
peat waters. All statistical analyses (Tables 2 and 3 in the the
Appendix) indicate that no significant difference exists between water
regimes. The extreme variation in NH¿^-N concentrations at any one
sampling time for each regime affects the statistical analyses in such a
way that no significant differences occur. Because the statistical data
do not support the hypothesis of a NH^-N difference among the regimes,
NH¿^-N is not considered an indicator species for any water regime.
The concentration of NH^-N in flowing surface waters did not
fluctuate dramatically on a seasonal basis. An average NH¿^-N
concentration of 0.01 mg/l was determined for West Brice Creek in this
study, which is comparable to that found by Noltemeier (1984) (0.03 mg/l)
further downstream.
In comparison to mineral and peat pore waters, standing surface
waters were lowest in concentration. Anaerobic and aerobic microbes
are resppnsible for the production of inorganic NH^ from organic N
(Smith, 1980). Conversion of to NO3 by aerobic nitrifiers occurs
rapidly in an oxygenated surface environment (Klopatek, 1975). The NH^
not converted to NO3 is probably not an appreciable amount, and this may
be reflected in the low NH^ concentrations in surface waters.
The higher concentrations in peat pore waters reflect an
accumulation of ions produced in the conversion of organic N to
inorganic NH^ by anaerobic bacteria (Richardson et al., 1978). The
concentrations present in mineral pore waters (ave = 0.15 mg/l) are also
52
in response to the conversion of organic N to inorganic NH^. The lower
levels reflect the lack of abundant organic matter in the mineral
sediments as opposed to the abundance of organics composing peat. There
may be more NH^-N present in the mineral pore waters than is detected
because cation exchange sites of the clay minerals can be occupied by
NH¿j-N ions. PMA did not determine levels in their analyses, so no
comparisons are possible.
Total Dissolved Phosphorus
Total dissolved phosphorus (TDP) content was low, ranging from less
than 0.01 mg/1 to 0.18 mg/1, in all of the environments samples. Mineral
waters averaged 0.03 mg/1; peat waters averaged 0.06 mg/1; and surface
waters averaged 0.02 mg/1. Catfish Lake, West Brice Creek, and canal
waters all averaged 0.01 mg/1 TDP.
The histogram of monthly average TDP concentrations for each regime
(Figure 13a) exemplifies the difference between peat and surface water
concentrations. As can be seen in the figure, there is an overlap of
ranges among regimes, and the statistical data (Tables 2 and 3 in the
Appendix) indicate this. According to one set of analyses (Table 2 in
the Apendix), only mineral and surface waters are significantly
different. The results from a second statistical analysis (Table 3 in the
Appendix) indicate that only peat and mineral waters are significantly
different. Because of this lack of agreement between the two statistical
tests and because of the range overlap between regimes, TDP is not
considered a good indicator for any specific water regime.
Figure 13a. Monthly average TDP concentrations (mg/1) for the three water
regimes. See the Appendix for explanation of missing data.
mineral peat = standing surface
y + standard deviation of the mean
Figure 13b. Monthly average Mg concentrations (mg/1) for the three water
regimes. See the Appendix for explanation of missing data.
mineral peat = standing surface
+ standard deviation of the mean
53
(TmD(gMmP/Ig) /gI) 0O01..,0.J5j05O50
54
Correlation data (Table 4 in the Appendix) for drainage waters is
not very meaningful as the canals fluctuated between 0.00 (below
detection limits) and 0.03 mg/1 TOP. Creek waters varied only 0.01 rag/1
between months, averaging 0.01 mg/1. This is close to the 0.02 mg/1
average TOP content reported by Noltemeier (1984). Total dissolved
phosphorus in all drainage and surface waters is quite low, suggesting
that pocosins are very phosphorus poor systems.
Peat pore waters contained the highest amounts of TOP (ave = 0.06
mg/1). This is to be expected as TDP is associated with the slowly
decaying plant material. A large portion of TDP that is released by
decomposition may not be lost in runoff but possibly retained by the high
ion exchange capacity of the colloidal peat (Moore and Bellamy, 1974).
TDP may, therefore, either be bound within the organic structures of the
decaying peat, or retained superficially by ion exchange (Moore and
Bellamy, 1974). Noltemeier (1984) reported that for West Brice Creek,
approximately 50% of the TDP was reactive, meaning that it is in a form
available for plants, the other 50% probably being organically bound.
Most of the TDP that enters the system via rainwater is probably in the
reactive form. In any case, peat pore waters contain greater amounts of
TDP than do surface waters.
Mineral pore waters contained less TDP than did peat waters (ave =
0.03 mg/1). The TDP content of the mineral layers is possibly related to
the organic material associated with the fine grained sediments. The
colloidal nature of not only the organics, but of the clays which are
common in the mineral sediments, allows TDP to be easily bound within
this layer (Moore and Bellamy, 1974).
55
Peat Methanol Associates (1983) reported TDP concentrations for
wells in the fine sand unit and in the peat unit to be 0.15 mg/1 and 0.39
mg/1, respectively. Their data are notably higher in concentration than
that of this study. However, the average TDP content in the peat water
was greater than the average of mineral water, which is consistent with
the trend observed in this study.
Magnesium/Potassium/Sodium
The magnesium (Mg) concentration in all environments was similar.
The average for peat pore waters was 0.76 mg/1; for mineral pore waters
was 0.61 mg/1; and for standing surface waters was 0.51 mg/1. The
averages for Catfish Lake, West Brice Creek, and the canals were 0.56
mg/1, 0.5A mg/1, and 0.49 mg/1, respectively. The histogram of monthly
average Mg concentrations (Figure 13b), supported by statistical data
(Tables 2 and 3 in the Appendix), illustrates that Mg levels are not
significantly different between regimes. Therefore, Mg is not an
indicator element for any particular water regime.
Magnesium content for flowing surface waters did not vary
significantly on a seasonal basis. West Brice Creek and the canals
averaged 0.54 mg/1 and 0.49 mg/1, ranging from 0.31 to 0.80 mg/1.
Noltemeier (1984) reported a comparable average of 0.47 mg/1 Mg for the
stream. Peat Methanol Associates (1983) reported higher Mg values for
peat and sand waters (ave = 10.95 and 8.93 mg/1, respectively) than those
reported here (peat ave = 0.76 mg/1 and mineral ave = 0.61 mg/1).
Although the PMA averages are higher by an order of magnitude than those
56
determined during this study, a similar trend exists in which peat waters
contain slightly higher Mg concentrations than mineral waters. Daniel
(1981) observed the opposite trend, however, in which mineral waters had
higher average Mg concentrations (ave = 2.A3 rag/1) than peat waters (ave
= 0.A3 mg/1).
Mineral pore waters had higher average concentrations of dissolved
potassium (K) than did peat pore waters (ave = 0.64 mg/1 and 0.43 mg/1,
respectively). Standing and flowing surface waters contained lesser
amounts of dissolved K, with standing waters averaging 0.12 mg/1,
Catfish Lake averaging 0.30 mg/1, West Brice Creek averaging 0.22 mg/1,
and drainage canals averaging 0.20 mg/1.
Monthly K averages for each regime are presented in Figure 14a. As
is the case with the other "non-indicator" species, the overlapping of
ranges between regimes is so great that statistical differences can not
be demonstrated. Therefore, K can not be considered a good indicator for
any particular water regime. Statistical analyses (Tables 2 and 3 in the
Appendix) reflect the variation in K concentrations and, therefore,
support this conclusion.
Drainage waters and Catfish Lake exhibited a seasonal trend in K
concentration. The lake, canals, and stream all increased in K
concentration during the driest months of August to October (Figure 14b).
This trend was also observed by Noltemeier (1984) in her water
quality/flow study for peat draining streams, in which periods of low
flow were associated with an increase in K concentration. According to
Gorham (1967), with increased évapotranspiration and the subsequent
decline in water level during summer months, cation concentrations may
Increase. The average K concentrations of West Brice Creek and the
Figure 14a. Monthly average K concentrations (mg/1) for the three
water regimes. See the Appendix for explanation of
missing data.
f mineral peat = standing surface
y + standard deviation of the mean
Figure 14b. Monthly average K concentrations (mg/1) for standing
and flowing surface waters.
stream water — • canal water
— — lake water standing surface water
57
(Kmg/I)
01..527050
illlh
58
drainage canals (0.22 mg/1 and 0.20 mg/1) are similar to Che average K
concentration of 0.27 mg/1 reported by Noltemeier.
The trend of higher K concentrations in mineral waters, compared to
peat waters, as observed in this study, is similar to Daniel's (1981)
observation. Daniel reported slightly higher concentrations in mineral
waters (ave = 0.78 mg/1 as opposed to 0.6A mg/1 reported here), but did
not detect K in peat waters, whereas the average for peat waters in this
study was 0.43 mg/1 K. Peat Methanol Associates (1983) not only reported
significantly higher concentrations in both water regimes (ave = 1.19
mg/1 K in waters from their fine sand unit and 7.68 mg/1 k in waters from
their peat unit), but also noted the opposite trend, in which peat waters
contained higher K levels than mineral waters.
Sodium (Na) was present in all of the environments in comparable
concentrations. The average for mineral pore waters was 3.72 mg/1; for
peat pore waters was 2.99 mg/1; and for standing surface waters was 2.14
mg/1 Na. Catfish Lake, West Brice Creek, and the canals had averages of
2.78 mg/1, 2.75 mg/1, and 2.44 mg/1 Na, respectively. A histogram of
monthly average Na concentrations for each of the three regimes (mineral,
peat, and standing surface waters) is presented in Figure 15. Although
mineral waters contained higher levels of dissolved Na, the difference in
concentrations between regimes does not appear to be significant. The
lack of sufficient well water data prohibits the use of statistical
analyses in order to test significance between regimes. Therefore, Na
can not be considered a good indicator element for any of the water
types.
The Na concentrations of drainage waters (canals and creek), as well
as of lake waters, did not indicate any seasonal trends in outflow.
59
(Nmga/l)
21435.,.000
Figure 15. Monthly average Na concentrations (mg/l) for the
three water regimes. See the Appendix for explanation
of missing data.
f mineral peat = standing surface
+ standard deviation of the mean
60
Nolteraeler (1984) reported an average Na concentration of 2.78 rag/1 for
West Brice Creek, comparable to the 2.75 rag/1 Na recorded In this study.
Daniel (1981) reported Na concentrations for peat and mineral waters
collected in Pungo National Wildlife Refuge to be 4.00 and 5.51 rag/1,
respectively. Although these values are nearly twice as high as those
reported here, a similar trend exists in which mineral waters contained
higher average Na concentrations than peat waters. Peat Methanol
Associates (1983) recorded average Na concentrations of 13.25 mg/1 and
10.89 mg/1 for peat and sand units. Not only are their data more than
twice greater than those reported here and by Daniel, but a contrast in
trends exists in which peat waters had higher average Na levels than
mineral waters.
In orabrotrophic systems, such as the Croatan pocosins, the major
input of nutrients to the system is via precipitation (Moore and Bellamy,
1974). Thus, the low concentrations of the major cations (Mg, K, Na) in
standing surface waters are a result of this phenomenora. Sodium is a
good example of the close correlation between rainwater and surface
waters. The average Na levels in precipitation collected in the area by
Noltemeier (1984) was 2.21 mg/1, and the average for standing surface
waters collected in this study was 2.14 mg/1 Na. The low concentrations
of Mg in surface waters is due in part to a low input of Mg and possibly
to cation exchange by Sphagnum mats (Wetzel, 1983). Standing surface
waters contained lower levels of K compared to well waters, again
reflecting a low level of K input. Also, some K is lost to plant uptake
as K is considered an important nutrient, though seldom a limiting one
(Moore and Bellamy, 1974).
61
Cation concentrations in peat pore waters are most possibly related
to peat decomposition and to ion exchange. Living plants require
nutrient cations. With the later decomposition of organics, cations are
released. The colloidal nature of peat may also affect cation
concentrations by simple exchange.
Concentrations of Mg, K, and Na in mineral pore waters are related
to the solubility of the minerals and to the cation exchange capacity of
colloidal clays, meaning the cation exhange capacity of clays can lead to
the binding and release of cations into the interstitial pore waters
(Matthess, 1982). The K dissolved in the mineral waters may be derived
from the illite component of the clays associated with the mineral
sediments (Stumm and Morgan, 1981). Sodium ions may be released from the
crystalline structure of smectite, a sometimes Na-bearing clay also
present in the mineral soils. Some of the clays of the mineral soils are
comprised of vermiculite, a Mg-bearing mineral. The exchange of cations
(Mg for Ca) may account for the Mg content in mineral pore waters (Stumm
and Morgan, 1981).
Aluminum/Manganese/Titanium/Vanadium
Aluminum (Al) was present in detectable amounts only in the water
regimes associated with mineral sediments. Mineral pore waters contained
an average of 1.02 mg/1 Al; Catfish Lake an average of 0.28 mg/1 Al; and
West Brice Creek an average of 0.51 mg/1 Al. The Al concentrations of
both the creek and lake waters fluctuated very little seasonally, and so
no trends were evident. Aluminum data were not reported by Noltemeier
(1984), PMA (1983), or Daniel (1981); thus comparisons are not possible.
62
Manganese (Mn)<was present in detectable concentrations only in lake
waters (ave = 0.006 mg/1). Peat Methanol Associates (1983) reported
anomolous Mn concentrations of 0.27 mg/1 and 0.23 mg/1 for the fine sand
unit and peat unit, respectively. Manganese is detectable in most
groundwaters. Under reducing conditions, however, concentrations greater
than 1.0 rag/1 are rare (Matthess, 1982).
Titanium (Ti) and vanadium (V) were present in detectable
concentrations in mineral pore waters only, with averages of 0.010 mg/1
and 0.004 mg/1, respectively. Noltemeier ( 1 984), Peat Methanol
Associates (1983), and Daniel (1981) did not test for Ti or V in their
waters.
The A1 present in all the environments can be attributed to the
weathering of the alumino-silicates kaolinite, illite, vermiculite,
smectite and various feldspars (Stumm and Morgan, 1981), which are found
in the clay and sand size fractions of the mineral sediments. The mineral
sediments underlying the peat unit generally are either clayey sands or
sandy clays (Figures 5 and 6). The mineral pore waters contain A1 that
is possibly derived from the chemical weathering of these sediments.
Catfish Lake is also underlain by clayey sands in some areas, and West
Brice Creek drains mineral-dominated soils as well as peat-dominated
soils. The waters of these two systems also contain A1 in detectable
concentrations. It would seem logical for A1 to be considered an
indicator for mineral waters. However, since there is insufficient A1
data to compute valid statistical analyses, coupled with the
contamination problems associated with the trace element analyses, A1 can
not be considered an indicator for mineral pore waters.
63
Manganese is associated with various minerals, such as pyrolusite
(Mn02) and the mangani f e rous garnet, spessarite (Mn2Al2(SiO^)2) (Jones,
1945). The sand-size, detrital sediments underlying the lake and peat
deposits are derived from the weathering of crystalline rocks further to
the west. It is possible that the trace amounts of Mn in the lake waters
are related to the slow chemical weathering of these stable minerals. If
this is so, it seems likely that the mineral pore waters would contain
detectable Mn concentrations. However, due to contamination problems in
the lab, Mn concentrations in mineral pore waters are not reliable
values. Therefore, it was not possible to determine accurate Mn
concentrations in mineral pore waters.
Titanium (Ti) occurs in minerals such as rutile (TÍO2) and ilmenite
(FeTi03) (Jones, 1945). Detrital ilmenite and rutile, both chemically
and physically stable minerals, are common accessory minerals in local
beach sands. The mineral sediments of the study area do contain very
fine grained opaque minerals, some of which may be ilmenite. Therefore,
trace amounts of Ti in the pore waters might be expected.
Vanadium (V) is incorporated in minerals such as vanadinite
(Pb5(VO¿^)3Cl) and carnotite (K2(UO2)2(VO¿^)2.3H20) (Jones, 1945). If
these and other V-bearing minerals are associated with the mineral
sediments underlying the peat deposits, then the trace amounts of V
reported could be accounted. However, no such minerals were observed by
visual or microscopic inspection in the cores or grab samples during
sediment analyses. Since it was beyond the scope of this project, no
attempt was made to look for traces of V-bearing minerals by other more
sensitive methods. It is interesting to note that V is commonly present
in organic remains in varying amounts (Jones, 1945). Therefore, it is a
64
possibility that some V is present in the peat, but that it it tightly
bound to the organics and not detectable in the pore waters.
The trace elements detected in the mineral and lake waters (Mn,Ti,
and V) are all widely distributed in Igneous and sedimentary rocks, but
only in small quantities (Jones, 1945). It is feasible, then, to detect
trace amounts of these elements in groundwater as the result of slow
chemical weathering of detrital sediments.
65
SUMMARY
"The chemistry of water draining wetlands reflects the source of
water, whether from direct precipitation, surface inflow, groundwater, or
in some cases mixing of water from these different sources" (Daniel,
1981). Because of the nature of an ombrotrophic system, the regionally
elevated topography of accumulated organic remains tends to isolate a
pocosin from the influence of the regional groundwater system (Daniel,
1981). When peat accumulation forms an elevated surface, the water table
within the peat also becomes slightly elevated above the regional
groundwater system and removes itself from the chemical influence of
groundwater in the underlying mineral sediments (Daniel, 1981).
The objectives of this study were (a) to determine the chemical
characteristics of three water regimes (mineral, peat, and surface
waters) of the Croatan National Forest, and (b) to document any seasonal
fluctuations of those parameters in the outflow waters. In answer to the
first objective, three elements (TOC, Si, and Ca) appear to act as
indicator species for the three major water regimes. Mineral pore waters
contain the highest amounts of dissolved Si (ave = 3.55 mg/1),
intermediate levels of Ca (ave = 1.51 mg/1), and the lowest
concentrations of TOC (ave = 42.7 mg/1). Peat pore waters contain the
highest concentrations of TOC (ave = 212.5 mg/1) and Ca (ave = 2.09
mg/1), and the lowest Si content (ave = 0.89 mg/1) of the three regimes.
Standing surface waters contain the lowest concentrations of dissolved Si
(ave = 0.45 mg/1) and Ca (ave = 0.20 mg/1) and have intermediate
concentrations of TOC (ave = 51.0 mg/1).
66
As an aid in determining seasonal trends in outflow waters, Pearson
product-moment correlation coefficients were used on the indicator
species to test for correlations between water regimes and drainage
waters. The statistical data indicate that during the times of lowest
water level (summer and fall), peat and mineral pore waters are the major
contributors to canal drainage. With the rise in water table during the
winter and spring months, related to the decline in évapotranspiration,
increased surface runoff dilutes the waters of the canals; and surface
water becomes the major contributor to drainage waters. The seasonality
of outflow waters (concerning the three indicator elements) is as
follows: Highest TOC, Ca, and Si concentrations correspond to periods of
lowest water table when peat and mineral pore waters are draining.
During the winter and spring, increased surface runoff and the subsequent
addition of its chemical components causes TOC, Ca, and Si concentrations
in the drainage canals to decline.
Potassium (K), although not an indicator species for any particular
water regime, exhibits a seasonality in outflow waters. Concentration of
K in drainage waters increases during periods of low flow (summer and
fall) and declines during periods of high flow (winter and spring).
Noltemeier (1984) also reported this trend for K. Gorham (1967) noted a
similar trend in the English Lake District and suggests that during
periods of lowered water table associated with increased
évapotranspiration, the K already present in drainage waters is
concentrated. As surface runoff increases, the K content of drainage
channels is diluted, and K concentrations of the canals decrease.
Other ions (NO^-N, NH^-N, TDP, Mg, Na, and trace metals) present in
the outflow waters, as well as in the well waters, are not indicative of
a specific water regime, nor do they exhibit a strong degree of
67
seasonality in outflow. Had this study been conducted somewhat
differently, these ions and perhaps some others not tested, might be
found to be indicator species. As an example, more frequent sampling
(every two weeks as opposed to the monthly sampling done here) would have
resulted in a larger number of samples as well as in a closer look at
fluctuations in water chemistries.
Based on the results obtained from this project, some
recommendations for further study can be suggested and are as follows:
(1) To do a thorough stratigraphic work-up on the underlying
mineral soil, including detailed petrographic and clay
analyses, in order to possibly relate mineral pore
water chemistry to changing lithologies.
(2) To concentrate solely on peat pore waters and analyze not
only the pore waters, but also the non-fi1terab1e
colloids to determine the relationships between the colloidal
organics and the interstitial waters.
This project did not include either detailed petrographic work or
organic geochemistry; but hopefully with the results of this baseline
study, additional interest in wetland environments will be generated.
68
REFERENCES CITED
Brinson, M., D. Noltemeier, and M. Jones, 1980. Coastal plain stream
types and their influence on estuarine water quality. Unpublished
report prepared for the Environmental Protection Agency, 28 pp.
Clymo, R., 1963. Ion exchange in Sphagnum and its relation to bog ‘
ecology. Annals of Botany New Series, v.27: 309-324.
Damman, A., 1978. Distribution and movement of elements in ombrotrophic
peat bogs. Oikos, v.30: 480-495.
Daniel, C., Ill, 1981. Hydrology, geology, and soils of pocosins:
a comparison of natural and altered systems, pp. 69-108. *
In; Richardson, C., (ed.) Pocosin Wetlands. Hutchinson
Ross Publishing Company, Stroudsburg, Pennsylvania, 364 pp.
Daniels, R., E. Gamble, W. Wheeler, and C. Holzhey, 1972. Some details
of the surficial stratigraphy and geomorphology of the coastal
plain between New Bern and Coats, North Carolina. Carolina
Geol. Soc. and Atl. Coastal Geol. Assoc. Field Trip Guidebook,
36 pp.
Daniels, R., E. Gamble, W. Wheeler, and C. Holzhey, 1977a. The strati-
graphy and geomorphology of the Hofmann Forest pocosin. Soil Sci.
Soc. America Jour., v.41; 1175-1180.
Daniels, R., E. Gamble, W. Wheeler, and C. Holzhey, 1977b. The Arapahoe
Ridge - a Pleistocene storm beach. Southeastern Geology, v.l8:
231-247.
Drever, J., 1982. The Geochemistry of Natural Waters. Prentice-Hall,
Inc., Englewood Cliffs, New Jersey, 388 pp.
Dubar, J., and J. Solliday, 1963. Stratigraphy of the Neogene deposits,
lower Neuse estuary. North Carolina. Southeastern Geology, v.4:
213-233.
Elliot, R. and A. Porter, 1971. A rapid cadmium reduction method for the
determination of nitrate in bacon and curing brines. Analyst, v.96:
522-527.
Environmental Protection Agency, 1976. Methods for chemical analysis of
water and wastes. Environmental Monitoring and Support Labs.
Cincinnati, Ohio. 298 pp.
Fallaw, W., and W. Wheeler, 1969. Marine fossiliferous Pleistocene
deposits in southeast North Carolina. Southeastern Geology, v.lO:
35-54.
Gambrell, R., J. Gilliam, and S. Weed, 1974. The fate of fertilizer
nutrients as related to water quality in the North Carolina
Coastal Plain. UNC Water Resources Research Institute, Raleigh,
North Carolina, Report No. 93, 151 pp.
Gilliam, J., and R. Skaggs, 1981. Drainage and agricultural develop-
ment: effects on drainage waters, pp. 109-124. In: Richardson,
C., (ed.) Pocosin Wetlands. Hutchinson Ross Publishing
Company, Stroudsburg, Pennsylvania, 364 pp.
Gorham, E., 1967. Some chemical aspects of wetland ecology. 12th
Annual Muskeg Res. Conf. Proceeding Techn. Mem. 90, N.R.C. Assoc.
Comm, on Geotech. Res. Canada, pp. 20-38.
Gosselink, J., and R. Turner, 1978. The role of hydrology in fresh-
water wetland system, pp. 63-78. In: Good, R., D. Whigham, and
R. Simpson, (eds.) Freshwater Wetlands: Ecological Processes
and Management Potential. Academic Press, Inc., New York, New York,
378 pp.
Heinselman, M., 1970. Landscape evolution, peatland types and the environ-
ment in the Lake Agassiz Peatlands Natural Area, Minnesota.
Ecological Monographs, v.40: 235-260.
Hem, J., 1959. Study and interpretation of the chemical characteristics
of natural water. Geol. Survey Water-Supply Paper #1473, US Govern-
ment Printing Office, Washington, D.C., 269 pp.
Hemond, H., 1980. Bi o ge o che m i s t r y of Thoreau's Bog, Concord,
Massachusetts. Ecological Monographs, v.50: 507-526.
Hobbie, J., and N. Smith, 1975. Nutrients in the Neuse River Estuary. Sea
Grant Publication UNC-SG-75-21, Raleigh, North Carolina, 183 pp.
Ingram, R., and L. Otte, 1981. Peat deposits of Croatan Forest; Craven
Jones, and Carteret Counties, North Carolina. US Department of
Energy DE-AC01-79ET-14693 and NC Energy Institute, Washington, D.C.,
20 pp.
Jones, M., 1984. Nitrate reduction by shaking with cadmium: alternative
to cadmium columns. Water Res., v.l8, no. 5, pp.643-646.
Jones, W., 1945. Minerals in Industry. Penquin Books Publishers, New
York, New York, 149 pp.
Kirby-Smith, W., and R. Barber, 1979. The water quality ramifications
in estuaries of converting forest to intensive agriculture.
UNC Water Resources Research Institute, Raleigh, North Carolina,
Report No. 148, 70 pp.
Klopatek, J., 1975. The role of emergent macrophytes in mineral cycling
in a freshwater marsh, pp. 367-393. In: Howell, F., J. Gentry,
and M. Smith, (eds.) Mineral Cycling in Southeastern Ecosystems.
ERDA Symposium Series, Conf. 740513, Oak Ridge, Tennessee.
Kuenzler, E., P. Mulholland, L. Ruley, and R. Sniffen, 1977. Water quality
in North Carolina Coastal Plain streams and effects of channelization.
70
UNC Water Resources Researh Institute, Raleigh, North Carolina, Report
No. 127, 160 pp.
Matthess, G., 1982. The Properties of Groundwater. John Wiley & Sons,
New York, New York, 406 pp.
McGlynn, J., 1974. Examination of waters: evaluation of methods for selected
characteristics. Australian Water Resources Council Technical Paper
No. 8, Australian Government Publishing Service, Canberra, 109 pp.
Mixon, R., and 0. Pilkey, 1976. Reconnaissance geology of the submerged
and emerged Coastal Plain Province, Cape Lookout Area, North Carolina.
Geol. Survey Prof. Paper 859, US Government Printing Office, Washington,
D.C., 45 pp.
Moore, P., and D. Bellamy, 1974. Peatlands. Springer-Verlag, New York, '
New York, 221 pp.
National Oceanic and Atmospheric Administration (NOAA). 1945-1982. Clima-
tological data for North Carolina. Annual Summary, v.50-87, no.13.
US Department of Commerce, National Climatic Data Center, Asheville,
North Carolina.
National Oceanic and Atmoshperic Administration (NOAA). 1983, 1984.
Climatological data for North Carolina, v. 88, no.6-12 and v.89,
no.1-5. US Department of Commerce, National Climatic Data Center,
Asheville, North Carolina.
Noltemeier, D., 1984. Comparison of water quality in coastal plain /
streams of North Carolina. Unpublished master's thesis. East Carolina
University, Greenville, North Carolina, 139 pp.
Otte, L., 1981. Origin, development, and maintenance of the pocosin
wetlands of North Carolina. Unpublished report prepared for the
NC Natural Heritage Program. NC Department of Natural Resources
and Community Development, Raleigh, North Carolina, 51 pp.
Otte, L., and L. Loftin, 1983. Water chemistry of the pocosins of the
Croatan National Forest, North Carolina. Unpublished report prepared
for the UNC Water Resources Research Institute, Raleigh, North Carolina,
Project #WRRI 84-02-62002, 31 pp.
Peat Methanol Associates, 1983. Summary environmental analysis. '
Unpublished report. Groundwater Section, pp. 104-115.
Rainwater, F. and L. Thatcher, 1960. Methods for collection and analysis
of water samples. Geol. Survey Water-Supply Paper #1454, US Govern-
ment Printing Office, Washington, 301 pp.
Ray, A., (ed.), 1982. SAS User's Guide: Statistics, 1982 edition.
SAS Institute, Inc., Cary, North Carolina, 921 pp.
Richards, H., 1950. Geology of the coastal plain of North Carolina.
Trans. Am. Phil. Soc., v.40, 83 pp.
71
Richardson, C., D. Tilton, J. Kadlec, J. Chamie, and A. Wentz, 1978.
Nutrient dynamics of northern wetland ecosystems, pp. 217-241.
In; Good, R., D. Whigham, and R. Simpson,(eds.) Freshwater
Wetlands: Ecological Processes and Management Potential.
Academic Press, Inc., New York, New York, 378 pp.
Schelske, C., E. Stoermer, D. Conley, J. Robbins, and R. Glover, 1983.
Early eutrophication in the lower Great Lakes: new evidence from
biogenic silica in sediments. Science, v.222: 320-322.
Schumann, M., 1977. The Living Land; An Outdoor Guide to North Carolina.
Dale Press, Chapel Hill, North Carolina, 178 pp.
Smith, R., 1980. Ecology and Field Biology. Harper & Row Publishers,
New York, New York, 835 pp.
Snyder, J., 1977. Vegetation and site factors of the Pond Pine-Pocosin
Research Natural Area, Croatan National Forest. Unpublished report '
submitted to the Southeastern Forest Experiment Station, Asheville,
North Carolina, 62 pp.
Snyder, J., 1978. Analysis of vegetation in the Croatan National Forest,
North Carolina. Unpublished master's thesis. University of North
Carolina, Chapel Hill, North Carolina. 151 pp.
Snyder, S., L. Mauger, and W. Akers, 1983. Planktonic foraminifera and
biostratigraphy of the Yorktown Formation, Lee Creek Mine, pp. 455-
482. Ray, C., (ed.). Geology and Paleontology of the Lee
Creek Mine, North Carolina. Smithsonian Institution Press, Washington,
D.C., 529 pp.
Soloranzo, L., 1969. Determination of ammonia in natural waters by the
phenol hypochlorite method. Limnol. Oceanogr., v.l4: 799-801.
Stainton, M., M. Capel, and F. Armstrong, 1974. The chemical analysis of
fresh water. Environment Canada Miscellaneous Publ. No. 25. Research
and Development Directorate, Freshwater Institute, Winnipeg, Canada.
Stumm, W., and J. Morgan, 1981. Aquatic Chemistry, 2nd edition. John
Wiley & Sons, New York, New York, 780 pp.
Teitz, P., 1981. Holocene sedimentation and evolution of the Great Lake
Area, North Carolina. Unpublished master's thesis. University of
North Carolina, Chapel Hill, North Carolina. 82 pp.
Tooker, W., 1899. The adopted Algonquian term "pocosin". Am Anthropol.
January, pp. 162-170.
van Cleemput, 0., W. Patrick, Jr., and R. Mcllhenng, 1975. Formation of
chemical and biological denitrification products in flooded soil '
at controlled pH and redox potential. Soil Biol. Biochem., v.7:
329-332.
Verry, E., 1975. Streamflow chemistry and nutrient yields from upland
peatland watersheds in Minnesota. Ecology, v.56: 1149-1157.
Weiss, C., and E. Kuenzler, 1976. The trophic state of North Carolina
lakes. UNC Water Resources Research Institute, Raleigh, North
Carolina, Report No. 119, 224 pp.
Wetzel, R., 1983. Limnology. CBS College Publishing, Philadelphia,
Pennsylvania, 858 pp.
Whitehead, D., 1972. Development and environmental history of the
Dismal Swamp. Ecological Monographs, v.42; 301-315.
Wilbur, R., and N. Christensen, 1982. Effects of fire on nutrient
availability in a North Carolina Coastal Plain pocosin.
Am. Mid. Nat., v.l07: 54-61.
APPENDIX
Statistical data and raw data for detectable elements and ions.
Sampling sites are expressed as "peat thickness", "transect", and
"water type".
Examples :
0AM site of 0 feet peat along transect A, mineral water
4BP site of 4 feet peat along transect B, peat water
2AC site of 2 feet peat along transect A, canal water adjacent
to wells 2AM and 2AP
STR West Brice Creek
LA.1 lake water from site 1 on Catfish Lake
Table lA. Precision and accuracy daca for water analyses. Precision Is expressed as the standard deviation
of the mean of "n" samples. Accuracy la expressed as X recovery of a known concentration control
or of a aplked aample.
Method Ion ad n Z Recovery
Jones, 198A N03-N +0.04^ 12 96Z^
Solorzano, 1969 NHA-N +O.01I 8 94*3
Scalnton, etal, 197A S102-S1 +O.O3I 13 sax;
EPA, 1976 TDPOA-P 40.01^ 11 lOOX^
Ca^
— — —
Mg«
kA
TOC 4ÏTÔ1 “8 99X^
1 as determined In this study based on "n" samples.
2 as reported In cited reference.
3 as reported In McGlynn, 1974,
A Preclslon/accuracy data not available. AA spectrophotometer was calibrated In the concentration
mode. When the analysis of standards and blank showed more than a 0.02 mg/1 discrepancy, the
calibration curve was updated with three readings of all standards.
3 as reported by Oceanography International Corporation.
75
Table 2A, Anova tables where data are derived from comparison of sur-
face, peat, and mineral waters, considering transect and water
types as factors. F-ratIo represents F = ss (factor)/df and
ss (error)/df
P-value Is the probability of observing a difference as
great or greater than that observed if the null hypothesis
(H ) is true,
o
H : no difference between water regimes,
o
Source SS df F-ratlo P-value
Transect 10969.48 1 2.68 0.1050
u Type 541243.50 2 66.05 0.0001*
H Transect*Type 6850.69 2 0.84 0.4365
Error 405609.59 99
Source SS df F-ratlo P-value
*H
CO Transect 1.85 1 0.72 0.3995
Type 224.32 2 43.43 0.0001*
os|
o Transect^Type 5.80 2 1.12 0.3294
•H
CO Error 268.60 104
Source SS df F-ratio P-value
s Transect 0. 15 1 0.50 0.4816
1 Type 0.98 2 1.66 0. 1955^
æ Transect*Type 0. 99 2 1.69 0.1896
3 Error 30.61 104
Source SS df F-ratio P-va1ue
;z: Transect 0.00 1 0.01 0.9311
1 Type 1.79 2 3.54 0.0326*^
o Transect*Type 0.13 2 0.25 0.7763
z Error 26.32 104
Source SS df F-rat lo P-va1ue
FU Transect
1 0.01 1 0.05 0.8214
cf Type 1.97 2 3.94 0.0225*
FU Transect*Type 0.08 2 0. 16 0.8514
O
H Error 26.07 104
Source SS df F-ratio P-va1ue
Transect 1.36 1 0. 64 0.4277
ca Type 38.25 2 8.96 0.0003*
o
Transect*Type 7.14 2 1.67 0.1942
Error 170.73 80
Source SS df F-rat io P-value
Transect 0.33 1 0.89 0.3486
Type 0.90 2 1.22 0.3001
Transect*Type 0.03 2 0.04 0.9602
Error 30. 16 82
Source SS df F-ratio P-va1ue
Transect 0.40 1 1.38 0.2431
GO
-r* Type 0.60 2 1.06 0.3526
Transect*Type 0.35 2 0.61 0.5475
Error 22.91 80
•significant at 0.05 level
76
Table 3A . Anova tables where data are derived from comparison of peat
and mineral pore waters, considering water type, season
(time), and peat thickness as factors. F-ratio represents
F = ss (factor)/df and P-value Is the probability of observing
ss (error)/df
a difference as great or greater than that observed if the
S null hypothesis (H ) is true,oH : no difference between water regimes,oSource SS df F-ratio P-valueType 262636.63 1 103.81 0.0001»u Time 24081.70 1 9.52 0.0033H Th ickness 24973.77 3 3.29 0.0276*Error 129029.20 51•H Source SS df F-ratIo P-value1 Type 100.69 1 24.42 0.0001*C Time 3.85 1 0.93 0. 3385•H Th Ickness 10.27 3 0.83 0.4858CO Error 218.57 53Source SS df F-ratio P-valueType 14.33 1 6. 76 0.0141*Ca Time 6.74 1 3.18 0.0842Th Ickness 1.25 3 0.20 0.8976Error 65.67 31Source df F-ratIo P-value
z
Type 0.06 1 9.29 0. 0036*
Time 0.00 1 0. 12 0.7251
o
z Th Ickness 0.02 3 1.27 0.2943
Error 0.34 53
Source SS df F-ratlo P-value
z
Type 0. 19 1 2.19 0.1447
•< Time 0.01 I 0.07 0.7915
æ
z Th Ickness 0. 24 3 0.91 0.4432
Error 4.62 53
Source SS df F-ratlo P-va1ue
Type 0.01 1 6.58 0.0131*
Time 0.00 1 0.08 0.7786
O
Oi Th ickness 0.00 3 0.36 0.7841
Q
H Error 0. 07 53
Source SS df F-ratlo P-value
Type 0.84 1 9.94 0.0036*
Time 0.18 I 2. 12 0. 1557
Th 1 ckness 0. 16 3 0. 64 0.5922
Error 2.62 31
Source SS df F-ratlo P-va1ue
Type 0.33 1 5.06 0.0317*
Time 0.01 1 0. 16 0.6917
Th Ickness 0.41 3 2.09 0.1216
Error 2.30 31
?significant at 0.05 level
Table 4A. Pearson product-moment correlation coefficients. Values In parentheses denote number of pairs used In
calculating the statistic.
June July Aug Sept Oct Nov Dec Jan Feb Mar Apr 11 May
M,C -.345(8) .646(9) .588(9) - - - 1.000(4) -.616(9) - - - -
DOX P.C .921(7) 1.000(6) 1.000(4) - - - 1.000(4) -.500(8) - - - -s,c .408(5) - - - - - - - .632(5) -.962(6) -.251(6) -
TS- M,C -.004(9) -.335(9) .316(9) - -.702(9) .724(9) - -
P.C -.160(7) 1.000(6) 1.000(4) - - - .885(8) .889(8) - - - -
OTS S.C -.362(5) - - - “- “ .984(6) .958(6) .243(6) “
M.C .271(9) .000(3) - - - - .061(9) .571(9) - . -
P,C .000(4) .000(3) - - - - .057(8) .500(8) - - - -
s.c .369(5) - - - - - - - .751(6) .840(3) .742(5)
N- M.C .224(9) -.828(9) .250(9) - -.309(9) -.750(9) - _ -
ON P.C .779(7) .000(6) .000(4)
- - - .000(8) .858(8) - - - -
S.C -.372(5) - - - - - - -.448(6) -.333(6) -1.000(6) -
N- M.C -.554(9) -.349(9) .022(9) . . .451(9) .864(9) - - - -
P.C .971(7) .000(6) .000(4) - - - -.562(8) -.411(8) - - - -
HN S.C -.372(5) - - - - - - - .000(6) .000(6) .000(6)
d- M.C -.762(9) -.643(9) .000(9) - -.378(9) .000(9) -
P.C -.125(7) .000(6) .000(4) - - - .447(8) .000(8) - - - -
OdQI S.C .371(5) - - - -- - - .894(6) .000(6) .000(6) -
M.C .391(9) - - . - .272(9) .061(9) - -
P.C 1.000(4) - - - - - .642(8) .801(8) - - - -
S.C -.361(5) - - - - - - - .827(6) .000(3) .963(5) -
M.C .206(9) - - - - -.240(9) .939(9)
P.C -1.000(4) - - - - - .511(8) .692(8) - - - -
S.C .368(5) - - - - - - - .000(6) .000(3) .112(6) -
78
Table 5A, Raw elemental data TOC
{mg/1 )
SITES JUNE JULY AUG SEPT OCT NOV DEC JAN FEB MAR APRU. MAY
0AM 76.9 43.4 40.1 * # 90.5 18.9 4 4 4 4
2AM 34.1 30.7 29.0 # « « 24.1 15.6 4 4 4 4
4AM 74.0 81.0 » * * 65.3 45.7 4 4 4 4
6AM 59.4 59.2 55.4 # « # 41.3 39.4 4 4 4 4
8AM 63.2 74.6 65.2 # 80.9 82.9 4 4 4 4
OBM 11.5 25.6 13.6 * 13.4 18.2 4 4 4 4
2BM 15.6 18.3 14.3 # * * 58.3 10.9 4 4 4 4
4BM 51.9 57.3 56.3 « « 54.4 43.9 4 4 4 4
6BM 40.2 23.7 * « 9.1 10.3 4 4 4 4
4AP 212.8 382.0 « « 26.0 45.8 4 4 4 4
6AP 195.9 262.5 259.0 « 216.3 154.3 4 4 4 4
8AP 274.7 267.4 302. 1 « * « 368.6 252.1 4 4 4 4
2BP ** «« # # * 4.8 36.8 4 4 4 4
4BP 284.7 582. 1 «« « « # 218.4 78.3 4 4 4 4
6BP 209.5 211.5 204.4 » « 149. 1 113.9 4 4 4 4
2AC 51.7 54.8 53.2 ## ** «44 444 54.3 48.1 52.2 48.8 55.2
4AC 54.0 68.9 56.9 62.7 «« 444 444 42.1 41.9 45.5 44.4 46.1
6AC 54.7 70.5 59.3 61.9 #« 444 54.3 43.2 40.5 45.6 44.4 46.2
2BC 58.9 64.6 49.4 59.3 444 60.2 46.6 46.0 48.9 50.3 49.9
4BC 56.2 65.4 52.9 52.0 444 444 46.6 44.9 49.6 49.4 49.6
STR 37.4 58.9 59.5 29.3 190.5 207.8 45.3 21.8 27.4 29.2 33.7 47.4
LAI 29.5 21.9 19.9 11.8 150.8 135.9 11.8 14.3 30.7 28.8 33.8 23.0
LA2 36. 1 15.4 20.8 12.5 129. 1 115.3 52.1 23. 1 27.3 30.8 27.7 27.3
6AS «« 44 44 41. 1 4 42.5 53.2 42.8
6AS2 *« 44 44 4 39.5 42.6 47.7 4
6AS3 44 44 4 39.9 42.7 47.8 4
8AS 69.9 ** 44 52.9 40.6 4 4 4 42.4
8A2S 47.5 44 44 4 4 4 4 4
4BS 54.6 it* 44 44 39.3 40.4 43.3 50.6 41.9
4BS2 34.6 44 44 4 40.4 43.6 46.7 4
4BS3 # 44 44 4 42.9 43.7 48.6 4
6BS 56.4 44 50.0 41.9 4 4 4 41.5
*no water collected
**no water present
»*»lab problem
79
Table 5A. (continued) SIO -SI
(mg71)
SITES JUNE JULY AUG SEPT OCT NOV DEC JAN FEB MAR APRIL MAY
0AM 2.96 2.96 3.32 * * « 2.40 2.37 * # * #
2AM 3.63 3.36 3.66 « « 3.35 2.94 # «
4AM 0.64 0.29 0.71 « * « 0.77 0.70 « * *
6AM 5.34 2.54 6.62 # * « 5.76 2.52 « * «
8AM 5.24 2.07 6.52 # # # 5.29 2.85 « »
OBM 4.27 3.56 3.86 « # 2. 19 1.52 * « *
2BM 5.03 1.66 7.70 * « « 2.49 3.47 Ik * * «
4BM 3.46 0.84 14.75 « * 1.02 3.62 « * * «
6BM 4.63 3.43 5.14 * * « 4.51 3.81 « « « *
4AP 0.58 0.51 « * # 0.78 0.70 « Ik
6AP 0.46 0.46 0.53 « * « 0.57 0.57 ft * « ik
8AP 0.68 0.68 0.88 « « 1.05 1.06 « * « Ik
2BP ## # # * 0.87 0.84 « * « Ik
4BP 0. 46 0.60 #« * « « 1.04 0.71 « * « «
6BP 0.39 0.35 0.51 * « 0.59 0.55 # * « Ik
2AC 0.35 0. 29 0.46 «« #« 0.40 0. 67 0.61 0. 58 0.48 0.43 0. 48
4AC 0.37 0.30 0.27 0.13 #« 0.35 0.59 0.53 0.53 0.41 0.39 0.40
6AC 0.35 0.26 0.29 0. 10 «« 0.33 0.57 0. 53 0.51 0.40 0.38 0.38
2BC 0.51 0.55 0.44 0.38 «« 1.31 0.82 0.80 0.87 0.57 0.50 0.50
4BC 0.42 0.35 0.38 0.67 ## 0.41 0.77 0.68 0.68 0.56 0.49 0.50
STR 0.78 0.67 1.43 1.74 2.58 1.58 0.61 1.95 2.16 1.85 1.64 0.66
LAI 0. 78 0.66 0.69 0.71 0. 14 0. 00 0.02 0. 16 0.21 0.27 0.44 0.46
LA2 0.81 0.62 0.68 0.65 0. 13 0.00 0.45 0.41 0.27 0.30 0.42 0.51
6AS ** ** ** 0.58 0.53 0.51 0.40 0.38 0.40
6AS2 «« «« «« « # 0.49 0.40 0.40 Ik
6AS3 «« ** Ik 0.48 0.41 0.39 *
8AS 0.37 #« #« 0.47 0.50 # « # 0.33
8A2S 0.32 #« * # « « « *
4BS 0.36 #* *« 0.63 0.61 0. 59 0.48 0.39 0.35
4BS2 0.35 ** # * 0.59 0.45 0.39
4BS3 « «« ** « * 0.59 0.47 0.40 *
6BS 0.41 #« #« #« 0.60 0. 59 # « ik 0.36
•no water collected
»*no water present
»**|ab problem
80
Table 5A, (continued) (îa
(mg/ I )
SITES JUNE JULY AUG SEPT (KT NOV DEC JAN FEB MAR APRIL MAY
0AM 2.09 0.93 « # * 1.08 1.03 * * a
2AM 2. 39 ««« « 0. 72 0.69 « # n
4AM 0.83 ««« « * 0.10 0.12 * # it
6AM 0. 58 0.73 « * » 0.47 0.35 « « * it
8AM 2.89 2.34 ««« # * # 1.20 1.10 # « # it
OBM 9.66 * * * 4.35 4.09 « * #
2BM 0.63 #«# * « # 0.44 0.35 « * * «
4BM 1.69 ««« « « « I. 13 1.18 It * it #
6BM 1.13 *** « « 0.35 0.72 # * it n
4AP 1.22 # « * 1.60 0.25 « « it it
6AP «#« 2.64 #«* » # « 0.75 0.80 # » it it
8AP 5.83 6.38 « H 1.08 0.52 * # it it
2BP #« * « « 1.07 0.92 « « it it
4BP 3. 14 » * » 2. 10 1.46 « « it it
6BP 4.09 #« # « 1.33 2.54 « » it it
2AC 0.48 ««* #« 0.46 0. 12 0. 17 0. 19 0. 12 0.06 0.20
4AC 0.23 «»« 1.13 «« 0.23 0.08 0.13 0.10 0.04 «*« 0.03
6AC 0.28 ««* »«« 0.63 0.35 0.08 0.08 0.28 0.04 0.20
2BC 0.38 »»« 1.89 »« 0.35 0. 18 0.12 0.46 0.07 0.24 0.13
4BC 0.23 ##« **# 0. 33 «« 0.20 0.10 0. 14 0.10 0. 19 0.05 0.50
STR 0.38 0.23 *«« 1. 13 0.78 0.55 0.33 0.14 0.35 0.09 0.08 0.21
LAI 0.73 0.46 ##« 1.24 0.88 1.00 0.72 0.66 0.64 0.47 0.51 0.50
LA2 0.68 0.46 0.93 1.74 0.85 0.35 0.65 0.74 0.52 0.42 0.50
6AS «*# ** 0. 13 0.08 0.07 0.06 0.02
6AS2 * 0.19 0.15 0.03 it
6AS3 ** « # 0.08 0.02 ititit it
8AS 1.13 «« ** #« 0.10 0.05 * H a 0.25
8AS2 1.54 »« ** # « # n a
4BS 0. 13 #«« ** 0. 14 0.09 0. 11 0.04 0.10 0.04
4BS2 0.28 ## ## «# # * 0. 14 0. 16 0.07
it
4BS3 « «« «« « # 0.11 0.05 0.15 it
6BS 0.43 #« ** 0. 10 0. 19 * « it 0.21
*no water collected
**no water present
»«*lab problem
81
Table 5A, (continued) NO -N
(mg/I)
SITES JUNE JULY AUG SEPT (xrr NOV DEC JAN FEB HAR APRIL MAY
0AM 0.01 0.00 0.01 * « « 0.01 0.02 « # « it
2AM 0.01 0.00 0.01 « 0.02 0.02 * « « it
4AM 0.04 0.03 0.03 « » « 0.02 0.02 # « it
6AM 0.03 0.01 0.01 « « « 0.05 0.04 # # « it
8AM 0.06 0.04 0.02 « # « 0.05 0.06 # * it
OBM 0.01 0.04 0. 02 « * 0. 10 0.08 « # #
2BM 0.01 0.01 0.01 * « « 0.04 0.02 * » * it
4BM 0.03 0.03 0.01 « * « 0.02 0. 02 * it
6BM 0.01 0.00 0.00 * * 0.03 0.05 * it it
4AP 0.08 0.06 * « 0.04 0.02 « « it it
6AP 0.22 0.04 0.04 » « 0.10 0.07 * « it it
8AP 0. 28 0.02 0.04 # 0.08 0.07 « « it it
2BP ** * « « 0.04 0. 03 « # it it
4BP 0.30 0.04 * * # 0. 58 0.18 « # it it
6BP 0.16 0.02 0.03 * * « 0. 16 0. 19 * * it it
2AC 0.06 0.03 0.03 0.05 0.05 0.04 0.05 0.02 0.04 0.05
4AC 0.05 0.02 0.03 0.06 ** 0.04 0.04 0.02 0.03 0.02 0.04 0.04
6AC 0.05 0.02 0.03 0.04 «« 0.04 0.04 0.02 0.02 0.01 0.02 0.02
2BC 0.04 0.03 0.03 0.06 ** 0.04 0.04 0.03 0.04 0.03 0.02 0.03
4BC 0. 07 0.01 0.02 0.04 0.03 0.04 0.04 0.04 0.03 0.04 0.04
STR 0.03 0.01 0.03 0.04 0.03 0.02 0.02 0.02 0.01 0.01 0.01 0.02
LAI 0.09 0. 11 0.07 0.11 0.09 0.05 0.06 0.07 0. 10 0.09 0.07 0.09
LA2 0.07 0. 11 0.07 0. 11 0.09 0.05 0.05 0.07 0.09 0.08 0.09 0.09
6AS ih* ## #« 0.02 0.02 0.02 0.02 0.02 0.02
6AS2 ** ## « 0.02 0.01 0.02 it
6AS3 ** * « 0.02 0.02 0.02 it
8AS 0.01 0.02 0.02 « it 0.03
8A2S 0.01 ** It a * # it it
4BS 0.01 «« #« ** 0.02 0.02 0.02 0.01 0.01 0.02
4BS2 0.01 «* ** # n 0.02 0.02 0.01 it
4BS3 * int * « 0.01 0.01 0.01 it
6BS 0.01 ** 0.02 0.02 « « it 0.02
*no water collected
**no water present
***lab problem
82
TabI e 5A, (continued) NH -N
(mg/1)
SITES JUNE JULY AUG SEPT OCT NOV DEC JAN FEB HAR APRIL MAY
0AM 0.19 0.24 0.19 * * 0. 14 0.00 « « « *
2AM 0.01 0.54 0.16 * * « 0.44 0.40 * « # «
4AM 0.08 0.07 0.09 # # 0.02 0.02 * *
6AM 0.40 0.21 0.23 « « * 0.03 0.02 * * *
8AM 0.14 0.16 0.11 « # # 0.12 0.25 * « # «
OBM 0.04 0. 19 0.27 « * « 0.02 0.03 « « « «
2BM 0.01 0.03 0.00 « « « 0.01 0.00 * * *
4BM 0.08 1.48 0. 12 a « 0.01 0.01 « * « «
6BM 0.04 0.02 0.00 # * * 0.00 0.00 « *
4AP 0.09 0.11 « « « 0.04 0.02 # « « *
6AP 0.27 0.15 0.10 « * 0.09 0.20 # #
8AP 0. 16 0.35 0.33 « # * 0.30 0.23 # « « *
2BP *# « * « 0.03 0.04 » «
4BP 0.71 0.47 # 1.02 0.68 « « * «
6BP 0.24 1.26 1.14 # 0.29 0.21 * » *
2AC 0.08 0.03 0. 13 «« «« 0.02 0.03 0.03 0.02 0.04 0.03 0.02
4AC 0.06 0.03 0.00 0.03 0.02 0.02 0.01 0.01 0.03 0.02 0.02
6AC 0.06 0.03 0.00 0.01 0.01 0.03 0.01 0. 00 0.02 0.01 0.01
2BC 0.06 0.04 0.01 0.01 0.01 0.02 0.02 0.01 0.04 0.02 0.02
4BC 0. 09 0.03 0.00 0.01 0.01 0.02 0.01 0.01 0.04 0.02 0.02
STR 0.03 0.02 0.01 0.00 0.02 0.00 0.01 0.00 0.00 0.01 0.00 0.01
LAI 0.04 0.02 0.01 0. 03 0.00 0.00 0.01 0.80 0.01 0.02 0.01 0.04
LA2 0.05 0.02 0.00 0.03 0.02 0.00 0.02 0.03 0.01 0.01 0.01 0.01
6AS «« «« ** 0.02 0.01 0.00 0.02 0.01 0. 00
6AS2 ## ** *» * « 0.00 0.02 0.02 *
6AS3 «« H * 0.00 0.02 0.02 «
8AS 0.05 0.02 0. 02 « « « 0.02
8A2S 0.02 ** ** » K * # « «
4BS 0. 03 ** 0.02 0.01 0.00 0.02 0.02 0.02
4BS2 0. 03 « 0. 00 0.02 0.01 *
4BS3 # * # 0.00 0.02 0.02 *
6BS 0.01 ** 0.02 0.02 * « 0.01
*no water collected
**no water present
***lab problem
83
Table 5A. (continued) TDPO -P
(mg/ I )
SITES JUNE JULY AUG SEPT OCT NOY DEC JAN FEB MAR APRIL MAY
OAM 0.06 0.04 0.01 # « * 0.01 0.01 * « #
2AM 0.03 0.08 0.01 * # 0.01 0.01 « # « #
4AM 0.10 0.02 0.03 n * « 0.01 0.01 H « « *
6AM 0.04 0.01 0.01 * « « 0.00 0.01 « « *
8AM 0.09 0.03 0.02 * « 0.02 0.02 # * « *
OBM 0.04 0.01 0.01 « * * 0.01 0.03 « »
2BM 0.07 0.02 0.01 « « « 0.01 0.01 » «
4BM 0. 10 0.03 0.01 0.01 0.02 # * *
6BM 0.06 0.01 0.00 * * 0.00 0.00 * * * *
4AP 0.06 «« 0.01 1k « # 0.02 0.01 « * «
6AP 0.12 0.02 0.02 * # # 0.02 0.01 * «
8AP 0.12 0.07 0.03 « « « 0.06 0.01 « # «
2BP #* # * 0.02 0.02 « * » *
4BP 0.15 0.02 «« « # 0. 18 0.11 « «
6BP 0.06 0.14 0.07 * * # 0.04 0.04 « * # «
2AC 0.02 0.01 0.01 «« 0.01 0.01 0.01 0.01 0. 00 0.00 0.01
4AC 0.01 0.02 0.01 0.01 ** 0.01 0.00 0.01 0.01 0.00 0.00 0.01
6AC 0.02 0.03 0.01 0.00 *« 0.01 0.01 0.01 0.00 0.00 0.01 0.01
2BC 0.03 0.03 0.01 0.00 0.01 0.00 0.01 0.01 0.01 0.01 0.01
4BC 0.01 0.03 0.01 0. 00 0.01 0.01 0.01 0.01 0.02 0.00 0.01
STR 0.01 0.01 0.01 0.00 0. 02 0.00 0.00 0.01 0.01 0.01 0.00 0.00
LA1 0.01 0.01 0.02 0.00 0.01 0.00 0.00 0.01 0.01 0.01 0.00 0.01
LA2 0.01 0.01 0.02 0.00 0.01 0.00 0.06 0.02 0.01 0.01 0.00 0.01
6AS «« ** ** 0.01 0.00 0.00 0.00 0.00 0.01
6AS2 ** #« # * 0.00 0.00 0. 00
6AS3 «* ** ** #* * * 0.00 0.00 0.00 «
8AS 0.14 «« 0.00 0.00 * « 0.00
8A2S 0.13 ** ** * » * «
4BS 0.07 0.00 0.01 0.01 0.00 0.00 0.01
4BS2 0. 08 #* « « 0.01 0.00 0.00 *
4BS3 * ** ** ## # * 0.02 0.00 0.00 *
6BS 0.09 ** 0.00 0.01 « « * 0.01
•no water collected
**no water present
•••lab problem
84
Table 5A. (continued ) Mg
(mg/ I )
SITES JUNE JULY AUG SEPT OCT NOV DEC JAN FEB MAR APRIL MAY
0AM 0.96 0.89 * « « 1.84 1.78 * » »
2AM 1. 11 ««« « * 0.75 0.73 * ik » »
4AM 0.91 * * » 0. 59 0.51 * » »
6AM 0.25 0.24 « « # 0.23 0.19 * « » »
8AM 0.64 0.68 « 0. 52 0.51 * « » »
OBM 0.89 # « 0.58 0.61 * * » »
2BM 0.23 * * 0.82 0.29 * * » »
4BM 0.32 * * « 0.33 0. 33 * it » »
6BM 0.27 * « * 0.16 0. 15 * » »
4AP 0.63 * # 0.92 0.49 « » »
6AP 1.05 ##« * « « 0. 52 0.50 * it » »
8AP 1.22 1.52 * « « 0.53 0.41 * it » »
2BP * * * 0.75 0.58 it » »
48P 1.01 ««« *#« * * » 0.86 0. 58 « it » »
6BP 0.75 * # « 0.68 0.65 it » »
2AC 0.42 ##» 0. 70 0. 59 0.50 0.53 0.44 0.37 0.46
4AC 0.33 0.36 «« 0.60 0.53 0.48 0.47 0.39 »»» 0.39
6AC 0.31 0.32 0.60 0. 53 0.45 0.54 it»* 0.37 0.43
2BC 0.43 »»« *«« 0.50 *« 0.69 0.66 0.70 0.68 0.49 0.50 0.46
4BC 0.34 0.35 0.68 0.66 0.55 0.55 0.50 0.43 0.48
STR 0.34 0.42 0.33 0.31 0.74 0.80 0.68 0.75 0.56 0.48 0.55
LAI 0.55 0.33 0.65 0.68 0.79 0. 68 0.60 0.68 0.68 0.66 0.58
LA2 0.51 0.33 0.65 0.68 0.78 0.55 0.68 0.65 0.70 0.56 0.62
6AS *** 0.55 0.50 0.51 »»» 0.39 0.42
6AS2 ** *## ** # 0.53 0.43 0.39 ?
6AS3 #« *« ## # # 0.50 0.40 »»» »
8AS 0.57 ## 0.52 0. 50 Ik » » 0.45
8A2S 0.49 «#« ** #« *# » Ik * » » »
4BS 0.38 0.72 0.68 0.55 0.50 0.45 0.45
4BS2 0. 37 ** n* ## #* Ik 0.58 0.50 0.43 »
4BS3 # ##« ## « * 0.63 0.49 0.45 »
6BS 0.49 ** »«# 0.68 0. 68 * » » 0.51
?no water collected
**no water present
•?•lab problem
85
Table 5A. (continued) K
(mg/I )
SITES JUNE JULY AU6 SEPT XT NOV KC JAN FEB MAR APRIL MAY
0AM 0.90 0.66 * « * 0.32 0.33 * # « #
2AM 1.59 « * 1.26 1.22 « # * #
4AM 0.52 ««« » * # 0.12 0.06 It * it #
6AM 0.61 1.02 * # 0.65 0.49 # « it #
8AM 0.83 0.83 « « # 0.67 0.68 « » it #
OBM 0. 56 « # 0.30 0. 43 « * it #
2BM 0.56 *** « # » 0.26 0.43 « « it #
4BM 0. 60 * « 0.44 0.34 It it #
6BM 1.37 # « 0.57 0.44 It « it #
4AP 0.43 * « * 0. 17 0.08 * * it #
6AP 0.73 *** * » « 0.30 0.22 * it #
8AP 0.89 0.90 * # * 0.51 0.47 « * it #
2BP ## # « # 0. 11 0.24 * it #
4BP 0.82 ««* « # * 0.48 0.31 « * it #
6BP 0.57 * « « 0.24 0.27 # it #
2AC 0. 29 *« «« 0.77 0.06 0. 1 1 0.07 0.07 0. 13 0. 13
4AC 0.22 #«* 0.76 «« 0.26 0.07 0.06 0.01 0.06 it*# 0.07
6AC 0.26 ««« *«« 0.89 ** 0.20 0.03 0.08 0.02 *«* 0.11 0.05
2BC 0.39 0.85 *« 0.43 0.04 0.07 0.02 0.05 0.09 0.06
4BC 0.20 «»# 0. 06 #« 0.31 0. 08 0.08 0.00 0.06 0.07 0.05
STR 0.32 0.14 0.48 0.75 0.30 0.06 0.11 0.03 0.07 0.09 0.07
LAI 0.39 0.38 0.40 0.81 0.38 0.25 0.22 0.20 0.30 0.24 0.29
LA2 0.36 0.35 *«« 0.39 0.30 0.38 0. 20 0.21 0. 18 0.28 0.28 0.27
6AS ** ** «*# #» 0.06 0. 06 0. 00 0. 10 0.08
6AS2 #* * 0.00 0.07 0.07 #
6AS3 ** ** #* * » 0.00 0.06 it## #
8AS 0. 76 #«« ** 0.01 0. 03 « # # 0.06
8A2S 0.64 * H » « # #
4BS 0. 18 0.04 0.05 0.00 0.04 0.05 0.06
4BS2 0. 12 »# » « 0.00 0.03 0.05 #
4BS3 « ** «# « « 0.00 0.03 0.05 #
6BS 0.79 #* ** 0.07 0.06 « # 0.07
*no water collected
**no water present
»**lab problem
86
Table 5A, (continued) Na
(mg/ I )
SITES JUNE JULY Aue SEPT OCT NOV DEC JAN FEB MAR APRIL MAY
0AM *#« *«* « * * 4.81 4.75 # « « 3.06
2AM ««« « # * 5.64 5.61 « « « 5.31
4AM « « 2.31 «»* « # ##«
6AM ««« « # * 2.94 2.86 « « « 2.32
8AM #»* « # # 2.39 #«« « *
OBM * * * 5.87 * * « 6.31
2BM »«* a # « *«« 2.22 * * « 3.09
4BM **« # # * 2.47 2.70 # * *«#
6BM *## * 2.20 « # *
4AP #«* « # « ##* # * « 2.73
6AP * # « »«« 3.07 * # « 2.41
8AP »«« # « * 3.66 3.81 « « « 3.62
2BP **« *#« * # * 2.13 *«« « « « 2.75
4BP «#» « * # 2.65 « # * 2. 16
6BP * » # 2.66 3.66 * « 3.57
2AC 1.88 #«# ««« #«« ««« 2. 17 2.05 2.35 ««« 2.66 «««
4AC ««« 2.37 #«* *«* 2.34 2.57 2.83 «#* ##* 2.61
6AC «#« ««# ««* ««« 2.33 ««« 2.74 2.69 1.90 1.98
2BC #«« **« *»« »«« 2.96 3.19 #«« *#« ««« 3.03
4BC ##« **« #*« ««« *«« 2.55 1.83 2.29 «««
STR ««* *«« 2.39 2.77 3.19 3.89 3.62 0.03 2.60 3.49
LAI ««« 3. 10 #«« »«« 3.33 2.88 3.28 3.12 »«» #»« ««# 2.94
LA2 #«« 3.69 **« #«* ««# 0.62 *«« 2.58 ««« *** 2.34
6AS ## #** ** ## 2.02 1.75 *««
6AS2 ##« *« *« # * 2.42 2.11 *
8AS #« *«* ** It* #» #«« 2.52 « 2.00
4BS ### #* #« 2.39 2.41 # # #«« 2.11
4BS3 ## «*« # * 2.32 2.07 1.99 «
6BS »# *«« #» «« 2.81 2.66 * « « 2.51
*no water collected
**no water present
***lab problem
87
TabI e 5A. (continued) AI
(mg/1)
SITES JUNE JULY AUG SEPT OCT NOV DEC JAN FEB MAR APRIL MAY
0AM «## » # » 1.76 1.95 « » « 1.36
2AM #«« » ft « 2.00 1.84 * » # 1.70
4AM #** » « # 0.11 # * « **«
6AM » # # 0.36 0.49 « » » 0.42
8AM ««# ##« « « # 0.45 # * « ##«
OBM #«# * # » 0.53 « « * 3.53
2BM «*« # * »# 0. 17 * # * 0.40
4BM * # # 0.23 0. 58 «
6BM IHHk )(«# *** # * # 0.36 *«« * # « #*#
STR 0.32 0.21 0.34 0.85 0.82 0.01 0.67 0.84
LAI 0.32 0.25 0.22 0.32 0.35 **# 0.32
LA2 0.33 «## 0. 00 0.09 ««« 0.29 0.27
Ml
(mg/I)
SITES JUNE JULY AUG SEPT OCT NOV DEC JAN FEB MAR APRIL MAY
LAI ««* 0.009 *«« 0. 009 0.009 0.009 0.014 *»« »** 0.008
LA2 ««« 0.010 *«# 0.000 0.001 0.000 0.031 «»* **# 0. 005
*no water collected
•*no water present
**»lab problem
88
Table 5A, (continued) TI
(mg/1)
SITES JUNE JULY AUG SEPT OCT NOV DEC JAN FEB MAR APRIL MAY
0AM 0.000 ### * « « 0.002 0.001 « « # 0.001
2AM #«# 0.000 » « * 0.002 0.001 « » # 0.001
4AM 0.000 * # 0.003 0.000 * « « 0.000
6AM 0. 000 « * 0.003 0. 004 * 0. 002
8AM 0.000 * * # 0.001 0.000 n * « 0.000
OBM 0.000 « # 0.020 0.000 « « « 0. 126
2BM 0.000 » « 0.000 0.001 « « 0.001
4BM 0.000 » « 0. 007 0.013 » * # 0.000
6BM ##« 0.000 »»# « * # 0.001 0.000 « # 0.000
V
{mg/1)
SITES JUNE JULY AUG SEPT OCT NOV DEC JAN FEB MAR APRIL MAY
0AM 0.000 «*« » « 0.004 0. 004 « « « 0.002
2AM 0.000 # # # 0.009 0.009 * « « 0.007
4AM 0.000 # * « 0.000 0.000 « 0.000
6AM 0.000 « # 0.008 0.008 « * « 0.008
8AM *#« 0.000 « # « 0.004 0. 000 « « 0.000
OBM 0.000 ### # » 0. 001 0.000 * * n 0.005
2BM 0.000 «## « « 0.000 0. 001 « It 0.001
4BM 0. 000 « 0.001 0. 001 # n it 0.000
6BM 0.000 # * # 0.001 0.000 # « # 0. 000
*no water collected
•*no water present
»**lab problem