EPA 600/3-81-046
July 1981
EARLY DIAGENESIS AND CHEMICAL MASS TRANSFER
IN LAKE ERIE SEDIMENTS
by
Gerald Matisoff
J. Berton Fisher
Department of Geological Sciences
Case Western Reserve University
Cleveland, OH 44106
and
Wilbert Lick
Department of Mechanical & Environmental Engineering
University of California
Santa Barbara, CA 93106
Contract No. R805716020
Project Officer
David Dolan
Large Lakes Research Station
Environmental Research Laboratory -- Duluth
Grosse lie, MI 48138
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
ENVIRONMENTAL RESEARCH LABORATORY
DULUTH, MN. 55804
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO.
EPA-600/3-81-046
2.
ORD Report
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
larly Diagenesis and Chemical Mass
Transfer in Lake Erie Sediments
5. REPORT DATE
July 1981
6. PERFORMING ORGANIZATION CODE
7.AUTHOR Gerald Matisoff, J. Berton Fisher and Wilbert
.ick*, Dept. of Mechanical & Environmental Eng., Univer-
sity of California. Santa Barbara. California 93106
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Department of Geological Sciences
Case Western Reserve University
Cleveland, Ohio 44106
10. PROGRAM ELEMENT NO.
1BA769
11. CONTRACT/GRANT NO.
R805716
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Large Lakes Research Station
Environmental Research Laboratory-Duluth
9311 Groh Road, Grosse He, Michigan 48138
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Vertical profiles of pore water and sediment solids chemistry were obtained from two
sites in Lake Erie. Samples were collected using both gravity coring and pore water
peeper" techniques. In general, concentrations of nutrients and toxic metals in sedi-
ment solids decreased with increasing depth. Comparison of pore water "peeper" data 1
gravity core data showed that "peeper" data provides higher resolution near the sedi-
ment-water interface. The thermodynamic tendency of metal phosphate and carbonate
mineral phases to precipitate in Lake Erie sediments has been calculated by means of an
ion-pair model of the interstitial water chemistry. The calculations suggest that de-
trital calcite, aragonite, and dolomite should be dissolving in the sediments, but that
iron and manganese carbonates should be precipitating. Regenerated phosphate should be
reacting with calcium, iron, manganese, and lead to form authigenic mineral phases.
Whitlockite (Ca3(P04)2) and not hydroxylapatite (Ca5(P04)30H) is the predicted mineral
phase controlling phosphate solubility. Rates of anaerobic decomposition of Lake Erie
sediments from one locality were determined for seven depth intervals at three tempera-
tures. Concentration increases of bicarbonate, phosphate, ammonium, calcium, magne-
sium, iron, and manganese in pore water within any given depth interval followed zeroth
order kinetics and exhibited Arrhenius temperature dependency. The observed release
rates decrease exponentially with depth in the sediment due to a corresponding decrease
in the amoim'LQf oxidizabl& organic matter and acid hvdrolvzable mineral phases.
17.
KEY WORDS AND DOCUMENT ANALYSIS
a.
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Diagenesis
Sediments
Decomposition Reactions
Geological Sedimentation
Precipitates
Fermentation
08/D
08/H
13. DISTRIBUTION STATEMENT
19. SECURITY CLASS (ThisReport)
21. NO. OF PAGES
20. SECURITY CLASS (This page)
22. PRICE
EPA Form 2220-1 (9-73)
-------
DISCLAIMER
This report has been reviewed by the Large Lakes Research Station, U.S.
Environmental Protection Agency, and approved for publication. Approval does
not signify that the contents necessarily reflect the views and policies of
the U.S. Environmental Protection Agency, nor does the mention of trade names
or commercial products constitute endorsement or recommendation for use.
ii
-------
Abstract
Vertical profiles of pore water and sediment solids chemistry were ob-
tained from two sites in Lake Erie. Samples were collected using both gravity
coring and pore water "peeper" techniques. In general, concentrations of
nutrients and toxic metals in sediment solids decreased with increasing depth.
Comparison of pore water "peeper" data to gravity core data showed that
"peeper" data provides higher resolution near the sediment-water interface.
Modifications of the present peeper are required to adequately sample easily
oxidizable materials (e.g. ammonia, ferrous iron). '
The thermodynamic tendency of metal phosphate and carbonate mineral
phases to precipitate in Lake Erie sediments has been calculated by means of
an ion-pair model of the interstitial water chemistry^ The calculations
suggest that detrital calcite, aragonite, and dolomite should be dissolving in
the sediments, but that iron and manganese carbonates should be precipitating.
Regenerated phosphate should be reacting with calcium, iron, manganese, and
lead to form authigenic mineral phases. Whitlockite (Ca_ (P0.)9) and not
•3 4r &•
hydroxylapatite (Ca5 (P04)3 OH) is the predicted mineral phase controlling
phosphate solubility. Zinc and cadmium are apparently controlled by other
mechanisms, perhaps by sulfide phases, mixed mineral phases, adsorption and/or
ion exchange equilibria.
Rates of anaerobic decomposition of Lake Erie sediments from one locality
were determined for seven depth intervals at three temperatures. Sealed
sediment sections were incubated under anoxic conditions and the interstitial
waters were sampled over a period of approximately 200 days. Concentration
increases of bicarbonate, phosphate, ammonium, calcium, magnesium, iron, and
manganese in pore water within any given depth interval followed zeroth order
iii
-------
kinetics and exhibited Arrhenius temperature dependency. The rates and ener-
gentics of these fermentation reactions are only slightly less than those
reported from sediments undergoing sulfate reduction. The observed release
rates decrease exponentially with depth in the sediment due to a corresponding
decrease in the amount of oxidizable organic matter and acid hydrolyzable
mineral phases.
s*^*
J
• A stoichiometric model was constructed utilizing the observed release
rates and assumed chemical reactions to predict the stoichiometry of the
decomposing organic matter and the natur.e of the hydrogen buffer.
y The flux of nutrients and metals from Lake Erie sediments to anoxic
overlying water were studied in laboratory microcosms. Three cases were
investigated: 1) homogenized sediment without worms, 2) homogenized sediment
V
preconditioned by tubificid worm activities, 3) natural lake cores. Flux
estimates were made using both direct (concentration changes in the overlying
water) and indirect (pore water concentration gradients) techniques.j- Tubi-
ficids were found to increase the flux of ammonia, and decrease the flux of
iron, soluble reactive phosphorous, and soluble reactive silica. The presence
of tubificids had no effect on bicarbonate flux. Fluxes observed in the
t
natural lake core experiment were similar to those observed in the homogenized
sediment with tubificids. Mineral equilibrium calculations performed for the
pore water data collected in these experiments showed that the laboratory
microcosms provided a reasonable representation of chemical conditions in Lake
Erie sediments. Comparison of direct and indirect flux estimates showed that
both types of estimate had high variability. £ln general, indirect flux esti-
mates were higher than direct flux estimates .-j
Field measurements and data from the production experiment were used to
calculate loads and losses of metals (Cu, Pb, Zn) and nutrients (organic
nitrogen) to and from Lake Erie sediments./ It was found that man's influence
iv
-------
on metals was Zn»Pb>Cu. The natural abundance ratio is Zn>Cu>Pb. Further,
it was found that the inclusion of organic decomposition in the calculation of
apparent anthropogenic nitrogen loading to Lake Erie sediments resulted in a
significant decrease (about a factor of two) in the estimate of anthropogenic
loading.
The time dependent behavior of ammonia and bicarbonate in the homogenized
sediments lacking a tubificid population could be well approximated by a one
dimensional time dependent transport-reaction model. More complex models are
needed to provide adequate descriptions for other parameters and situations
(real lake sediments, homogenized sediments with tubificids).
-------
Contents
Page
Disclaimer ii
Abstract iii
Figures viii
Tables xii
1. Introduction
2. Conclusions 2
3. Methods and Materials 5
Field Measurements 5
Study Area 5
Core Collection 7
Pore Water Peeper 8
Laboratory Experiments 8
Production Experiment 8
Flux Experiment 11
4. Results 16
Field Measurements 16
Production Experiment 28
Release Kinetics 28
Depth Dependency . 33
Temperature Dependency 40
Flux Experiment 44
Overlying Water Chemistry 44
Interstitial Water Chemistry 48
Chemistry of the Sediment Solids 53
5. Discussion 64
Vi
-------
Page
Mineral Equilibria 64
Calculation of Activities 66
Calculation of Ion Activity Products 68
Field Measurements 71
Production Experiment 75
Flux Experiment 78
Mineral Equilibria in Anoxic Lake Sediments 86
Stoichiometric Model 92
Solid Phase Prediction 97
Prediction of Pore Water Concentration 100
Direct and Indirect Flux Estimates 102
Laboratory Flux Experiments 103
Lake Core Experiments 107
Field Cores 108
Load and Loss Calculations 108
Metals 108
Nutrients 126
Prediction of Pore Water Chemistry 133
6. References 138
7. Appendices 147
I. Chemical Methods 147
II. Field Core Data 158
III. Production Experiment Data. ...... , . . . . 165
IV. Flux Experiment Data 171
-------
FIGURES
Number Page
1 Field sampling localities in Lake Erie ............. . . 6
2 Apparatus used for the flux experiment: 1. Sampling cover
2. Detachable top 3. Overlying water reservoir 4. Sediment
chamber 5. Extruder piston 6. Extruder base 7. Quick
connect valve 8. Turbine magnetic stirrer ............ 13
3 Total and organic carbon versus depth at Station 83 ........ 17
4 Total and organic carbon versus depth at Station Al ........ 18
5 Acid volatile sulfide versus depth at Station 83 ........ . . 19-
6 Total sulfur versus depth at Station 83 .............. 20
7 Total Kjeldahl nitrogen versus depth at Station 83 ........ 21
8 Total Kjeldahl nitrogen versus depth at Station Al ......... 22
9 Total phosphorous versus depth at Station 83 ............ 23
10 Total phosphorous versus depth at Station Al ............ 24
11 Vertical profiles of ferrous iron, ammonia, and SRP in the
interstitial water of sediments at Station Al determined
by both gravity coring and pore water "peeper" technique ..... 26
12 Vertical profiles of chloride, alkalinity, and SRS in the
interstitial water of sediments at Station Al determined
by both gravity coring and pore water "peeper" technique ..... 27
13 Ammonia concentration data and derived regression line from
the production experiment for three temperatures and four
depths .............................. 30
14 Ammonia production rate, Rf versus total Kjeldahl nitrogen ... 32
15 Ammonia production rate at three temperatures as a function
of depth ....................... ...... 34
16 Bacterial abundance as a function of depth at Station 83 ...... 39
Vlll
-------
17 Excess Pb-120 versus depth at Station 83 41
18 Concentration of ammonia, ferrous iron, alkalinity, SRP,
and SRS and pH in the overlying water of homogenized mud
experiment cores. C-with tubificide, T-without tubificide ... 45
19 Concentration of ammonia, ferrous iron, alkalinity, SRP,
and SRS and pH in the overlying water of a lake core
experiment core. (LCE-4) 47
20 Vertical profiles of ammonia, ferrous iron, alkalinity, SRP,
and SRS in the homogenized mud experiment cores without worms. . 49
21 Vertical profiles of ammonia, ferrous iron, alkalinity, SRP,
and SRS in the homogenized mud experiment cores with worms ... 50
22 Vertical profiles of ammonia, ferrous iron, alkalinity, SRP,
and SRS in the lake core experiment cores 52
23 Total and organic carbon in homogenized mud core R (+72 days)... 54
24 Total and organic carbon in homogenized mud core T (+169 days) . . 55
25 Total and organic carbon in lake core 6 (+13 days) 56
26 Total Kjeldahl nitrogen in homogenized mud core R (+72 days) ... 57
27 Total Kjeldahl nitrogen in homogenized mud core T (+169 days) . . 58
28 Total Kjeldahl nitrogen in lake core experiment core 6 (+13
days) 59
29 Total phosphorus in homogenized mud core R (+72 days) 60
30 Total phosphorus in homogenized mud core T (+169 days) ...... 61
31 Total phosphorus in lake core experiment core 6 (+13 days) .... 62
32 Saturation index versus depth for interstitial water from a
typical Station 83 core for calcide (CaCO3) , siderite (FeCO3) ,
rhodochrosite (MnC03) , smithsonite (ZnCC>3) , and cerrussite
(PbC03) 72
33 Saturation index versus depth for interstitial waters from
a typical Station 83 core for vivianite (Fe3 (PO^I 2 ' 8H20),
whitelockite (Ca3 (PC>4) 2) , reddingite (Mn3 (£04)2 ' 3H2O) ,
hydroxylapatite (Cac, (PO^) 3OH) , hydroxylpyromorphyte
(Pb5 (PO4) 3OH) , chlorophyromorphite (Pbs (PC>4) 3C1) ,
a-hopeite (Zn3 (PC>4)2 * 4H2O) , and struvite (MgNH4 PC-4 •
6H20) 73
IX
-------
34 Saturation indices over time for siderite, rhodochrosite,
calcite and magnesite observed in the production experiment . . 76
35 Saturation indices over time for vivianite, hydroxylapatite,
reddingite, whitlockite, fluorapatite, and struvite
observed in the production experiment ............. 77
36 Saturation indices for carbonate and phosphate phases
observed in the homogenized mud core experiment ........ 82
37 Saturation indices for carbonate phases observed in the
lake core experiment ..................... 84
38 Saturation indices for phosphate phases observed in the
lake core experiment ..................... 85
39 Rate of carbonate production (R'ncoP versus t^6 sum of the
rates of calcium (R'Ca), iron (R'pe^ an<* manganese (R'^n)
production for three temperatures ............... 95
40 Phosphorus flux estimates versus iron flux estimates
(direct and indirect) for the homogenized mud core
experiment (with and without worms) and the lake
core experiment ...... ....... ......-•• ........ 109
41 Schematic diagram of the simplified two-compartment model.
F]_ represents natural flux to the lake . F^2 represents
the flux of metal from compartment 1, the lake water
(dissolved and particulate) to the sediments. F2i is the
flux of metal in the opposite direction. F3 represents
export from the lake to the Niagara River and the Welland
Canal. F2 is the anthropogenic flux to the lake . . ..... 116
42 Model predictions of Zn, Pb, and Cu concentrations in
Lake Erie water to the year 2075 ............... 124
43 Model predictions of Zn, Pb, and Cu concentrations in
Lake Erie sediments (Station 83) to the year 2075 ....... 125
44 Water content (%H2O) versus depth at Station 83 ........ . 129
45 K^ as a function of time of sediment deposition at Station 83 . . 130
46 Present day particulate organic nitrogen concentration
(Np) and calculated organic nitrogen concentrations at the
time of deposition (Np0) plotted versus depth and time
of deposition at Station 83 .................. 132
47 Observed and calculated concentration profiles of ammonia
and alkalinity (HCCJ) for the homogenized mud core
experiment (without worms) ................... 137
-------
TABLES
Number page
1 Apparent rates of release of nutrients and metals
observed in the production experiment. All units
are millimoles/1/yr. 29
2 Depth dependencies of rates of production from the
equation R"(Z,T) = R1 (T) exp^r«(T) • A). Units are
R1 (millimoles/1/yr) and a (cm ) 35
3 Temperature dependencies of the rates of production
from the equation R1 (T) = A exp (-E*/RTX R' (T)
(millimoles/1/yr) is the production rate at the sediment
water interface; A (millimoles/1/yr) is the pre-
exponential factor, E* (KCal/mole) is the activation energy;
and R (1.9872 cal/mole/°K) is the gas constant 42
4 Comparision of averaged concentrations (-log molality)
and one standard deviation of major species given in
Nordstrom et al. (1979) and those of our model ASAME 69
5 Solubility constants of potential authigenic mineral phases. ... 70
6 Major ion data used in the mineral equilibrium calculations
for the homogenized mud core experiment 79
7 Major ion data used in the mineral equilibrium calculations
for the lake core experiment 80
8 Stoichiometric model of decomposing Lake Erie sediments 93-94
9 Comparison of calculated solids distribution based ,pn eq.
(12) to observed data. Depth interval 0-7 cm 99
10 Diffusion coefficients used for indirect flux calculations .... 104
11 Direct and indirect flux estimates obtained from the
lake core experiment and-the homogenized mud core
experiment. SRP flux is reported as moles P/m^/day.
*tubificid population present, **zero values not con-
sidered in calculation of mean and variance (before
anoxia) . n.d. =. not determined . 105
XI
-------
12 Indirect flux estimates determined from field measurements
at Station 83 and Station Al. Units are xlO"^ moles/m^/day . . 110
13 Summary of the direct flux calculations for Stations 83 and Al.
Negative fluxes indicate a flux to the sediments. Units
are xlO~6 moles/m2/day .................... Ill
14 Inventory of sources and sinks of heavy metals in Lake Erie . . . 118
15 Values for D, K, R, Co, and Ci used in the time dependent
ammonia and bicarbonate transport-reaction calculations .... 136
xii
-------
Introduction
Lacustrine sediments are known to play an active role in the biogeo-
chemical cycling of materials. Freshwater sediments act as both a source and
a sink for many biologically important materials, notably nutrients, such as
phosphorous, carbon, nitrogen, sulfur, and silicon. Further, sediments are
known to play an important role in resulting cycles of trace metals, radio-
nuclides, and xenobiotics. Because of this, knowledge of the early chemical
diagenesis of sediments, that is those reactions occuring during and after
burial, is essential to an understanding of materials cycling in freshwater
environments.
The present study has focused on the chemistry of nutrient and metal
release during early diagenesis, rates of nutrient and metal regeneration,
rates of materials' movement across the sediment-water interface, and the
effects of bioturbation on materials' cycling.
An experimental approach has been used in this study; the chemistry of
nutrient and metal regeneration has been studied in laboratory microcosms. In
addition, field observations of pore water chemistry and sediment solids
chemistry have been used to verify and test results obtained in the labora-
tory.
-------
Conclusions
Thermodynamic modeling of interstitial waters is a useful technique for
suggesting possible mineralogical controls on trace metals. Application of
such a model to Lake Erie pore waters reveals that iron and manganese carbo-
nates, phosphates, and sulfides are all forming in the sediments as well as
chlorophyromorphite, a lead phosphate. No mineralogical controls for zinc and
cadmium were clearly identified. The inclusion of organic complexes or com-
plexes yet to be discovered in the thermodynamic model will not significantly
improve the results. Additional complexing only serves to lower the ion
activity products, and hence the saturation indices. Predicted supersatura-
tion would be decreased, but not by more than about a factor of two. Under-
saturation of zinc and cadmium phases would increase. Mixed and sulfide
mineral phases are the most likely mineralogical controls on zinc and cadmium.
It is also possible that the controlling reactions could be adsorption or ion
exchange equilibrium. Until more sophisticated techniques for examining
sediment solids are employed and until existing thermodynamic data is critic-
ally compiled and adopted, no further progress on this problem can be made.
Knowledge of the rate of anaerobic decomposition of organic matter and
subsequent release of nutrients and metals to pore waters is essential to an
understanding of early diagenesis and chemical mass transfer in sediments. In
Lake Erie, anaerobic decomposition proceeds via fermentation reactions, pri-
marily methane fermentation. The rates and energentics of these fermentation
reactions are only slightly less than those reported from sediments in which
sulfate reduction is the primary diagenetic pathway.
Both direct and indirect estimates of the flux of ammonia, iron, soluble
reactive phosphorous (SRP), soluble reactive silica (SRS), and bicarbonate
from lake sediments to anoxic overlying water exhibit a high degree of vari-
-------
ability. Further, indirect flux estimates for redox sensitive materials (i.e.
ferrous iron and SRP) may grossly underestimate the actual flux. The initial
flux of iron and phosphorous from sediments to anoxic overlying water is
strongly dependent on conditions at the sediment-water interface prior to
anoxia in the overlying water. Sediments preconditioned by the activities of
tubificid oligochaetes exhibited a higher flux of ammonia, but a lower flux of
iron, SRP, and SRS. The presence of worms had no effect on bicarbonate flux.
The higher flux of ammonia in the presence of worms appeared to be due to a
worm-caused ammonia source in the upper zone of sediment. Reduced fluxes of
iron and phosphorous in the presence of tubificids is most likely due to their
continuous subduction of surficial oxidized material prior to anoxia. The
reduction of SRS flux in the presence of worms cannot be presently explained.
Inclusion of diffusive loss of trace metals from sediments in mass bal-
ance calculations shows that Cu, Pb, and Zn are lost from sediments in roughly
the same molar ratio as they accumulate in sediments. Even if the loading of
Cu, Pb, and Zn to Lake Erie were to increase exponentially for the next 100
years, the concentration of these metals in the lake's waters would increase
by only a factor of three to five. ^
Consideration of organic decomposition in the calculation of anthropo-
genic loading of nitrogen to Lake Erie sediments decreases the estimate of
anthropogenic loading by about a factor of two. Estimates of anthropogenic
loadings of labile materials (carbon, phosphorous, sulfur) to lake sediments
cannot ignore organic decomposition.
A one-dimensional time dependent reaction-transport model which considers
only production, adsorption, and diffusion was found to adequately predict
ammonia and bicarbonate profiles in laboratory microcosms containing homo-
genized Lake Erie sediment and no tubificids. More complex models are
-------
required for other parameters (iron, phosphorous silicon) and situations
(homogenized sediment with tubificids, real lake sediments).
-------
METHODS AND MATERIALS
Field Measurements
Study Area
The data reported in this study were collected from stations 83 and Al in
the central basin of Lake Erie (Fig. 1 ). Station 83 is located in about 16
meters of water. The 1-3 meter thick hypolimnion is often anoxic in August
and September. The sediments are derived from the eastward transport of
sediment from the western basin and from shoreline erosion. They consist
primarily of silt sized quartz with minor amounts of illite, kaolinite, dolo-
mite, magnetite, and feldspar. Cesium-137 and lead-210 profiles at this site
2
yield a sedimentation rate of 0.077 g/cm /yr, and exhibit a biogenically mixed
zone 4.5 cm thick (J.A. Robbins, pers. comm.). Porosity decreases ex-
ponentially from about 90% at the surface to about 60% at 20 cm. The macro-
benthic community at Station 83 consists primarily of tubificid oligochaetes
22 2
(~9500/m ), sphaerid clams (~2000/m ), and chironomid larvae (~1600/m )
(Fisher and Matisoff, in press). These organisms live in the top 5 cm of sedi-
ment and are capable of producing the observed mixed zone (top 4.5 cm) and of
increasing chemical exchange between sediments and lake water (McCall and
Fisher, 1980; Fisher et al., 1980). Station Alls located in about 24 meters
of water. The 3-5 meter thick hypolimnion in often anoxic in August and
September. Sediments at this locale are derived from the eastward transport of
sediment from the western basin and from shoreline erosion. They consist of
approximately equal amouts of silt sized quartz and illite. Minor amounts of
kaolinite, dolomite, magnetite, and feldspar are also present. Sedimentation
rates determined from cores taken in the vicinity of station Al yield sedi-
2
mentation rates of 0.119 g/cm /yr (Kemp, et. al, 1976). Porosity decreases
-------
-43'
i I I I I
83'
82'
80'
i
79'
_j
Figure 1. Field sampling localities in Lake Erie.
-------
linearly with depth from about 95% at the surface to about 85% at 20 cm. the
2
benthos at this site is dominated by tubificid oligochaetes (~4400/m ), sphaerid
2 2
clams (~4000/m ), and chironomid larvae (~100/m ). The biogenically mixed
zone at this site is - 2 to 3 cm in thickness.
Core Collection
Cores were collected in a plastic core liner (cellulose-acetate-
butyrate, 7.3 cm O.D., 7.0 cm I.D.) with a Benthos gravity corer or by SCUBA
divers. The collected core was then capped and held in a vertical position,
measured and described. A 125 ml sample of overlying water was collected
immediately above the sediment-water interface. This water sample was capped
and filtered. The remaining overlying water was then siphoned off. To min-
imize oxygen contact, a plastic bag was taped over the end of the core liner
and nitrogen was continuously blown into the core liner during these mani-
pulations . A hydraulically driven rubber piston was inserted into the bottom
of the core liner and the top of the core liner was placed in a N,-filled
£*
glove bag. Sediment samples were sliced off as the core was pushed up by the
rubber piston. Ten samples were taken at five consecutive 2 cm intervals in
the upper 10 cm, a 3.5 cm layer at about 15 cm, and 5-6 cm intervals at about
20, 30, 40 cm and deeper depending upon the length of the core. Samples were
transferred to nylon squeezers which were covered and clamped into a squeezing
fi P
rack. A 3.4 atm (3.45 xlO dynes/cm ) pressure of N_ acting against a thin
rubber diaphragm ("Dental Dam") (Reeburgh, 1967) forced the interstitial water
through two circles of nylon mesh supports overlain by two 0.45 pm Whatman
Filters and one 0.22(Jm Millipore filter (75 mm diameter). Since the sample
was not exposed to oxygen at any time, the intersitital water closely ap-
proximated the natural system.
-------
Pore Water Peeper
Intersititial waters were collected at Station A-l by both the preceeding
gravity core technique and with a pore water peeper, (originally designed by
Hesslein (1976)). The pore water peeper consisted of a thick, meter long
plexiglass plate with machined slots filled with deionized water and covered
with a polycarbonate membrane filter. The slots were located at 1 cm inter-
vals near the sediment-water interface. The distance between slots increased
as one moves further away from the interface. Diffusion of chemical substances
across the membrane proceeds until the concentration of all species inside the
slots equals their concentrations in the surrounding sediment pore water,
usually 10-30 days. Long-term natural disturbances in the upper part of the
sediment column, such as biotrubation and sediment mixing due to bottom cur-
rents, would create changes in the concentration gradient near the sediment-
water interface; these would be reflected in the peeper data. An accurate
representation of the effects of bottom disturbances could then be assesed in
greater detail.
The peeper was deployed at Station A-l in August, 1979 and retrieved for
a comparison with gravity core data in September, 1979.
Laboratory Experiments
Production Experiment
The chemistry of intersitital waters is driven by the decomposition of
organic matter (Berner, 1974). Bacterially mediated decomposition of organic
matter in recent sediments releases nutrients to interstitial waters (Siever
et al., 1965,- Murray et al., 1978; and others). These regenerated nutrients
undergo diffusion, advection, and chemical reaction. Nutrients escape across
8
-------
the sediment-water interface, undergo ion exchange and adsorption with sedi-
ments, and precipitate as authigenic mineral phases (Nriagu and Dell, 1974;
Holdren et al., 1975; Bricker and Troup, 1975; Emerson and Widmar, 1978;
Rosenfeld, 1979; Suess, 1979; and others). Quantitative description of chem-
ical mass transfer in recent sediments requires knowledge of the kinetics and
rates of material regeneration (Berner, 1974; Lerman, 1975; Aller, 1977).
In nearly all recent organic rich sediments, the so-called oxidized zone,
i.e., the zone in which iron is present as FeOOH, is never more than a few
centimeters thick unless the sediment is actively bioturbated (Siever et al.,
1965; Sholkovitz, 1973; Emerson, 1976, Bricker et al., 1977; Aller, 1977;
Murray et al., 1978; McCall and Fisher, 1980; and others). Thus, a major
portion of the observed sediment column falls within the zones of sulfate
reduction, methane fermentation, and CO reduction (Martens and Berner, 1974;
Reeburgh and Heggie, 1977; Goldhaber et al., 1977.)
Nutrient regeneration in anoxic waters by bacterial sulfate reduction has
been described by Richards (1965):
(CH20)1()6(NH3)16H3P04 + 53SO~2 -» 106CC>2 + 53S~2 + H3PC>4 + 106^0 + 16NH3 (1)
In previous work, rates of organic decomposition in sediments were measured by
the rate of sulfate reduction or ammonium production (Gunkel and Oppenheimer,
1974; Goldhaber e_t al., 1977; Aller, 1977; Rosenfeld, 1977). These workers
incubated sediments in jars under anoxic conditions and analyzed serially
taken samples for sulfate and/or ammonium. Concentration changes in a given
depth interval are used to determine reaction rate and kinetics at that depth.
Sulfate concentrations in Great Lakes' waters (~200 |JM) are much lower than
that of seawater (-27,600 pM) (Weiler and Chawla, 1968; Weiler, 1973; Nriagu,
1975; Tisue, 1980). Consequently, sulfate is not a important electron
-------
acceptor in Great Lakes' sediments. In this regard, the decomposition of or-
gainic matter in Great Lakes sediments is significantly different from that
occuring in marine and estuarine sediments. This report presents the first
experimental results from jar experiments on the rates of release of nutrients
and metals during anaerobic decompositon of freshwater sediments.
Twenty-five gravity cores were collected on September 18, 1978, at Sta-
tion 83 in the central basin of Lake Erie (Fig. 1). No attempt was made to
remove organisms before preparing the sealed jars. The effects of entombment
on pore waters has been discussed by Goldhaber e_t al. (1977).
The cores were sectioned in a helium-filled glovebag. Each of the depth
intervals (0-2, 2-4, 4-6, 6-10, 10-14, 14-18, and 35-39 cm) were pooled in
separate plastic buckets. After all the cores were collected and sectioned,
the mud in each bucket was completely homogenized. Aliquots of sediment from
each depth interval were placed in screw capped glass specimen jars (140 ml).
The jars were sealed with electrical tape and paraffin. The sealed jars were
submerged in mud-filled plastic containers to prevent oxidation and evap-
oration. Jars were stored at 5°C, 10°C, and 18°C, and serially sampled over a
period of approximately 200 days.
Pore waters were expressed under a nitrogen atmosphere using a Reeburgh-
type squeezer (Reeburgh, 1967). Expressed pore waters were analyzed for pH,
soluble reactive phosphorous (SRP), carbonate alkalinity, ammonium, chloride,
ferrous iron, manganese, calcium, and magnesium. Standard analytical methods
were modified for small sample volumes, and all analyses, except calcium, mag-
nesium, and manganese, were performed on the day of collection. The analy-
tical proceedures are given in Appendix I.
10
-------
Flux Experiment
The flux experiment represents part of an attempt to understand in detail
the mechanisms that control the flux of nutrients and other materials from
lake sediments. The experiments conducted were similar to Mortimer's classic
laboratory experiment (Mortimer, 1941; 1942), i.e., sediments and overlying
water were placed in a container; the container was sealed and the overlying
water was allowed to become anoxic; various chemical parameters were measured
as a function of time. In our experiments, however, chemical parameters were
measured not only in the overlying water (as done by Mortimer) but also in the
interstital waters and sediment solids. Thus, flux estimates using both
direct (overlying water concentrations) and indirect (interstital water con-
centration profiles) techniques could be made. Further, data from these
experiments and the production experiment could be used to develop and test
time dependent production-transport models for various chemical species in the
sediment pore waters.
Two series of flux experiments were conducted. In the first series of
experiments (Homogenized Mud Core Experiment), well mixed Lake Erie (Station
83) sediment was used. This was done to eliminate depth variations in the
chemistry of the solid phase and reduce core to core variability. In this
5 2
experiment, half (10) of the cores received a population (10 individuals /m )
of tubificid Oligochaetes so that the effects of their activities (bioturba-
tion, excretion, etc.) on materials release and pore water chemistry could be
investigated. In the second series of experiments (Lake Core Experiment)
intact sediment cores recovered from Station 83 were used. This allowed inves-
tigation of the processes occuring in naturally deposited Lake Erie sediments.
11
-------
Apparatus
Microcosms used in the experiments are illustrated in Fig. 2. These
devices were comprised of four main components: 1) an overlying water res-
ervoir, 2) a sediment chamber, 3) an extruder base, and 4) an extruder piston.
Coupling of components 1,2, and 3 was accomplished by 0-ring (Viton) and
silicon grease sealed threaded connections. The extruder piston was moveable
and a Viton 0-ring was used to provide a seal between the piston and the
sidewalls of the sediment chamber. The minimum wall thickness of any com-
ponent was 0.635 cm. Materials used for construction of the microcosms were
acrylic, PVC, Viton, Teflon, and stainless steel. The contained water and
sediment contacted only non-metallic surfaces.
The overlying water reservoir was a cylindrical acrylic container having
an inside diameter of 25.4 cm and a height of 10.16 cm. The total contained
volume of the overlying water reservoir was ~5 1. This volume corresponds
to a water column ~100 cm in height having the diameter of the sediment cham-
ber. The top of the overlying water reservoir was sealed by an 0-ring fitted
detachable top and sampling cover. The detachable top was secured to a flange
on the overlying water reservoir using eight %-20 x 2" stainless steel cap
screws, while the sampling cover was secured to the detachable top by eight
ij-20 x 2V PVC cap screws. Water within the overlying water reservoir was
constantly stirred using a well-mounted Nalge 6600 series Star Head magnetic
stirbar (2.22 x 1.59cm) driven by a water turbine magnetic stirrer.
The sediment chamber was a cylindrical acrylic container having an inside
diameter of 7.62 cm and a height of 50 cm. The extruder base was a solid
block of acrylic 15.28 x 15.28 x 7.62cm machined to accept the threaded base
of the sediment chamber. The extruder based was fitted with a Swagelock
"Quick-Connect" valve. When a microcosm was sacrificed for analysis, the over-
12
-------
Figure 2. Apparatus used for the flux experiment:
1. Sampling cover 2. Detachable top 3. Overlying
water reservoir 4. Sediment chamber 5. Extruder
piston 6. Extruder base 7. Quick connect valve
8. Turbine magnetic stirrer.
13
-------
lying water reservoir was emptied and removed from the sediment chamber and
extruder base assembly. A pressurized water line was then corrected to the
"Quick-Connect" valve of the extruder base. By admitting water through this
valve, the extruder piston was forced upwards and the desired intervals of
sediment were collected by means of a slicer fitted to the top of the sediment
chamber.
Preparation of Homogenized Mud Cores
In the Homogenized Mud Core Experiment well mixed Lake Erie (Station 83)
sediment was used. This sediment was collected with a grab sampler and sieved
through a 250pm mesh screen to remove benthic macrofauna and large debris.
Before the sediment was placed in the microcosms, it was homogenized by stir-
ring. In all, twenty microcosms were prepared. Tubificids (456 Limnodrilus
spp./microcosm) collected from Cleveland Harbor were added to ten of the
5 2
microcosms to simulate a population density of 10 individuals /m . The re-
maining microcosms were maintained without worms. The tubificids were allowed
to rework the sediments for thirty days before all microcosms were sealed and
helium purged. The concentrations of nutrients and other materials in the
overlying water of each microcosm was monitored regularly. At intervals, two
microcosms (one with worms and one without) were sacrificed to determine pore
water gradients of dissolved materials. The water used to fill the overlying
water reservoirs was deionized.
Preparation of Lake Cores
In the Lake Core Experiment, undisturbed sediment cores were collected by
SCUBA drivers at Station 83. The cores were collected using the sediment
chambers of the microcosm apparatus as coring tubes and the PVC extruder
pistons as bottom caps. In all, twelve cores were collected. The cores were
14
-------
returned to the laboratory, sealed and helium purged. In this experiment, the
water used to fill the overlying water reservoirs was filtered (0.45 pm) Lake
Erie water collected 1m above the bottom at Station 83. Again, the concen-
trations of nutrients and other materials in the overlying water were moni-
tored regularly. Individual microcosms were sacrificed at intervals to deter-
mine pore water gradients of dissolved materials.
In both experiments, the temperature of the microcosms was maintained by
immersing them in a large, specially constructed water bath. During both
experiments, microcosm temperature was maintained between 14° and 16°c.
-Sampling Procedures
Overlying water samples were withdrawn using a syringe. In the Homog-
enized Mud Core Experiment, the sampling cover was removed, and the samples
were withdrawn by inserting a tygon sampling tube into the central part of the
overlying water reservoir. During sampling, a helium atmosphere was main-
tained by enclosing the apparatus in a glove bag. In the Lake Core Experi-
ment, the sampling covers were modified so that helium under a small positive
pressure was admitted to the overlying water reservoir. Admission of the
helium expelled water through a sampling tube into the sampling syringe. With
this arrangement, the sampling cover could remain in place during sampling.
When a core was sacrificed for pore water analysis, the sediment was
gently extruded using the indwelling extruder piston. The desired depth
intervals were collected, homogenized, and placed in Reeburgh type squeezers
(Reeburgh, 1967). Sediment collection and squeezing were carried out using the
techniques described for the field cores. Chemical parameters measured in the
interstitial water were pH, soluble reactive phosphorous (SRP), carbonate
alkalinity, ammonium, chloride, ferrous iron, soluble reactive silica (SRS),
15
-------
manganese, calcium, and magnesium. Standard analytical methods were modified
for small sample volumes, and all analyses except calcium, magnesium, and
manganese, were performed on the day of collection. Analytical proceedures
are given in Appendix I. After the necessary pore fluids were collected,
sediment remaining in the squeezers was collected into plastic bags. These
sediment samples were stored in a frozen condition until they were analyzed.
Results
Field Measurements
The chemistry of sediment solids reflects the chemical depositional
history of sediments and the post-depositional processes affecting them.
Since small changes in sediment solid chemistry can often cause large changes
in pore water chemistry, the chemistry of interstitial fluids is studied to
identify mineral reactions taking place in sediments. Knowledge of both sedi-
ment solid and pore fluid chemistry is required for a complete understanding
of chemical processes taking place in sediments.
Chemical data obtained from both Station 83 and Al is given in Appendix
II.
Sediment solids data includes total C, organic C, Kjeldahl N, total P,
total S, acid volatile S, and total Ca, Mg, Mn, Fe, Zn, and Cu. Pore water
+2 +2 + + +? +? +? +? +? +? -
data includes Fe , Mn , Na , K , Ca , Mg . Zn , Cd , Pb , Cu , HC03,
SRP, SRS, NO-NO Cl , NH , and pH. In addition, pore water concentrations at
station Al obtained by both the gravity coring technique and the pore water
peeper technique are given in Appendix II.
In general, concentrations of nutrients and toxic metals in the sediment
solids decrease with depth (Appendix II, Figs. 3 to 10). This may be due to
16
-------
millimoles Carbon / g dry sediment
C
0
10
1
j= 20
"B.
-------
0
millimoles Carbon / g dry sediment
I 2 3
0
10
20
OL
CD
a 30
40
50
1
Total
6 Organic
A1-6-9-79
Figure 4. Total and organic carbon versus depth at Station Al.
18
-------
0
0 -
Acid Volatile Sulfide
(millimoles S / kg dry sediment)
5 10 15
1 1 1
20
1
10
20
a.
CD
O
30
40
o
O
X)
O
L O
83-4-11-77
Figure 5. Acid volatile sulfide versus depth at Station 83.
19
-------
Total Sulfur
(milIimoles S / kg dry sediment)
10 20 30 40
50
i
83-4-11-77
Figure 6. Total sulfur versus depth at Station 83.
20
-------
Total Kjeldahl Nitrogen
(millimoles N / kg dry sediment)
150300
0
10
20
-g 30
^o
-C
"ci
Q 40
50
60
7n
i i i
\
1 i
.
I
i
i
i
i
i
.
i
83-4-11-77
Figure 7. Total Kjeldahl nitrogen versus depth at Station 83.
21
-------
Total Kjeldahl Nitrogen
(millimoles N / kg dry sediment)
0 100 200 300 400
r~
Or
10
E 20
o
Q.
(D
Q 30
40
A1-6-9-79
50
Figure 8. Total Kjeldahl nitrogen versus depth at Station Al.
22
-------
Total Phosphorous
(millimoles P / kg dry sediment)
0 10 20 30 40 50
f
0
10
20
o
30
a.
-------
0
Total Phosphorous
(millimoles P / kg dry sediment)
10 20 30 40
1 r~
50
T
0
10
E
o
a.
CD
20
30
40
50
H
H
H
H
H
A1-6-9-79
Figure 10. Total phosphorous versus depth at Station Al.
24
-------
increased loading of these materials over time and/or due to post-depositional
chemical mass transfer (see p. 125ff). The concentrations at station 83 of
those species undergoing increased loading or post-depositional alteration
to 5-7 cm generally decrease rapidly to 5-7 cm and are approximately
constant below 7 cm. At station Al, the concentrations increase nearly
linearly from a depth of about 40-50 cm. This would imply that the sedi-
mentation rate at station Al is substantially higher than at station 83.
Calcium and magnesium decrease in the top 10-40 cm at station 83 and
then increase at depths greater than 40 cm. Inorganic carbon (Fig. 3) shows
a similar increase below about 40 cm. This would imply that calcium and
magnesium carbonates are precipitating in the sediments below about 40 cm.
The results of the interstitial water comparison between the gravity
coring technique and the pore water peeper technique are shown in Appendix II
and Figs. 11 and 12. The pore water peeper permits higher resolution around
the sediment-water interface and thus enables a more accurate evaluation of
the concentration gradient at the sediment-water interface. This may re-
present the fact that the pore water peeper collects data that is an average
over the time interval of deployment, while a gravity core collects data that
is an instantaneous representation of the system. Thus storm wave activity
stirring up the bottom is averaged in the peeper case and appears anomalous
2+ +
in gravity coring case. Sampling of reduced species such as Fe and NH
and gases such as CH requires modification of our peeper, since oxidation
and loss to the atmosphere occurred for these species.
25
-------
NJ
CTi
Fe
+2
100
-10-
0-
_ 10-
J 20~
z 30-
a. 40-
UJ
0 50-
60-
70-
80-
90-
SWI
%
o
o
o PEEPER
I CORE
200
100
I
200
300
SRP (JIM)
20 30 40 50 60
SWI
Figure 11. Vertical profiles of ferrous iron, ammonia, and SRP in the interstitial water of
sediments at Station Al determined by both gravity coring and pore water "peeper" technique.
-------
1
o
I
Q.
UJ
O
-10-
0_
10-
20-
30-
40-
50-
60-
70-
80-
90-
Cl' (JIM)
0 500 I0<
i i
o 1
SWI o J
°o°<
\
°)
0 '
o
o
o PEEPER
1 CORE
ALKALINITY (meq/J)
H4Si04
4
1 1
0
i
400
800
i
; 10'6 M
SWI
£_'*-
Figure 12. Vertical profiles of chloride, alkalinity, and SRS in the interstitial water of
sediments at Station Al determined by both gravity coring and pore water "peeper" technique.
-------
Production Experiment
Release Kinetics
Concentrations of all nutrients and metals in the jar pore waters in any
given depth interval increased with time, and the rate of increase was greater
at higher temperatures. Over the course of the experiment, pH remained at
approximately 7.2 (Fisher and Matisoff, in press). Initial review of the data
(Appendix III) indicated that a linear relationship existed between concen-
tration and time. Accordingly, linear regressions were calculated for all
concentration versus time plots. The ammonium data and the derived regression
lines are presented in Fig. 13. The results of the regression analysis are
given in Table 1. Bicarbonate, calcium, iron, and magnesium all display
concentration versus time behavior similar to that shown for ammonium.
Soluble reactive phosphorous was the only parameter studied which displayed a
significant deviation from this trend. The concentrations of SRP from the 2
day and sometimes the 22-24 day 0-2 cm and 2-4 cm depth samples were higher
than those determined at all succeeding times from these depths. These ele-
vated levels of SRP probably result from release of sorbed phosphorous from
dissolving ferric oxyhydroxides. Iron does not show a similar pattern because
of the concurrent dissolution of other iron phases (See Stoichiometric Model).
The reduction in SRP concentration after this initial release is not well
understood, although it probably represents the precipitation of a phosphorous
phase. After 71-73 days, the concentration of phosphorous increased linearly.
The apparent linearity of the concentration versus time plots suggests
zeroth order kinetics for the decomposition:
3P
= R' (Z,T) (2)
28
-------
Temp.
18°C
10°C
5°C
Depth
Interval
0-2cm
2-4
4-6
6-10
10-14
14-18
35-39
0-2cm
2-4
4-6
6-10
10-14
14-18
35-39
0-2cm
2-4
4-6
6-10
10-14
14-18
35-39
K
4.93
2.38
1.38
.323
.137
.483
.373
2.59
1.48
.946
.190
.359
.467
.188
1.16
1.45
.783
.209
.123
.319
.255
HC03
19.8
17.6
15.0
12.3
12.1
7.05
7.45
15.5
11.2
10.1
8.80
10.9
3.54
6.25
7.82
10.6
9.75
4.82
5.08
4.60
7.20
*P04
.045
.055
.075
.148
.030
.003
.033
.062
.035
.049
.099
.038
.006
.041
.064
.062
.067
.037
.022
.007
.026
Ca
4.08
4.49
4.15
3.35
3.55
3-02
1.96
2.80
2.41
2.69
2.53
1.69
1.75
1.59
1.69
2.34
2.47
1.59
1.40
1.25
1.30
Fe++
.992
1.32
.710
.578
.656
.179
.148
.988
.613
.638
.576
.435
.372
.230
.580
.571
.373
.130
.116
.116
.135
Mn
.187
.168
.128
.105
.107
.069
.052
.137
.100
.091
.083
.054
.045
.041
.071
.074
.074
.040
.013
.030
.032
Mg
1.50
1.70
1.42
1.30
1.33
.858
.861
1.15
1.03
.993
1,07
.791
.676
.679
.729
1.14
.977
.731
.573
.562
.464
Table 1. Apparent rates of release, R'(Z,T), of nutrients and metals from the
jar experiments. All units are millimoles/year.
29
-------
• 0-2 CM
0 2-4
6-10
—14-18
« 0-2 CM
0-—2-4
* 6-10
A 14-18
50 100 ISO
TIME (DAYS)
200
50 100 150
TIME (DAYS)
200 0
•—0-2 CM
0--2-4
X 6-10
* 14-18
0.8-
50 100 150 200
TIME (DAYS)
Figure 13. Ammonia concentration data and derived regression line from the production experiment
for three temperatures and four depths.
-------
where C is concentration, t is time, R1 is the apparent rate of release, Z is
depth in the sediment (positive downward), and T is temperature . Integrating
C = R1 (Z,T) * t + constant (3)
The slope of the linear regression line is interpreted as the apparent
rate of release, and the intercept of the regression line represents the pore
water concentration at t=0. The intercepts for any given parameter/depth
interval should not be and are not significantly different among the different
temperatures. Maganese displays the best degree of linearity (r = .85 - .999)
while SRP exhibits the greatest deviation from linearity (r = .5 - .95).
The production rate of ammonia is also directly proportional to the
amount of organic nitrogen in the sediment (Fig. 14). Kjeldahl nitrogen
measures organically bound nitrogen in the trinegative state plus ammonia
nitrogen. It does not include nitrate, nitrite, or organic nitrogen in highly
refractory heterocyclic compounds such as nicotinic acid and pyridine and
certain compounds containing N-N and N-0 linkages. These are all minor frac-
tions of organic nitrogen in sediments (Bremner, 1965). Hence, Kjeldahl
nitrogen is approximately total organic nitrogen. The linear relationship in
Fig. 14 indicates that the overall decomposition kinetics for nitrogen is
first order, and that the observed zeroth order release rates (Fig. 13) are
only an approximation of the overall kinetics. Sequential zeroth order reac-
tions such as microbial degradation often result in an overall reaction that
is first order (Berner, 1974, 1980). A similar relationship exists for or-
ganic carbon.
It must be remembered that the observed rates of release in our experi-
ments (apparent rates) are less than the "true" rates of release (Berner,
1976; Rosenfeld, 1979). Some of the regenerated material may absorb onto
31
-------
O
0>
U)
o:
o
I T
I I
0 100 200 300
KJELDAHL NITROGEN
(mMoles/Kg dry sed)
Figure 14. Ammonia production rate, R , versus total Kjeldahl nitrogen.
4
-------
sediment particles, undergo ion exchange, and/or precipitate as authigenic
mineral phases. The net result is that some of the released material does not
appear as an increase in concentration. The rates reported here are not
corrected for any of these processes.
Depth Dependency
The release rate for most parameters decreased rapidly with depth. This
is illustrated in Fig. 15 where the rates of ammonium release have been plot-
ted as a function of depth for each temperature. Release rates tend to de-
crease exponentially with depth. This depth denpendency is a function of
temperature and may be described by an equation of the form
R' (Z,T) = R'Q(T) e~a(T)*Z (4)
where R1 (T) is the apparent release rate at Z=0 and a is a constant. These
parameters may be estimated by an exponential fit to the R1(Z,T) versus depth
data.
Equation (4) results from the steady-state requirement that the rate of
burial of potentially oxidizable organic matter be equal to the rate at which
it is decomposed (Berner, 1974, 1980). The distribution of oxidizable organic
matter thus decreases exponentially with depth in the sediment (Berner, 1974).
Since production rates depend directly upon the amount of oxidizable organic
matter, they too must decrease exponentially with depth.
Results of the exponential fits to the apparent release rate data are
given in Table 2. Included for comparison in Table 2 are results from similar
studies conducted in estuarine sediments. The greatest difference between the
two environments is the process of sulfate reduction in the estuarine sedi-
ments. Data from Tisue (1980) indicates that sulfate in the Great Lakes is
also undergoing reduction in the sediments. Applying the model of Berner
33
-------
01
OJ
E
o
Q.
(U
(mM/yr)
234
10-
20-
30-
rtO-
',"
I I8°C
RNHU+ (mM/yr) RNHU"*" (mM/yr)
501234 501 2 34 5
-
-
r
T io°c
-
-
i;'
T 5°C
Figure 15. Ammonia production rate at three temperatures as a function of depth.
-------
Fit
Temperature Parameter NH.
22°C R'
a°
22°C R' 23
a° 0
20°C R' 22
a° 0
j
" 18°C R' 6
a° 0
10°C R' 3
a° 0
5°C R' 1
a 0
.405
.8
.419
.90
.342
.31
.259
.60
.149
HC03
IPO.
4
Ca
Mg
_ Location
Fe Mn S0~ Reference
-81. Long Island Sound
0.24 Goldhaber et al (1977)
-129. Long Island Sound
0.33 Aller (1977)
19.2
0.042
13.5
0.038
8.17
0.013
2
0
0
0
0
0
0
0
.41
.410
.796
.315
.266
.205
.189
.198
-96.0 Long Island Sound
0.344 Rosenfeld (1977)
4.44
0.022
2.71
0.019
2.09
0.018
1.60
0.021
1.10
0.018
0.954
0.025
1.25
0.079
0.879
0.053
0.727
0.143
0
0
0
0
0
0
.185
.052
.130
.059
.082
.069
Lake
This
Lake
This
Lake
This
Erie
Study
Erie
Study
Erie
Study
Table 2. Depth dependencies of rates of production from the equation R'(Z,T)
exp (-a(T)Z). Units are R (xlO~3 raoles/yr) and a (cm"1).
-------
—8 —1
(1974) to his data, one may calculate K ^ 9 x 10 sec for Lake Erie
s
-11 -1 -9
sediments. Berner (1974) calculates K of 7.5 x 10 sec and 5.6 x 10
S
sec for Santa Barbara Basin and Sommes Sound sediments, respectively.
Although sulfate reduction is more rapid in Lake Erie sediments than in these
two environments, the small quantities of sulfate do not enable sulfate re-
duction to be a quantitatively major process of decomposition in the sedi-
ments. Hence, methane fermentation would be expected to be the dominant
reaction in Lake Erie sediments. Production rates for ammonium and phos-
phorous are only about a factor of three less in Lake Erie sediments. This is
not particularly surprising since Stumm and Morgan (1970) indicate that sul-
fate reduction is only slightly more energetic than methane fermentation.
The behavior of ammonium is well described by equation (4). This is not
true for all of the parameters. Phosphorous displays a suppressed release
rate in surficial sediments. As suggested earlier, this may be the result of
the precipitation of a phosphorous phase. The large increases in calcium and
ferrous iron in surface samples probably supresses the apparent production of
phosphorous via precipitation of whitlockite, apatites, and/or vivianite
(Nriagu and Dell, 1974; Troup, 1974; Matisoff et al., ms). Other species,
such as bicarbonate, have release rates which are greater than zero at depth.
The depth dependency of bicarbonate release is better described by
R'(Z,T) = R^(T) + RLed(T)e~a(T)*Z (5)
where R'(T) is the apparent rate of release at Z=°°, and R'(T) + R1 ,(T) =
oo oo sect
R'(T) = the apparent production rate at Z=0. At depth, diffusive transport is
small for short time intervals. Consequently, the contribution of R'(T) will
result in a large, continuous increase in concentration with depth. This is
not observed. Accordingly, the true R^(T) must be a very small positive
number. The higher values observed in the jar experiments may be an experi-
mental artifact (Goldhaber e_t al., 1977). A small amount of oxidation of
36
-------
sulfide minerals during preparation of the jars could easily generate an
initial excess bicarbonate release by sulfate reduction upon sealing the jars.
This would appear as a positive value of R^(T). Hence, eq. (4) will be used
to represent the experimental data.
Regeneration of metals during the decomposition experiment is interpreted
as pH buffering mechanisms resulting from the dissolution of carbonate and
sulfide phases. Calcite, aragonite, and dolomite have been clearly identified
in these sediments by optical and x-ray diffraction studies. A substantial
acid volatile sulfide phase is also present. As shown in Fig. 5, the amount
of acid volatile sulfide decreases rapidly with depth. The pattern is prob-
ably the result of conversion of acid soluble monosulfide phases to less
reactive disulfides as shown by Bernsr (1970). Although FeCO- and MnC03 might
also be dissolving, because of the J/igh levels of acid volatile sulfide in the
surface samples, we have interpreted the production of ferrous iron and maga-
nese in these experiments resulting from dissolution of mackinawite (FeS) and
alabanite (MnS) in response to H production during decomposition (Pankow and
Morgan, 1980). Accordingly, the production rates of iron and manganese should
follow the distribution of acid volatile sulfide, and they do (Appendix II)
The profiles of total and organic carbon at the study site are shown in
Fig. 3 . Organic carbon comprises 75% - 100% of the carbon present. Carbon
content decreases exponentially with depth to a constant value of approxi-
_2
mately 1.2% (~1 x 10 mole/gram dry sediment) at a depth of 7 cm. The term
oxidizable carbon will refer to total carbon less the observed 1.2% constant
at depth. Under a steady-state approach, we conclude that below 7 cm the
decomposition reactions are probably incomplete and proceeding at a much
slower rate than those above this depth. Production rates of metals and
nutrients follow the distriubtion of oxidizable carbon (Fig. 3; Tables 1
37
-------
and 2). The distribution of bacterial abundance (Fig. 16) at this site is
also correlated with the vertical profile of oxidizable carbon. The coin-
cidence of these two profiles is not due to the carbon content of the bacteria
—8 -7
population. Assuming a weight of 10 to 10 (jgC/bacterium (Bancroft et al.,
1976), the bacteria themselves account for only 0.1-1% of the organic carbon
found in the sediment. The depth profile of bacterial abundance probably also
reflects the profile of metabolic activity of the bacterial population at this
site. Tobin and Anthony (1978) used tritiated thymidine incorporation to
measure DNA synthesis in a sediment core from this part of Lake Erie and took
DNA synthesis as an indicator of cell proliferation and microbial activity.
They found that microbial activity was highest in the top 7 cm of the core,
decreasing exponentially by about a factor to ten from the sediment-water
interface to a depth of 7 cm. This is similar to the pattern we report here
for bacterial number. They also found a high correlation of activity with
bacterial biomass and cell numbers.
The observed decreases with depth for carbon and bacteria are sharper
than those of the production rates. In part, this may be the result of sampling
procedures. Total carbon, organic carbon, and bacterial abundance were all
determined at 1 cm intervals on single cores. In contrast, production rates
were determined over large intervals on homogenized sediment sections obtained
from 25 different cores. Homogenization might tend to diminish the sharpness
of depth variation.
Sediment homogenization, however, cannot account for the weak exponential
decrease in calcium production. Calcium is released to the pore waters by a
two-stage process. Hydrogen ion and organic acids are produced by the pro-
posed decomposition reaction. Calcium is released when the excess acid is
38
-------
Bacteria Cells
Q
(10 cells/gm dry sediment)
0 5 10 15 20 25 30
i i i i i i
10-
20-
30-
i
i
i
Depth (cm)
Figure 16. Bacterial abundance as a function of depth at Station 83.
39
-------
neutralized by dissolution of calcuim-bearing carbonate phases. Thus, mineral
equilibria, ion exchange, and adsorption all tend to modify the effects of the
decomposition reactions. When coupled with the result that the composition of
the sediment and the decomposition reaction rates are approximately uniform
below 7 cm, it is reasonable that some of the parameters have rate distri-
butions that are only partially described by equation (4).
It is interesting to note that the profiles of some parameters such as
210
Pb (Fig. 17) show a mixed zone 4.5 cm thick, while other parameters such as
organic carbon and total Kjeldahl nitrogen do not indicate mixing. The reason
that the two observations are still consistent lies in the relative rates of
210
mixing, decomposition, and loading of the parameters. Pb has a 22 year
half life and has a constant supply to the sediment while loading of organic
carbon has increased by about a factor of two since 1947. Apparently, the
increased loading of organic carbon swamps the biogenic mixing signal.
Temperature Dependency
The rates of release of all parameters studied increased with increasing
temperature. The temperature dependency of a single chemical reaction may be
described by an Arrhenius rate law. The release of nutrients and metals
during decomposition is clearly the result of numerous biotic and abiotic
reactions, both partial and complete. Nevertheless, the temperature data may
be characterized by the Arrhenius function, "apparent activation energies"
calculated, and some information about the kinetics of reactions extracted.
Table 3 gives the results of applying the Arrhenius rate law to the production
rate data.
The apparent activiation energy for ammonium production is slightly
larger than the 16.6 KCal/mole value calculated from ammonium data given by
Rosenfeld (1977) and is only slightly less than those reported for sulfate
40
-------
Excess Pb-210 (dpm/g)
5
10
10
^o
JC
"a.
Q
20
30
Figure 17. Excess Pb-210 versus depth at Station 83.
41
-------
++
A
o
E*
NH4 HC03
32.78 20.79
17.82 10.29
P°4
31.13
18.17
Ca
17.71
9.38
Mg
11.69
6.50
Fe
11.88
6.74
Mn
15.34
9.83
Table 3. Temperature dependencies of the rate of production from the equation
R'(T) = A exp (-E*/RT). R'(T) (xlO~3 moles/yr) is the production rate at the
sediment-water interface,- A (xlO moles/yr) is the pre-exponential factor;
E* (KCal/mole) is the activation energy; and R (=1.9872 cal/mole • °K) is the
gas constant.
42
-------
reduction (Vosjan (1974), 22 KCal/mole; and Rosenfeld (1977), 24.6 KCal/mole).
This suggests that bacterial energetics in Lake Erie sediments are comparable
to those of estuarine and marine sediments. This agrees well with our pre-
vious conclusion that the energetics of methane fermantation are only slightly
less than those of sulfate reduction. The lower apparent activation energies
for bicarbonate, calcium, magnesium, ferrous iron, and manganese suggest that
these species may also be involved in chemical reactions, such as mineral
dissolution, that are much less energetic. The high activation energy deter-
mined for phosphate is based upon depth dependencies which assume no loss of
phosphate to mineral reactions in the surface sediments. The agreement in
activation energies between ammonium and phosphate supports this assumption for
phosphate .
If the effects of adsorption are considered, the rates obtained are
greater than the observed "apparent rates" (Rosenfeld, 1979):
RT(Z,T)
where R (Z,T) is the total or true release rate. Using a value of K ~l-2 for
ammonium (Rosenfeld, 1979), the calculated true rates of release are a factor
of 2-3 greater than the reported apparent rates, but the activation energy
remains essentially unchanged. Thus, adsorption does not affect the previous
conclusion that of the species studied only ammonium does not significantly
enter into mineral dissolution or precipitation reactions.
43
-------
Flux Experiment
Overlying Water Chemistry
Homogenized Mud Core Experiment Typical concentration versus time plots
for ammonium, alkalinity, pH, SRP, SRS, and ferrous iron in the homogenized
mud experiment (with and without worm cases) are given in Fig. 18-. On the
whole, changes observed in the overlying water are those expected from the
work of Mortimer (1941; 1942). The overlying water, which was initially
deionized, quickly reached an apparent steady state composition with respect
to both SRP and alkalinity. After the microcosms were sealed and purged with
helium (+44 days), alkalinity in the overlying water increased markedly.
Alkalinity reached somewhat higher values in microcosms with tubificid popu-
lations. Immediately prior to (or concurrent with) the rapid increase in
ferrous iron in the overlying water, alkalinity decreased slightly. The exact
time at which dissolved oxygen in the overlying water was totally depleted was
uncertain. Electrode measurements of dissolved oxygen indicated no dissolved
oxygen present prior to other chemical changes (e.g. increase in ferrous iron
concentration) known to accompany anoxia. Consequently, the presence of
detectable level of ferrous iron was taken to indicate anoxia in the overlying
water. Ammonia was first detected in the overlying water after the microcosms
were sealed and helium purged. In general, the concentration of ammonia
increased with time. Microcosms with tubificid populations exhibited higher
ammonia concentrations compared to that of those microcosms lacking a worm
population. Silica concentrations increased with time. The pattern of silica
increase, however, appeared to be unrelated to anoxia or the presence of
tubificids. In contrast, Mortimer (1941) observed a large increase in silica
44
-------
Sfr
pue
q.noqq.TM-1 'spTOTj-cqnq.
pnui pazyueBouioq jo JCS^BM 6uTA-[j3AO aqq. UT yd pire SHS
'UOJT snojcaaj '^TUOUIUIB jo uoT^BJ^ueouoo "Q-[
Q IRON pM (o)
0 AMMONIA /xM (A) |QQ
20
20
Q AMMONIA
(A) |QQ
Q IRON ^M (o)
20
0 ALK mM (V)
0 ALK mM (v)
Q SRP
(•)
SRP
(•)
SILICATE
200
SILICATE M (») 200
6 PH (*)
8
(*)
8
O
en
O
m
o
Is
en
O
-------
concentration concurrent with anoxia. Ferrous iron remained at low levels
(i.e. below detection) then sharply increased in concentration. Microcosms
lacking a tubificid population tended to experience a sharper initial rise in
ferrous iron concentration compared to those microcosms having a tubificid
population. The concentration of SRP increased after the microcosms were
sealed. The pattern of SRP concentration increase, however, did not neces-
sairly mirror that of ferrous iron. Often, SRP concentration increased sharply
prior to the detection of ferrous iron in the overlying water. The pH of the
overlying water exhibits an initial increase (first 20 days). After this, pH
appeared to decrease. Reasonably sharp pH decreases occurred concurrently with
the initial rise in ferrous iron concentration .
Lake Core Experiment A typical concentration versus time plot for
ammonium, alkalinity, pH, SRP, SRS, and ferrous iron in the lake core
experiment is given in Fig. 19. Again, the observed chemical changes are
similar to those described by Mortimer (1941; 1942). After sealing, ammonium
increases sharply. Alkalinity tends to increase, but not to the same extent
as seen in the Homogenized Mud Experiments. Silica displays a strong increase
over time. Again, the pattern of silica concentration increase appears to
bear no relationship to anoxia (i.e. presence of ferrous iron). Ferrous iron
displays a pattern of sharp concentration increase. SRP concentration changes
appear to coincide with the increases and decreases of ferrous iron concen-
tration. Again, pH of the overlying water tends to decrease with time.
46
-------
o
o
9=
CO
O
tr
cr
LU
u_
E
>-
Q_
cr
CO
OLOLOLOLOLU>
UJ
CO
•*
X
a.
CORE 4
0
60
TIME (DAYS)
120
Figure 19. Concentration of ammonia, ferrous ironf alkalinity, SRP, and SRS and
pH in the overlying water of a lake core experiment core. CLCE-4) .
-------
Interstitial Water Chemistry
Homogenized Mud Core Experiment
Interstitial water profiles of ammonia, ferrous iron, alkalinity, SRP
and SRS determined in the homogenized mud experiment for microcosms lacking a
tubificid population are shown in Fig. 20 . Chemical profiles from microcosms
with a tubificid population are shown in Fig. 21 . (Numerical data are given
in Appendix IV).
Ammonium. In microcosms lacking worms, ammonia concentration increased
exponentially to a depth of ~10cm then remained constant. In these micro-
cosms, ammonia profiles did not exhibit a strong time dependency. Ammonia
concentration at depth did tend to increase slightly with time. The absolute
value of ammonia concentration at depth is similar to that observed in cores
collected at Station 83. In microcosms with a tubificid population, the
observed ammonia profiles are strikingly different. Ammonia concentrations
exhibit far less depth variation and there is some evidence that a subsurface
source of ammonia is present. The ammonia concentration at depth is similar
to that observed in Lake Erie cores (station 83).
Ferrous Iron. In both with and without worm microcosms, ferrous iron con-
centration profiles show an increasing trend with depth, and there is no consis-
tent evidence for time dependency. Ferrous iron profiles from microcosms with
tubificids are somewhat more chaotic than those from microcosms without worms.
Alkalinity. Depth profiles of alkalinity from both with and without worm
microcosms are very similar. Alkalinity profiles exhibit an exponential
increase with depth to ~10cm then increase monotonically with depth. Pro-
files of alkalinity do exhibit substantial evidence of time dependency.
. 48
-------
Alk (meq/t) SRP (/tM) H4Si04(MM)
c
o-
5-
10-
15-
20-
25 J
0-
5-
10-
15-
20-
25 -^
5-
g 15-
~ 20-
x 25-1
Q.
U
5-
10-
15-
20-
25J
0 -i
5-
10-
15-
20 -
25 J
0 -i
5-
10-
15-
20-
25J
o o
10 O
3 CM 10
1 1
"•i
I
1
1 1 1
f\
1
1
f\
1
1
\
1
1
1 1
r\
1
1
1 ,
' ','
'l
1
1
(
_
-
"
-
-
-
-
_
_
-
-
.
_
-
-
.
-
-
-
.
1
.
.
-
-
_
8 o
D - CO
1 1
Y
i
I
1 i i
''',
l
l
i i i
Mi
i
l
l
I
i i
V
l
l
, ,
•^
*»(
\
l
l
c
-
—
-
-
J
-
-
-
-
-
-
.
..
-
-
-
_
-
-
-
-
1
-1
—
-
-
-
-
3 ,0 9
1 |
\'\
1
1
1 |
"\ '
|
1
i i
"\
I
1
1 1
*•
1
1
i i
'\
1
1
1 l
' 's
1 '
1
1
(
-
J
-
-
J
-
-
.
_
-
-
»
.
-
-
J
-
-
-
-
-1
—
-
-
-
-
o
IO IO
3 1^ —
1 1 1
• X
'l
1
1 1 1
f
1
1
1 1 1
v
1 •;
i
i
i
i
i i
' ';•. •
i
i
, ,
• y
i
'i
1
1
c
-
-
-
-
J
-
.
_
_
-
-
^
_
-
-
_
-
-
_
_
-1
-
-
-
-
_
o 8
> p 2
1 f
• \
i
f
72 DAYS
, ,
' \
1
1
99 DAYS
1 f
'" 1
.
•
1
120 DAYS
i i
• * .
1
1
133 DAYS
i i
' "\
1
1
147 DAYS
i i
• • .
\
'
1
1
169 DAYS
Figure 20. Vertical profiles of ammonia, ferrous iron,
alkalinity, SRP, and SRS in the homogenized mud experi-
ment cores without worms.
49
-------
NH4+(/iM) Fe2+(/iM) Alk (meq/I) SRP (/iM) H4Si04(/iM)
88
o cj mo
E
o
I
CL
Ul
O
f- - O t- ^
t 1 I
O-i
5-
10-
15-
20-
25J
' '\
'',
i i i
O-i
5-
10-
15-
20-
25J
t
}
1
1
1 • '
O-i
5-
15-
25-
|'l
1
1
1 1 1
O-i
5-
10-
15-
25-
i
/
1
1
! i i
0-]
5-
15-
20-
25 -1
"\
1
1
I ' *
O-i
5-
10-
15-
25-
?
1
1
1 t 1
«
"
•
.
"'j
1,
i i i
•
-
.
J
• i
\
|
1
i • '
•
•
_
•<;
i
i
i > >
-
;
-
.
,
'',
1
1
i > '
-
-
.
.
1
1
' >
•
•
.
f \
1
1
.
"
•
j
A
1
i i i
i
;
•
.
-
. .
1
I
1
i i i
-
•
.
'\
1
1
i i i
„
;
-
.
. ,(
\
I
1
i i i
-
-
.
.
"\
i
1
i « >
•
-
„
1
1
.
.
-
.
r' '
1
1 ' '
i
;
•
.
.
i,
)'
1
1
t • '
-
•
.
'(
1
1
. , ,
~
m
-
.
,
|
1
i • <
-
-
.
.
\
I
1
t * *
-
-
_
'\
1
1
.
:
-
.
1
71 DAYS
i i i
-i
;
-
_
j
4,
'\
I
1
100 DAYS
i i i
-
-
.
\
1
1
120 DAYS
i i i
-
_
-
_
'l
1
1
1
134 DAYS
i i i
-
.
J
\(i
1
1
148 DAYS
F l I
•
•
.
' "'!
i
i
169 DAYS
Figure 21. Vertical profiles of ammonia, ferrous iron,
alkalinity, SRP, and SRS in the homogenized mud experiment
cores with worms.
50
-------
Profile curvature changes with time and the alkalinity at depth increases with
time.
SRP. Profiles of SRP show no consistent trend with either depth or
time for either with or without worm cases.
SRS. Silica concentrations in microcosms both with and without tubificids
generally exhibit a decrease with depth to a depth of ~ 5cm and thereafter
maintain a fairly constant concentration. Silica concentration at depth tends
to increase with time in both with and without worm casas.
Lake Core Experiment
Interstitial water profiles of ammonia, ferrous iron, alkalinity, SRP,
and SRS obtained in the lake core experiment are given in Fig. 22. (Numerical
data are given in Appendix IV).
Ammonia. Ammonia profiles observed in the lake cores exhibit a subsur-
face maximum indicating subsurface production of ammonia. Below the subsur-
face maximum, ammonia concentrations are fairly constant with both depth and
time. The lake core experiment ammonia profiles are similar to the homogen-
ized mud core (with worms) ammonia profiles, but show greater consistency.
Ferrous Iron. Depth profiles of ferrous iron obtained from the lake
core experiment exhibit a subsurface concentration maximum with increases
in magnitude with time. Ferrous iron concentrations at depth are constant
*
over time.
51
-------
Fe^ (/iM)
0 0
Alk(meq/i) SRP(/iM) H4Si04(/iM)
o Q o
o _ «n m
O cu in
l i i
O-i
5-
10-
15-
20-
25-J
• • .
,i "•
i
1
1
i i i
O-i
5-
10-
15-
20-
' 25-1
' I
,»
i
1
1
i i i
Q —
e 5-
0 J
~- 10-
H I5~
0 25-
. ,
?
1
1
1 i i
O-i
5-
10-
15-
20-
25-"
• «,(
.1
1
1
1
1 i l
O-i
5-
10-
15-
20-
25-I
,
,''
'
1
1
o -- <*>
ill
-i
_
-
-
_
-
,i
('
1
1
i i i
-
-
-
-
-
-
•s
, I1'
1
1
1
i i i
_
-
-
-
_
,
,1 ''
1
1
1
i i i
-
-
-
-
_
_
• l
t
1
1
1
i i i
..
-
-
-
—
-
. ,
. •• ' '
'
i
1
o «n -
i i i
-i
_
-
-
_
-
••,,
,'
|
1
1 '
i i i
-i
-
-
-
-
-
•-,
S|
1
1
1
i i i
_
-
-
—
_
1,
',
1
1
i i i
-
-
-
-
_
_
"'•l
«
'
1
1
t i i
_
-
-
-
—
-
• i
\
l
1
1
o ^ —
1 1 1
-1
.
-
-
.
-
•• .,
,•'
1
1
1
1 1 1
-1
-
-
-
-
J
• - ;
, • ''
1
1
1
1 1 1
—
-
-
-
_
.,
, • '•
1
1
1
1 1 1
-
-
-
-
•
„
• 1 (
•l1
1
1
1
1
1 1 1
_
-
-
-
-
-
. .
, .'
l
•'
o ^ —
iii
-i
-
-
-
-
-
' • .
i1
1
1
1
13 DAYS
i i i
-
-
-
-
-
-J
• • \
.•''
'
1
1
29 DAYS
i i i
_
-
-
—
_
. ,
'l
. 1
1
1
1
66 DAYS
i i i
-i
-
-
-
-
_
• ' ,
f
1
1
84 DAYS
i i i
..
-
-
-
-
-
• 1
,.•'
1
1
1
92 DAYS
Figure 22. Vertical profiles of ammonia, ferrous iron, alkalinity,
SRP, and SRS in the lake core experiment cores.
52
-------
Alkalinity. Vertical profiles of alkalinity also show a subsurface maximum
which increases with time. Alkalinity at depth in the cores is constant with
time. The location of the alkalinity maximum coincides with the depth at which
ferrous iron reaches a constant value.
SRP. Vertical profiles of SRP appear to follow those of ferrous iron.
The location of the SRP subsurface maximum coincides with that of the ferrous
iron, and SRP concentration is reasonably constant at depth in all cores. The
magnitude of the SRP maximum increases up to 66 days,then decreases. This is
in contrast to the ferrous iron maximum which consistenly increases.
SRS . As in the cases of the other parameters, silica profiles also
exhibit a subsurface maximum. The vertical location of the maximum is coinci-
dent with those of ferrous iron and SRP. Unlike the other parameters, silica
concentrations tend to increase with both depth and time below the maximum.
Chemistry of the Sediment Solids
Results of chemical analysis of the sediment solids from two homogenized
mud core experiment cores, HMC-R and HMC-T (both without worms, +71 days and
+169 days after settling in the microcosms, respectively), and one lake core
experiment core LCE-6 (+13 days after collection) are presented in
Figures 23-31 (numerical data is given in Appendix IV). The measured para-
meters include total carbon, organic carbon, Kjeldahl nitrogen, total phos-
phorous, and total calcium, magnesium, manganese, zinc, and copper.
53
-------
millimoles Carbon / g dry sediment
01 2345
0
5
10
1
jz 15
"a.
-------
millimoles Carbon / g dry sediment
012345
1 1 1 1
Or
10
l5
CL
-------
millimoles Carbon / g dry sediment
0 I 2 34 5
1 1 1 1 1 1
0
5
10
1
£ l5
"QL
o>
Q
20
25
•*n
0 1
0 1
0 1
- y
o
p
- 1
0
I
1
__
j Total
1 O Organic
O
- 1
CORE 6
Figure 25 . Total and organic carbon in lake core 6 (+13 days)
56
-------
Total Kjeldahl Nitrogen
(millimoles N / kg dry sediment)
0 125 250
Or
10
o
Q.
Q
15
20
25
30 L
I.
CORE R
Figure 26. Total Kjeldahl nitrogen in homogenized mud core
R (+72 days).
57
-------
Total Kjeldahl Nitrogen
(miHi moles N / kg dry sediment)
0
Or
10
E
o
Q.
a>
0
15
20
25
30
125
250
CORE T
Figure 27. Total Kjeldahl nitrogen in homogenized mud core
T (+169 days).
58
-------
Total Kjeldahl Nitrogen
(millimoles N / kg dry sediment)
0
125
250
Or
10
15
Q.
Q
20
25
CORE 6
Figure 28. Total Kjeldahl nitrogen in lake core experiment
core 6 (+13 days).
59
-------
Oi-
10
20
a.
o>
Q
30
40
Total Phosphorous
(millimoles P / kg dry sediment)
10 20 30 40
—i 1 1 r~
H
50
~i
CORE R
Figure 29. Total phosphorous in homogenized mud core R (+72 daysl,
60
-------
0
r~
Or
10
E
o
Q.
CD
O
20
30
40
Total Phosphorous
(millimoles P / kg dry sediment)
10 20 30 40
50
CORE T
Figure 30. Total phosphorous in homogenized mud core T (+169 days).
61
-------
Total Phosphorous
(millimoles P / kg dry sediment)
0 10 20 30 40 50
i i i ii i
0
10
1
£ 20
"a.
0)
Q
30
d.n
,
H
H
—
H
—
CORE 6
Figure 31. Total phosphorous in lake core experiment core 6 (+13 days)
62
-------
Homogenized Mud Core Experiment
In the homogenized mud cores, there is no depth variation in total car-
bon, organic carbon, Kjeldahl nitrogen, total phosphorous, calcium, mag-
nesium, or iron. This result indicates that sediments in the homogenized mud
core experiment were well mixed. Vertical profiles of total manganese, zinc,
and, to some extent, copper exhibit significant departures from homogeneity.
Solid phase manganese shows an exponential decrease in abundance moving from
the sediment surface to depth, while both zinc and copper have surface and
subsurface maximum. These patterns are observed in both HMC-R and HMC-T. The
only possible explanation of such patterns is the mobilization of those metals
and their precipitation as authigenic minerals. The specific mineralogy of
these precipitates could not be identified, and results from the mineral
equilibria calculations did not indicate a large degree of supersaturation for
any pure carbonate or phosphate phase for manganese, zinc, or copper at the
depths their solid phase maxima are observed.
Lake Core Experiment
Solid phase analyses of the lake core experiment core, LCE-6 show no
significant departure from other observations made for cores at station 83.
The suite of cores collected at station 83 for the lake core experiment are
representative of the locale.
Comparison of Homogenized Mud Cores to Lake Cores
Comparison of solid phase data for the homogenized mud cores and the lake
cores shows that the homogenized mud core sediments are significantly lower in
water content and exhibit a smaller gradient in porosity change with depth.
The homogenized mud cores vary in water content from ~60% at the sediment-
63
-------
water interface to ^52% at 22-26 cm. In contrast, the lake cores vary in
water content from ^86% at the sediment-water interface to ^56% at 22-26 cm.
Examination of the total carbon, organic carbon, and Kjedahl nitrogen data
indicates that the homogenized mud core sediments are chemically similar
(gross comparison) to sediments from a depth of 10 to 15 cm.
Discussion
Mineral Equilibria
Many trace metals can be toxic if their concentrations reach sufficiently
high levels in living plants and organisms. It is necessary to understand
their behavior and transport to determine the magnitude of their threat in a
specific environment. In order to understand chemical behavior, concentra-
tions and reactivities must be determined in the different physical compart-
ments—overlying water, sediments, and interstitial waters --before exchange
between the systems can be investigated. In this section, the interstitial
waters in Lake Erie sediments were investigated in an effort to identify
post-depositional mineral equilibria of calcium, magnesium, maganese, iron,
zinc, cadmium and lead carbonate and phosphate minerals.
The formation of these mineral phases results in an increased removal and
storage of trace metals and phosphate pollutants in sediments. Thus the
indentification of authigenic mineral phases increases our understanding of
the lake's response to chemical stress and aids managers in quantitatively
evaluating pollutant loading rates.
Little work has been done on the interstitial water chemistry of zinc,
cadmium, and lead. A large body of data has accumlated describing their
distribution in solid sediments (Wheeler and Duriming, 1976; Presley, et al.,
64
-------
1972; Semkin and Kramer, 1976; Jenne, 1968) and surface waters (Chawla and
Chau, 1969; Zirino and Healy, 1970; Riley and Taylor, 1972; Bachmann, 1963;
Bradford, 1972; O'Connor and Renn, 1964; Hem, 1972), but these studies were
not able to elucidate their chemical behavior in the sediment-interstitial
water system. Several investigators have attempted to determine the control-
ling mechanisms responsible for the distribution of zinc, cadmium and lead by
analyzing their concentrations in various aqueous and solid fractions in one
system (Walters et al., 1974; O'Connor and Renn, 1964; Bachmann, 1963; Cline
and Upchurch, 1973; Gardiner, 1974a, 1974b). These studies have made headway
towards understanding the relative reactivities of metals with organics and
their adsorption on ferric hydroxides and clay mineral surfaces. Such meth-
ods, however, could not identify individual mineral phases. Therefore, these
studies only conclude that the metals were complexed by other dissolved species
or were in equilibrium with an unidentified mineral phase (Presley et al.,
1972). Analysis of metals in the sediment fraction alone cannot be used to
discern processes controlling their distribution.
More recently, researchers have studied interstitial water chemistry
(Muller, 1969; Brooks et al., 1968; Troup 1974; Duchart et al., 1973) in the
hope of understanding the processes controlling the behavior of the metals in
sediments. The interstitial water is the medium connecting the solid phases
in the sediments and the soluble species in the overlying water. Because any
change in metal behavior will alter the chemical composition of the pore
waters, pore water chemistry is a sensitive indicator of chemical reactions
and equilibria between solid phases and dissolved species (Berner, 1971).
The residence time of metals in interstitial water is greater than in overly-
ing water. Consequently, metals in pore waters are more apt to approach a
state of equilibrium with solid phases in the sediments. Thus, the chemical
65
-------
composition of the interstitial waters can be used to identify and study
mineral-water equilibrium reactions in sediments. Once these reactions are
determined, chemical mass transport can be studied.
A thermodynamic equilibria approach has been used successfully to study
the behavior of ferrous iron in pore waters (Troup, 1974; Bray, 1973;
Nriagu, 1972a; Emerson, 1976). Siderite (FeC03), vivianite (Fe3(P04)2. 8H20)
and mackinawite (Fe S) were found to be supersaturated in many intersitital
A"rX
waters, and their presence has been confirmed by x-ray diffraction studies.
This supports the conclusion that these minerals control the dissolved ion
centrations of ferrous iron, carbonate, and phosphate, and that thermodynamic
studies of the interstitial waters may be used to predict the controlling
mineral equilibria. Lu and Chen (1977) examined the seawater overlying sedi-
ments for release of trace metals under oxidizing and reducing conditions.
They concluded that solid phase sulfides, carbonates, and silicates were
controlling trace metal behavior.
Since zinc, cadmium, lead, and manganese share some chemical properties
with ferrous iron, a thermodynamic approach may be applicable to the study of
these trace metals. These metals (like ferrous iron) may form relatively
insoluble metal carbonates and/or phosphates under conditions found in anoxic
lake sediments. Hence, the thermodynamic approach that has been used study
ferrous iron behavior was followed in this study.
Calculation of Activities
The ionic strength, activity coefficients, and activities of calcium,
magnesium, ferrous iron, manganese, zinc, cadmium, and lead were calculated by
a computer program, ASAME ( Aquatic Speciation And Mineral Equilibria), to
determine the speciation of the metals and the ion activity products of the
metal carbonates and phosphates. The computer program developed in this study
66
-------
is based on procedures outlined by Garrels and Thompson (1962) and is of the
'continued fraction1 type discussed in Nordstrom et al. (1979). In this
study, all the measured species (Na , Ca , Mg , K , Zn, Cd, Pb, Fe , Mn ,
NH4+, CL~, NO ~ ZP04/ and HCC>3~ and estimated values for SO^Weiler, 1973))
were used to calculate the initial ionic strength, I, of each pore water
sample:
I = | I n± Z\ (7)
where m. is the measured analytical concentration and Z. is the charge of a
species i. Activity coefficients were then calculated from the Debye-Hiickel,
the Guntelberg, or the Davies equations, according to the ionic strength
limits given in Stumm and Morgan (1970).
The details of the calculation procedure are given in Garrels and Thompson,
(1962), and Nordstrom et al. (1979). Briefly, the activities of the free
ions were calculated by a successive approximation. A mass balance equation
is written for each analyzed species. Each measured concentration (e.g. total) is
equal to the sum of the free ion concentration and the concentrations of all
ion pairs containing that species. Mass action equations (at 25°C) for all
the ion pairs are substituted into the mass balance equations. The mass
balance equations are rearranged and solved for the concentration of the free
species in terms of association constants, activity coefficients, and the
uncomplexed concentrations of the other species. For the first iteration, the
total concentrations are set equal to the free concentrations, and the com-
plexed concentrations are calculated from the mass action equations for use in
recalculating a new ionic strength. New activity coefficients are then cal-
culated. The concentrations of the complexed species calculated in the pre-
vious iteration are included in the mass balance equations, and a new so-
lution is obtained for the mass balance and mass action equations. Approxi-
mately six iterations are required for convergence.
67
-------
ASAME incorporates 93 aqueous species; these include the free species,
acids and bases of these species, and all known inorganic ion pairs. The
effects of organic complexes are not considered in ASAME. Table 4 compares
standard river water and sea water data calculated by ASAME with the results
given for the same data in Nordstrom e_t al. (1979). In general, results
obtained with ASAME are within one standard deviation of the mean of those
reported for other equlibrium programs. Where it is not, or where the stan-
dard deviations are large, it is because the number of programs is small (n<5)
and the reported thermodynamic data is in significant disagreement. There are
no 'correct' answers, since each program is based on tabulated thermodynamic
data.
Calculation of Ion Activity Products
The precipitation of a mineral phase from a supersaturated solution is a
function of the reaction kinetics of the precipitation and the free energy
drive of the reaction. Little information is available concerning the kine-
tics of precipitation of the trace metal solid phases of interest. Conse-
quently, precipitation kinetics will not be considered. The free energy drive
of heterogeneous reactions involving calcium, magnesium, ferrous iron, mang-
anese, zinc, cadmium and lead will be used to determine the saturation state
of any solution with respect to a mineral phase. The minerals considered in
this study and their respective solubility products are listed in Table 5.
The saturation index, SI, is defined by Troup (1974) as
SI = log <> (S)
sp
where IAP is the ion activity product, K is the solubility constant, and n
sp
is the stoichiometric coefficient of the metal in the solid phase under con-
sideraton. The saturation index is a measure of the free energy drive of a
68
-------
Species
Ca2+
Mg2+
Na+
K+
S°4~
CL"
HC03
cof
H+
OH"
Mn2+
Fe2+
Zn2+
Cd2+
Pb2+
p°r
4
H2P°4
Ionic
Strength
River Water
Test Case
3.530 ±
3.523 ±
3.282 ±
4.447 ±
4.132 ±
3.540 ±
2.920 ±
5.231 ±
7.991 ±
6.237 ±
7.256 ±
8.476 ±
8.087 ±
10.041 ±
11.636 ±
10.133 ±
5.798 ±
6.644 ±
.00251
.006
.007
.001
.002
.014
.046
.006
.120
.008
.288
.384
2.527
.510
1.370
1.688
.070
.018
.031
±.0002
ASAME
3.529
3.520
3.283
4.446
4.127
3.554
2.913
5.158
7.985
5.965
7.137
6.893
9.143
9.427
11.992
10.036
5.770
6.655
0.00241
Sea Water
Test Case
2.057 ±
1.334 ±
.334 ±
1.994 ±
1.870 ±
.257 ±
2.940 ±
4.597 ±
8.097 1
5.630 ±
8.817 ±
-
7.596 ±
10.853 ±
11.530 ± 1
10.309 ±
6.904 ±
8.412 ±
0.653 ±
.056
.040
.024
.018
.096
.021
.242
.314
.053
.141
.330
.398
.309
.354
.204
.129
.073
.043
ASAME
2.058
1.332
.332
1.994
2.003
.256
2.854
4.580
8.092
5.652
8.725
7.810
8.076
10.733
11.581
10.447
6.905
8.308
.652
Table 4. Comparison of averaged concentrations (-log molality) and one
standard deviaton of major species given in Nordstrom et al.
(1979) and those of our model ASAME.
69
-------
Mineral
Calcite
Aragonite
Dolomite
Siderite
Rhodochrosite
Magnesite
Nesquehonite
Smithsonite
Otavite
Cerussite
Hydroxylapatite
Vivianite
Whitlockite
Reddingite
a-Hopeite
—
Hydroxypyromorphite
Fluorapatite
Chloropyromorphite
Struvite
Formula
CaC03
CaC03
CaMg(C03)2
FeC03
MnC03
MgC03
MgCO 3H 0
«J £*
ZnC03
CdC03
PbC03
Ca5(P04)3OH
Fe3(P04)2 • 8H20
Ca3(P04)2
Mn3(P04)2 - 3H20
Zn (P04)2 • 4H 0
cd3(po4)2
Pb5(P04)3OH
Ca5(P04)3F
Pb5(po4)3ci
MgNH PO • 6H 0
K
P sp
8.42
8.36
16.70
10.68
10.37
7.46
5.62
10.26
13.74
13.13
55.6
36.00
28.5
34.6
35.2
32.6
76.8
36.08
84.43
13.15
Reference
Jacobson and Langmuir, 1974
Christ et al. , 1974
Sillen, 1971
Smith and Martel, 1976
Morgan, 1967
Smith and Martel, 1976
Wagman et al. , 1968
Wagman et al. , 1968
Smith and Martel, 1976
Smith and Martel, 1976
Stumm and Morgan, 1970
Nriagu, 1972a
Duff, 1971
Nriagu and Dell, 1974
Nriagu, 1973
Sillen and Martell, 1964
Nriagu, 1972b
Roberson, 1969
Nriagu, 1974
Taylor et al., 1963
Table 5. Minerals considered by ASAME and pK values used.
sp
70
-------
metal phase to precipitate or dissolve. For example, a value of SI=2 indi-
cates that the mineral phase is two orders of magnitude supersaturated; a
value of SI = 0 indicates that the mineral phase is saturated with the solu-
tion; and a negative value indicates that the mineral phase is undersaturated.
Using equation (8), the degree of saturation of solutions with respect to
several authigenetic metal phases can be compared. Saturation index calcu-
lation were performed for calcium, magnesium, ferrous iron, manganese, zinc,
cadmium, and lead carbonates and phosphate for the intersitital waters of
Lake Erie sediments.
Field Measurements
Saturation index values were calculated for cores collected at the samp-
ling stations located in Fig. 1. The results for station Al do not differ
significantly from those obtained for Station 83. Consequently, they are not
presented in this report. The saturation indices for carbonate and phosphate
phases from a typical core at Station 83 are presented in Fig. 32 and Fig. 33,
respectively.
Smithsonite, otavite and cerrussite, the zinc, cadmium and lead car-
bonates, respectively, are clearly undersaturated and would not be expected to
control the pore water concentrations of zinc, cadmium, or lead. Calcite is
undersaturated in the upper portions of the cores, and levels off at a SI=M).5
at depth. Slight supersaturation of calcite in hard waters and organic rich
waters in not uncommon (Morton and Lee, 1968; Otsuki and Wetzel, 1974; Chave,
1965; Suess, 1970). The interpretation, then, is that the diffusional loss of
calcium and bicarbonate from the sediments is balanced by the dissolution of
calcite, and that a slight supersaturation is maintained due to organic col-
loidal interactions. Siderite is at saturation in the top 10 cm of the cores
and reaches an SI of ~0.6 at depth. Rhodochrosite is about one order of mag-
71
-------
*^
0-
20-
1
o
~ 40-
CL —
0>
Q60-
80-
Saturation Index
4-3-2-101234
i i i I i i i i
1 IT V
A + «b •
• i
H A • <
HA *
It A 4
* A +
>0 •
• 0 •
0.
•0.
• Calcite
• Siderite
O Rhodochrosite
A Smithsonite
+ Otavite
* Cerrussite
1 1 1 1 1 1 1
-
-
-•
-
-
Figure 32. Saturation index versus depth for interstitial water from a typical
Station 83 core for calcite (CaCO ), siderite (FeCO ), rhodochrosite (MnCO ),
sraithsonite (ZnCO ), otavite (CdCO,), and cerrusite (PbCO_).
72
-------
Saturation
4 -3 -2 -I
20-
80-
2
_l_
3 4
i
o * 0
* *o *
mO *
• Vivianite
• Whitlockite
OReddingite
AHydroxyiapatite
*Hydroxylpyromorphite
HChloropyromorphite
^) «*-Hopeite
O Struvite
I
I
I
I
T
I
I
Figure 33. Saturation index versus depth for interstitial waters
from a typical Station 83 core for vivianite (Fe (PO.)'
8H20),
whitlockite (Ca.fPO ) ), reddingite (Mn (PO4)2 • 3H20), hydroxylapatite
(Caj. (PO.) ^OH) , hydroxylpyromorphite (Pb- (PO.) .OH) , chloropyromorphite
(Pb5(P04)3Cl) , ct-hopeite (Zn3(P04)2 • 4H2) ) , and struvite
73
-------
nitude supersaturated throughout the core. These phases would be expected to
precipitate in recent, organic, and metal rich anoxic sediments (Troup, 1974;
Robbins and Callender, 1975; Holdren et al., 1975), and should also be preci-
pitating in these sediments.
Other carbonate phases listed in Table 5, but not shown in Fig. 31,
include dolomite, aragonite, magnesite, and nesquehonite. The solubility
difference between aragonite and calcite is small, and the general observa-
tions for calcite may be applied to aragonite. Aragonite is clearly present
in many surface sediments as shell fragments, and must be dissolving, as very
few shell fragments are found at depth. Dolomite is sourced to the sediments
as a detritial particle and behaves like calcite and aragonite. A portion of
the sediment diagenesis may be modeled as a dissolution of CaCO, and dolomite
(see 'Stoichiometric Model1). Magnesite and nesquehonite are 1-4 orders of
magnitude undersaturated and would not be expected to be present in these
sediments.
Table 5 cites 11 known pure end-member phosphate phases that are possible
authigenic precipitates. Struvite, hydroxylpyromorphite, and a-hopeite are
all clearly undersaturated and should not precipitate in the sediments
(Fig. 33). Also undersaturated but not shown for the purpose of clarity in
Fig. 33 are fluorapatite (5 x 10~6M F~;SI~-2.5), Pb (PO ) (SI>-10), and
Cd_(P04)_ (S]>-4). Chloropyromorphite shows considerable scatter about the
saturation boundary, and may be the controlling mineral for lead concen-
trations as suggested by Nriagu (1974). The calcium phosphate, whitlockite, is
at saturation throughout most cores. The other calcium phosphate, hydro-
xylapatite, is about an order of magnitude supersaturated. Hydroxylapatite is
a common detrital mineral phase in Lake Erie, so that its supersaturation and
presence in the sediment does not imply authigenic growth. Furthermore, its
74
-------
nucleation kinetics are such that several orders of magnitude supersaturation
may presist unless large quantities of relatively pure fine grained calcite
are present (Martens and Harriss, 1970; Stumm and Morgan, 1970). Stumm and
Leckie (1971) demonstrate that hydroxylapatite will form epitaxially on the
surface of calcite, but the rate of reaction is very slow under normal sedi-
mentary conditions. Thus, whitlockite and not hydroxylapatite would be ex-
pected to be the pure end-member calcium phosphate-controlling mineral phase
in these anoxic sediments. Vivianite is a little more than an order of magni-
tude supersaturated and reddingite has a saturation index of about 0.5. These
phases would be expected to precipitate in recent, organic and metal rich
anoxic sediments (Nriagu and Dell, 1974; Aller, 1977), and should also be
precipitating in these sediments.
Production Experiment
The continued release of alkaline earths, transition metals, and nut-
rients during the decomposition experiments suggests that a number of mineral
phases may reach or exceed saturation during the course of the experiment.
The results of mineral equilibria calculations for carbonate phases are given
in Fig. 34 and for phosphate phases in Fig. 35-
The data in Figs. 34 and 35 show that the saturation indices of the
minerals considered remain roughly constant over the length of the experiment.
In general, the variations in the saturation indices are no greater than the
spatial variability of the saturation indices from several cores obtained at
the same location. Calcite is undersaturated initially, but eventually levels
off at an SIMD.S. This level of supersaturation is also observed in the
field cores. The interpretation, then, is that calcium and bicarbonate are
released to the pore waters during decomposition by dissolution of calcite,
75
-------
4n
0
_C
c °
o
§-2
4-»
05
-4-
•D
D
J3t?0dochrosite
Calcite
Magnesjio. -Q-
50
100
150
200
Time (Days)
Figure 34. Saturation indices over time for siderite, rhodochrosite,
calcite, and magnesite observed in the production experiment.
-------
Vivianite
oxylapatite
. _ -^ --C
6 —
c
g
-2 H
CO
CO
-4H
>- o
50
100
150 200
Time (Days)
Figure 35. Saturation indices over time for vivianite, hydroxylapatite,
reddingite, whitlockite, fluorapatite, and struvite observed in the
production experiment.
-------
and that a slight supersaturation is maintained due to organic and collodial
interactions. Siderite is one to two orders of magnitude supersaturated and
rhodochrosite about one order of magnitude supersaturated.
Fig. 35 includes 6 known pure end-member phosphate phases that are poss-
ible authigenic precipitates. The calcium phosphate, whitlockite, is nearly
always at saturation. The other calcium phosphate, hydroxylapatite, is about
an order of magnitude supersaturated.
As discussed for the field cores, whitlockite and not hydroxylapatite
would be expected to be the pure end-member calcium phosphate-controlling
mineral phase in these anoxic sediments. Vivianite is a little more than an
order of magnitude supersaturated and reddingite has a saturation index of
about 0.5. These phases would be expected to precipitate in recent organic
and metal rich anoxic sediments, and should also be precipitating in these
sediments, but we have been unable to verify their presence.
Flux Experiment
The saturation state of various carbonate and phosphate mineral phases in
both the Homogenized Mud and Lake Core experiments was examined by calculating
mineral saturation indices using the computer program ASAME (described
eariler). Major ion data used in this calculation are given in Tables 6
(Homogenized Mud Core Experiment) and 7 (Lake Core Experiment). For the
Homogenized Mud Core Experiment major cations were determined for core H.
Sulfate data was estimated from Weiler (1973). For the Lake Core Experiment,
major cations were determined for a core taken at Station 83 (GASP XXXII-83).
Again, sulfate was estimated from Weiler's (1973) data. Unless major ion
data was available for a particular core, the above data was used to estimate
major ion concentrations in the Homogenized Mud Core Experiment and Lake Core
Experiment, respectively. ^b
-------
Depth
cm
0
0-1
1-2
2-3
3-4
4-5
5-6
6-8
8-10
14-18
22-26
Ca
pm
219.9
998.8
1145.5
1275.7
1372.8
1409.7
1420.4
1494.5
1562.1
1673.7
1722.8
Mg
pm
48.0
239.7
274.7
307.1
332.7
345.8
357.3
379.3
397.4
430.2
450.9
Na
pm
90.8
398.0
401.1
490.7
552.4
563.3
522.0
594.2
649.4
698.1
729.5
K
(jm
5.86
39.6
30.1
44.7
51.5
45.5
46.2
54.3
55.8
60.2
64.0
Mn
(jm
4.67
34.6
40.5
44.1
47.2
48.3
50.5
53.2
55.7
61.0
62.7
S°4
(Jm
29.8
15.4
14.3
13.2
12.3
11.4
10.6
9.5
8.1
4.8
2.6
Table 6 . Major ion data used in thermodynamic calculations. Homogenized
mud core experiment data determined for HMC-H. Sulfate
data estimated from Weiler (1973).
79
-------
Depth
cm
0
0-1
1-2
2-3
3-4
4-5
5-6
6-8
8-10
14-18
22-26
Table 7 .
Ca
|jm
884.2
906.2
912.0
918.4
924.4
930.5
936.6
945.7
957.9
997.0
1046.1
Major
core
Mg
(jm
227.7
234.9
234.9
236.2
226.2
226.2
214.3
218.7
219.7
282.5
304.3
Ion Data used
experiment data
Na
pm
480.2
527.6
526.5
525.5
524.4
523.4
522.3
520.7
518.6
511.7
503.2
K
|jm
44.9
49.8
50.3
50.9
51.4
52.0
52.5
53.3
54.4
57.9
62.3
in thermodynamic
determined from
Mn
fjm
3.1
21.0
21.4
21.9
22.3
22.7
23.1
23.7
24.6
27.3
30.7
S°4
urn
298.0
154.0
143.0
132.0
123.0
114.0
106.0
95.0
81.0
48.0
26.0
calculation. Lake
GASP XXXI II -83
Sulfate data estimated froim Weiler (1973).
80
-------
Homogenized Mud Core Experiment
The degree of saturation calculated for the various mineral phases did
not change over the course of the experiment; between core variations were
minimal. The results of the mineral equilibrium calculations for Core H are
for both carbonate and phosphate phases are shown in Fig. 36.
Carbonate Phases. Both magnesite and nesquehonite were significantly
undersaturated (S.I. = ~-l to -2 and -3 to -4 respectively). Rhodochrosite
and siderite are both supersaturated (S.I. = ~0 to 0.5 and +1 to +1.5 res-
pectively). Calcite was found to be very near saturation (S.I. = --0.3 to
+0.1) throughout the core. Calcite, siderite, and rhodocrosite all increased
their degree of saturation to a depth of 8 to 10 cm. Thereafter, their degrees
of saturation were constant.
Phosphate Phases. Flourapatite and struvite were both greatly under-
saturated throughout the core (S.I. = ~-2.5 and -4.0 respectively). Vivianite,
hydroxylapatite, and reddingite were all supersaturated throughout the core
(S.I. = ~+1.5, +1.0, and +0.5 respectively). Whitlockite was very near satura-
tion throughout the core (S.I. = -0.18 to +0.10). Vivianite, hydroxylapatite,
reddingite, and whitlockite all showed a higher degree of saturation above a
depth of 6 cm. Below this depth, the saturation indices were lower.
The above observations on the degree of saturation of various carbonate
and phosphate minerals in the homogenized mud core experiment are comparable
both in magnitude and pattern to observations made on cores taken from station
83.
Lake Core Experiment
In the Lake Core Experiment time variation in the degree of saturation
for some minerals was observed, but this degree of variation was small
(maximum A.S.I. = -1.0). The highest degree of time variation in saturation
81
-------
00
-5
0
10
e
o
a. 20
UJ
Q
30
0
SATURATION INDEX
+5 -5
'O' *?'A
0
-------
index was observed for siderite and vivianite. A lesser degree of variation
was seen for whitlockite and reddingite. No discernable variation was found
for magnesite, hydroxylapatite, rhodocrosite, or calcite. The pattern of
variation was the same for all minerals exhibiting variation. For these
minerals, the degree of saturation observed between depths of 2 and 6 cm was
less in the first core (LCE-6) than all succeeding cores. This probably
reflects the temperature increase the cores were subjected to for the experi-
ment (experiment temperature = 16°C; temperature at collection = ~10°C).
Carbonate Phases. Saturation indices observed for the carbonate phasts
considered are shown in Fig. 37. Both magnesite and nesquehonite (not shown)
were undersaturated at all depths (S.I. = -1.6 and -3.5 respectively). Both
rhodochrosite and siderite were found to be supersaturated. Their degree of
saturation increased with depth to a maximum at ~4cm then decreased to a more
or less constant value by 10 cm (S.I.'s at depth = 0.2 to 0.4 and 0.2 to 0.5
respectively). Calcite was slightly undersaturated at most depths, but tended
to be slightly supersaturated between 2 and 4 cm.
Phosphate Phases. Saturation indices for phosphate phases are shown in
Fig. 37. Not shown are observations for struvite and flourapatite which were
both greatly undersaturated (S.I. = -3.0 to -4.0 and -2.0 to -3.0 respectively).
The degree of saturation of all phosphate phases shown increases with depth to
a maximum value at - 4cm then decreases to more or less constant value at a
depth of ~10cm. Hydroxylapatite, in contrast to all other phosphate phases, was
supersaturated at all depths. Further, it was found to be near saturation in
overlying water. Vivianite tended to be supersaturated at all depths below 1
-------
-2
0
SATURATION INDEX
+2 -2
+2
0
10
20
J 30
X
& 0
o
10
20
30
4 A A
• A
,
-
-
SIDE
o ' '
• *Si^
/DA
• A «0k
RITE
-\'
«o
. «*
- °
-
_ ^
1 1
• LCE - 6
OLCE-7
A LCE - 8
ALCE-9
4 LCE -12
M AGNES ITE
' sl
3f°*
g^
of
"
,
CALCITE
'
-
-
-
-
I.
*
3k,
„
Oft,
RHODOCHROSITE
Figure 37. Saturation indices for carbonate phases observed in
the lake core experiment.
84
-------
-2
0
SATURATION INDEX
-1-2-2
0
+2
0
10
20
J 30
X
& 0
o
10
20
30
> A A '
• A A
<
-
-
VIVI/!
0 ' '
A $ Qk
HA 'o
•
01
«2G
*
-
<
A
-------
cm. Both whitlockite and reddingite tended to be undersaturated (S.I. -0.2
to -0.4 and -0.01 to -0.2 respectively) at all depths except a narrow range
(2-4 cm and 1-6 cm respectively).
Again, the above observations are entirely consistent with earlier obser-
vations (Station 83 cores, Production Experiment, Homogenized Mud Core Experi-
ment). This overall consistency indicates that the precipitation dissolution/
chemistry taking place in Lake Erie is adequately replicated in both the
production and flux experiments.
Mineral Equilibria in Anoxic Lake Sediments
In a multicomponent solution, such as interstitial water of sediments,
precipitated authigenic mineral phases are likely to be chemically mixed.
Suess (1979) indentifies a mixed Mn-carbonate [(MnQ o5CaQ oinMgn 05^C03-' and a
mixed Fe-phosphate t(Fe0 oc.Can -IA)^*®*)?] ^n anoxic Baltic Sea sediments.
Suess (1979) notes that in the case of the mixed Fe-Ca-phosphate the ion
activity product is lower than that of pure Fe (PO )„. So, for example, it is
possible that a mixed Fe-Ca-phosphate might be at saturation while the calcu-
lated value for vivianite is about an order of magnitude supersaturated. The
slight deviation from calculated saturation in Lake Erie sediments could be
due to the precipitation of mixed phases. A similar problem exists for the
apatite phases. Atlas and Pytkowitz (1977) show that the solubilities of
apatities vary widely, and interpret the results in terms of complex coatings
formed on the surface of the apatites. They conclude that apatites cannot be
described by the end member phases examined here. They suggest a surface
coating of a general form
CaA(HP04)B(P04)c(OH)D(F)E
Kramer (1964) suggested that francolite, a carbonate fluorapatite, is the
86
-------
solid phase which occurs in seawater. He lists the stoichiometry of the
mineral as NaQ 29Cag 59(P04>5 3?(S04)0 3Q (C03)Q 33 p2 Q4. In anaerobic
sediments this phase would probably not crystallize because of the reduction
of sulfate to sulfide. Saturation indexes may be calculated, however, from
the data in Appendix II. Saturation indices are slightly less than 1 in the top
of the cores and slightly greater than 1 at depth. These results indicate
that although whitlockite, the pure end member phase, appears to control the
interstitial waster concentrations of phosphate, various mixed apatites are
probably also forming in the sediments. Reliable thermodynamic data for these
mixed phases is non-existent. The indentification of thermodynamic and kinetic
data for the phases is essential for further progress in this area.
In addition to carbonate and phosphate mineral phases, sulfide mineral
phases are probably forming in the sediments. The free sulfide level is below
-4 =
the detection limit of the sulfide electrode, i.e., less than 10 MS at pH 7.
Therefore, it was not possible to include sulfide mineral calculations in the
thermodynamic model. Tisue (1980) reported that sulfate is reduced to sulfide
in Great Lakes sediments. In Lake Erie he found that sulfate did not diffuse
into the sediments for more than 5 cm before being reduced. Acid volatile
sulfide data (Fig. 5) is as high as 15 pM/gram dry sediment, values comparable
to those in marine sediments. Sulfide may precipitate as metal sulfides.
Mackinawite, Fe S, is thought to be important in controlling the ferrous
J.
-------
Therefore, control of metal concentrations by sulfide is not obvious. By sub-
stituting a sulfide activity of 10 M into the cadmium sulfide equilibrium
— R Q
expression, cadmium equals 10 M, a realistic value based on data from this
study. Mixed metal sulfide mineral phases are common and probably are also
forming. Further investigation of the levels of sulfide in the interstitial
water is warranted.
Verification of the thermodynamic calculations is difficult. There is
sufficient evidence that recent sediments are not in thermodynamic equili-
brium, but are driven by kinetic and diffusion mechanisms (Berner, 1974). In
addition, some of the thermodynamic constants are not well known and may be
very temperature sensitive. Troup (1974) indentified the iron carbonate phase
in Chesapeake Bay sediments by x-ray diffraction and Nriagu and Dell (1974)
report sand-sized vivianite 'nodules' in Lake Erie sediments at a location not
far from station 83. The small grain size and low abundance of the predicted
phases makes direct detection difficult. We have been unable to verify any of
the predicted phases by optical microscopy, x-ray diffraction, scanning elec-
tron microscopy, or x-ray fluourescence.
Neither inorganic complexation nor mineral-solution equilibria with
carbonate and phosphate adequately explain the concentration-depth trends
exhibited by zinc, cadmium or lead. Organic complexation may be important in
solubilizing these trace metals (Rashid and Leonard, 1973). After analyzing
the top centimenter of Lake Erie sediments, Kemp (1969) found that humic and
fulvic matter comprised 30 percent of the total organic carbon and humin com-
prised about 60 percent. Nissenbaum e_t^ al. (1971) reported that polymerized
organic matter accounts for nearly half of the total dissolved organic matter
in the interstitial water of Saanich Inlet sediments. Further, he found that
this matter was high in trace metals, especially iron, copper and zinc.
88
-------
Fulvic and humic acid complex metals at carboxyl and phenolic hydroxyl sites
and, when present, are probably responsible for the increased solubilities of
trace metals (Gamble and Schnitzer, 1973; Rashid and Leonard, 1973).
In sulfide rich water, simple organics, (i.e., amino acids and hydro-
xycarboxylic acids), do not significantly complex trace metals relative to
inorganic complexation (i.e. bisulfide) (Gardner, 1974). In systems containing
little sulfide, however, a significant portion of the trace metals are com-
plexed with simple organics. If the same ratio for organic complexation in
sulfidic waters is used in the low sulfide system, formation of metal-organic
complex follows the reaction
R _ (metal)(organic ligand) , ^
(metal-organic complex)
Thus, the ratio of free metal-organic complex is independent of sulfide activ-
ity. Application of Gardner's (1974) ratio to Lake Erie sediments results in
approximately 20% of the zinc being organically complexed. This calculation
indicates that if simple organics are present in low sulfide waters in similar
concentrations, they may contribute to the solubilities of the trace metals.
Complexation with simple organics is insignificant for Pb, which is highly
complexed with inorganic ligands and for Cd, which still exists predominantly
in the free state, but is significant for Zn, increasing the amount of total
complexation by up to 40 percent.
The metals in near surface pore waters may result from (1) release of
adsorbed cations from unstable hydrous oxides under reducing conditions, (2)
release of the cation from a mineral surface due to a change in adsorbing
capacity under low pH and low alkalinity conditions, (3) organic complexation
of released metals from hydrous oxides, and (4) release of metals from organic
material.
89
-------
For intersitital water in marine sediments, Brooks et_ al_. (1968) believe
that it is not improbable that the increased content of three metals (Cd, Zn
and Cu) in the interstitial water can be supplied completely from the marine
biosphere. They determined that cadmium and copper concentrations could be
accounted for by decomposing biological matter. Since organisms are the source
of 90 percent of the organic matter in Lake Erie (Harlow, 1966), biological
matter could supply the pore waters with the high levels of metals at the top
of the sediment column. Zinc is concentrated in the biota, both on the sur-
faces of and within algae and other organisms, and is released during bio-
degradation (O'Connor, 1968). Since up to 66 percent of the total zinc com-
plexed with proteins is loosely bound, it may be leached, transported and
become available for mineral deposition (Zajic, 1969).
The ferrous iron concentrations in the Lake Erie pore waters are 2000
time grater than those found in Santa Cruz Basin pore waters; it is not un-
reasonable to suspect that the release of adsorbed ions from ferric hydroxides
under reducing conditions is also an important source of trace metals (Duchart
et al., 1973). Lead and zinc often showed surface maxima in Lake Erie cores.
This phenomenon may be due to their release from hydrous oxides under reducing
conditions. Decaying organic matter, most concentrated in surficial sediments,
may then complex metal ions. Upon reduction of hydrous oxides with the ad-
dition of plant material, Kee and Bloomfield (1961) found that the metals were
present mostly as soluble organic complexes which could not be removed from
the fermented extracts by cation exchange resins.
Many of the cores show maximum metal concentrations at the surface and
from 5-10 cm below the surface. Walters and Wolery (1974) report a surface
minimum of trace metals in the Lake Erie sediments and a maximum at 7-10 cm in
the sediments. The interstitial maximum at 7-10 cm may be caused by solution-
90
-------
mineral equilibria and the surface interstitial maxima may be the result of
intense biodegradation and ferric hydroxide reduction. Humic by-products,
generated by biodegradation, can complex trace metals and hold them in so-
lution. Cline and Upchurch (1973) explain the surface maxima by the upward
transport of metals on bubble surfaces which have been found to concentrate
trace metals. They suggest that gases released by bacteria to form bubbles in
sediments could act as a transport mechanism for heavy metals released by the
same bacteria from chelation sites on the organics. Upon reaching the bio-
logically active surface sediments, the metals are immobilized as they form
new complexes or inorganic precipitates. In Lake Erie, most organics, such as
lipids, proteins and carbohydrates, decrease in an exponential manner with
depth as chemi-specific bacteria metabolize functional groups capable of
complexing cations (Cline and Upchurch, 1973). Contact of the adsorbed metals
with the complexing organic matter may cause the metals to become immobilized
in the sediment rather than passing out of the sediments on the bubble sur-
faces. Since many of the cores collected contain pockets of gas, the upward
movement of gas bubbles (e.g., CO,,, NH,, CH ) carrying adsorbed metals may be
another mechanism of upward metal transport.
91
-------
Stoichiometric Model
A stoichiometric model may be constructed utilizing the observed release
rates and assumed chemical reactions to predict the stoichiometry of the de-
composing organic matter and the nature of the hydrogen buffer. The model is
presented in Table 8. The principal driving reaction is anaerobic fermen-
tation. Reduction of C0_ is also a driving reaction, but it is quantitatively
half as important as fermentation (Wetzel, 1975). These two reactions release
nutrients, methane, and dissolved CO to the interstitial water which is buf-
fered by ammonium equilibria and the dissolution of solid carbonate and sul-
fide phases. In this scenario for anoxic sediments, ammonium and phosphate
are produced only by the decomposition of organics, while bicarbonate is
produced by both organic decompositon and metal carbonate dissolution. Since
quantitative data regarding their aithigenesis is lacking, precipitating phases
predicted by the thermodynamic model are not included in the stoichiometric
model. This exclusion will not quantitatively affect the results for the
major species (carbon, nitrogen, magnesium, calcium) but could underestimate
the release rates of the minor species (phosphate, iron, manganese).
A solution for this set of equations can be obtained from the relative
rates of release of the nutrients and metals. A typical relationship is given
in Fig. 39, where the apparent rates of release of calcium + iron + manganese
are plotted versus the producton rate of bicarbonate. The slope of the re-
gression line gives the relative rate of release. Strictly speaking, the data
from the different temperatures should not necessarily fall on a single line.
The above treatment assumes that the thermal kinetics of all the reactions are
the same -- an assumption which is obviously incorrect. There is, however,
sufficient scatter in the data to warrant the use of this simplifying assump-
92
-------
EQUATIONS:
(1)
AY
AZ
Anaerobic Fermentation
(2) -|X C02 + 2AX H
-fx CH4
H20
CO?Reduction
(3) AY NH3 + AY H
AY NH.
Ammonia Equilibria
(4) AZ H3P04
AZ HP04 + 2AZ H
Phosphate Equilibria
(5) -
-jX HCO" + -|x H*
4 34
CO Hydration
(6) B CaCO, + B H
J
(7) C CaMg(C)3)2 + 2C
(8) D FeS + D H+
(9) E MnS + E H+
B Ca++ + B
C Ca + C Mg ++ + 2C HCO,
D Fe++ + D HS
D Fe"*"* + D HS"
Calcite/Aragonite
Dissolution
Dolomite Dissolution
Machinawite Dissolution
Alabandite Dissolution
NET REACTION:
A(CH 0) (NH ) (H PO ) + 2AX H + B CaCO + C CaMg (CO ) + D FeS + E MnS
3'Y v 3
(AY + B + 2C + D + E - 2AZ - AX) H+
3'2
AY
AZ
(B + C) Ca++ + C Mg++ + D Fe++ + E Mn++
+(-|x + B + 2C) HCO~ + (D + E) HS"
-------
STOICHIOMETRIC CONSTRAINTS:
(1) (-|x + B + 2C)
(2) 4.77 AY
(3) 5.24 (B + C)
(4) 3.08 C
(5) 3.74 (B + C + D + E)
(6) 7.78 E
(7) 38.9 AZ
(~X + B + 2C)
(-|X + B + 2C)
(B + C)
(-|x + B + 2C)
+ B + 2C)
Bicarbonate Production
Ammonium vs. Bicarbonate
Calcium vs. Bicarbonate
Magnesium vs. Calcium
Calcium + Iron + Maganese
vs. Bicarbonate
Manganese vs. Iron
Phosphate vs. Bicarbonate
SOLUTION:
AX
AY
AZ
B
C
D
E
29.9
2.10
0.257
1.29
0.620
0.679
0.087
Table 8.
Stoichiometric model for decomposing Lake Erie sediments.
-------
ID
I80C
IO°C
5°C
HCO
- (meq/yr)
Figure 39.
of calcium
Rate of carbonate production (R'TTP/-V~) versus the sum of the rates
3
(R',,,) , iron (R',., ) and manganese (R'Mn) production for three temperatures.
-------
tion. Of course, the validity of the assumption is greatest for reactions
having similar thermal kinetics.
This system of equations does not permit an independent evaluation of the
parameter 'A' . Consequently the results are expressed as per mole of de-
composing organic matter. The solution is given in Table 8 .
One test of the model is the determination of the effectiveness of the
hydrogen buffer in the net reaction. The hydrogen buffer is given by the
expression (written as a reactant):
(AY + B + 2C + D + E - 2AZ - -| X). (10)
If the value of this expression is zero, there is no net consumption or pro-
duction of hydrogen ions; pH remains fixed. Substitution of the results of
the stoichiometric model gives a value for equation (9) of -2.59. This value
is significantly different from zero and indicates a net production of hydro-
gen ions equivalent to that of ammonium or calcium. Pore waters in Lake Erie
and in the jar experiments, however, appear to be well buffered at pH = 7.2.
Consequently, an additional hydrogen consuming reaction must be taking place.
The most likely possibility is that more ammonia is released than is measured
(Rosenfeld, 1979). If all of the excess hydrogen ion were consumed by liber-
ated ammonia, the total release rate of ammonium would be (2.10 + 2.59)/2.10 =
2.2 times the observed rate. This corresponds to an adsorption coefficient of
1.2, a reasonable value for recent sediments (Rosenfeld, 1979).
The proportion of the observed bicarbonate increase that is the direct
result of organic decomposition, -7 X, is given by:
7 X
4 (11)
A X + B + 2C
4
96
-------
where the denominator of equation (10) represents the total release rate.
Substituting values obtained from the stoichiometric model indicates that
of the observed release of bicarbonate is from organic decomposition and
of the release is from the dissolution of carbonate mineral phases.
Another test of the model examines the rate of release of methane in the
net reaction. Unfortunately, we were unable to measure methane in the jar
experiment throughout the course of this study. However, D. Adams (pers.
comm.) has made available interstitial water methane data from Station 83 and
Station A-l. Since the relative diffusivities of methane and ammonium are
approximately equal, and since transport of methane by diffusion and by bub-
bles deep in the cores is small, the molar concentration ratio of methane to
ammonium deep in the cores is given in the net reaction:
CH4 : NH* = | AX : AY. (12)
Substituting values shows that the molar ratio of methane to ammonium at depth
in the sediment should be about 10:1. The concentration ratios are
500|jm:100|jm = 5:1 at Station 83 and 2000(jm:200|jm = 10:1 at Station A-l. These
results are in good agreement with the stoichiometric model.
Solid Phase Prediction
The release of nutrients and metals to the pore water is necessarily ac-
companied by a decrease of these materials in the sediment solids. The depth
dependent formulations of the rates may be used to calculate the vertical
distributions of these materials in the sediments. The amount of a substance,
i converted from the solid phase to the solution phase over a depth interval
z to z was calculated assuming steady loading and constant porosity:
97
-------
2
v T *. r -i-j MFW 1 liter 1 r 2 „-„,. , .,.,
Ii lost from solids =— x 10QOg H Q x- S R(Z) dz (13)
2 Z
where uu is the sedimentation rate, MFW is the mass fraction of water, and MFS
is the mass fraction of sediment, and all other terms are as previously de-
scribed. Since most of the variation in the production rates, bacterial
abundance, and sediment solids data occur over the top 7 centimeters, equation (13)
was integrated over the depth interval 0-7 cm. The results of the calcu-
lations are given in Table 9. The data from 0 and 7 cm is based on single
samples (except carbon, obtained from Fig. 3). The C:N:P ratio of lost
material is ~135:9:1 while that of the calculated lost material is ^84:11:1.
The ratios are in reasonable agreement. Further, predictions of nutrient loss
derived from the apparent rate data are in accord with measurements of nutri-
ent regeneration in Lake Erie (Burns, 1976; Burns et al., 1976; Burns and
Ross, 1972). The ratio of C:N:P of the decomposing organic matter calculated
from the stoichiometric model is given by the AX:AY:AZ ratio and is 103:8:1.
The C:N:P of the sediment solids can be determined from the data in Table 9.
At the sediment-water interface, C:N:P is ^100:7:1 and at 7 cm is ~72:5:1.
These results indicate that carbon is recycled as fast or faster than nitrogen
and that both are recycled faster than phosphorous. This can also be seen in
the organic carbon to organic nitrogen and organic nitrogen to organic phos-
phorous ratios with depth in the sediment where the C:N ratio is approximately
constant and the N:P ratio decreases with depth.
In general, the model predicts that metals should be lost from the sedi-
ment in far greater quantities than is observed. This means that the release
of the metals by the dissolution of carbonate and sulfide phases must be
accompanied by the precipitation of other metal phases. The relative rates of
98
-------
ID
Parameter, i
Carbon
Nitrogen
Phosphorous
Calcium
Magnesium
Iron
Manganese
[i], Z=0
xlO Moles/g dry sed
2914
200
29.1
145
598
5.46
[i] 2=7
xlO Moles/g dry sed
1166
85.7
16.1
97.3
555
5.46
I [i] Lost
xlO Moles/g dry sed
Obs. Calc.
1749 1482
114 192
12.9 17.7
47.4 317
129
43.0 91.6
0 13.1
Table 9. Results of the solid phase prediction calculation compared to
the observed solid phase.
-------
release by decomposition and of removal by precipitation are such that release
has dominated over the course of this experiment. This single mass balance,
however, requires that most of the metals be retained by the sediments in
natural conditions. Presumably, some or all of the thermodynamically pre-
dicted authigenic phases are forming.
Prediction of Pore Water Concentration
The concentrations of materials in sediment pore waters may be also be
predicted from the decomposition rates. For example, the vertical distri-
bution of ammonia dissolved in interstitial waters can be determined by solv-
ing the equation (Rosenfeld, 1979)
§C _ D §!c _ , , 3C m -cf(T)Z
3t ~ D . 2 (1 K) w 3Z ot (T' e (14)
oZ
where C is concentration, t is time, D is the coefficient of diffusivity
(adjusted for porosity and tortuosity), Z is depth is the sediment (measured
positively downwards), K is the coefficient of adsorption, u) is the sedimenta-
tion rate, R1 (T) is the true rate of ammonia production at temperature T and
z = 0, and ct(T) is the decay constant for the exponential at temperature T as
defined earlier. Assuming steady state (i.e. 3C/8t = 0), and the boundary
conditions
C(z=0) = CQ
2)
the following solution is obtained:
Dd(T) + U>a(T)(l+K)
100
-------
As Z •» », and substituting typical values for the parameters, eq. (15) reduces
to
c =
Da(T)2
For ammonium, the calculated concentration profile is in reasonable agreement
with observed ammonium concentrations for depths less than ^5-7 cm, but the
calculated concentrations are a factor of 2-5 too high as Z -» ». Since the
rates were used successfully in the previous section to calculate the loss of
material from the solid phases by decomposition, there is every reason to
believe the reported rate data. Thus, the assumptions underlying the pore
water calculation are suspect. In particular, the assumption of steady state
must be evaluated.
One factor which regulates the supply of oxidizable organic matter to the
sediment-water interface is biological productivity in the overlying water. In
turn, biological productivity is regulated by nutrient supply. In Lake Erie,
nutrient loading has increased over time as has biological productivity (Sly,
1976). Nutrient loading to Lake Erie exhibits a sharp increase beginning
around 1947. This data corresponds to a depth of ~5 cm (uj = .168 cm/yr) for
station 83. This evidence suggests that the system is not at steady state;
that is, R1 (T) is not constant in time. Eq. (16) shows that by applying a
present-day value of R1 (T) which is substantially larger than the value when
the sediment at depth was deposited, the calculated value of C^ will also be
too high. Hence, application of the rate data reported in this study to the
prediction of pore water concentration profiles and fluxes requires additional
information on the historical record of R1 (T, time). The fact that the ob-
served decreases in concentration of the nutrients were accurately predicted
101
-------
by the rate formulations implies that decomposition and not loading quan-
titatively accounts for the majority of the changes in concentration of sedi-
ment solids in the near surface sediments. Calculation of the pore water
profile for depths larger than 5-7 cm requires the determination of a loading
curve at the sediment surface, from which R' (T, time) may be determined.
Direct and Indirect Flux Estimates
The flux of an material across the sediment water interface may be esti-
mated directly by observing concentration changes in a known volume of water
overlying a known area of sediment surface or indirectly by applying Pick's
first law to concentration gradients observed in the sediment column.
V Di 5 „ <17>
z=0
F. is the flux of component i across the sediment-water interface (moles/unit
area/ unit time). D. is the effective coefficient of diffusivity for compo-
2
nent i (length /unit time; adjusted for porosity, tortrosity, and tempera-
ture). The concentration gradient (moles/length /length), dc/dz is that at
the sediment-water interface (i.e. z = 0) and is generally approximated as the
concentration difference between the overlying water and the pore water con-
centration in the first sampling interval (0-1 or 0-2 cm depth) divided by the
mean depth of the sampling interval. Under field conditions, it is generally
inconvenient to attempt direct estimates of chemical fluxes, consequently the
indirect means described above is used. Unfortunately, few comparisons
between direct and indirect estimates of chemical fluxes have been made. One
goal of the flux experiments was to do just this.
102
-------
Laboratory Flux Experiments
In the flux experiments, direct flux estimates were based on observed
concentration changes in the overlying water. Estimates of flux were made for
ammonia, ferrous iron, SRP, bicarbonate, and SRS. For redox insensitive
materials (SRS and bicarbonate) all of the data collected following the
closure of the microcosms was used. For redox sensitive materials (ammonia,
ferrous iron, and SRP) only data taken during the initial concentration in-
crease observed for those paramenters was used. Since crashes in concen-
tration of redox sensitive materials following the initial rise in concen-
tration were probably due to the inadvertant admission of small quantites of
oxygen during sampling operations, it was felt that the most reliable esti-
mates of direct flux could only be made from initial rise data.
Indirect flux extimates were made using observed concentration profiles
at the time of core sacrifice. For purposes of the flux calculations, concen-
tration gradients were assumed to be linear at the sediment water interface.
The concentration in the 0-1 cm sampling interval was taken to be the concen-
tration at 0.5 cm and the concentration in the overlying water was taken to be
the concentration at the sediment-water interface. Indirect flux extimates
were made using Pick's law. Diffusion coefficients used in these calculations
are given in Table 10. The results of the direct and indirect flux calcu-
lations are given in Table 11.
Ideally, if the process controlling the movement of a material across the
sediment water interface is simple diffusion, and the concentration gradient
at the interface is well approximated by the difference in concentration
between overlying water and the 0-1 cm sampling interval, direct flux esti-
mates should equal indirect flux estimates. Indirect flux extimates for all
parameters except ferrous iron are greater than the direct flux estimates. In
the case of ferrous iron, the data show no clear pattern. Further, between
103
-------
COEFFICIENT OF DIFFUSIVITY (Cm2/sec)
ION
NH4
Fe++
HC03
Cl"
S°4
HPO^
H2P°4
Ca++
Mg++
Na+
K+
Mn++
H4si04
Zn+2
Cd+2
Pb+2
Cu+2
T=8°C
10.15
3.53
5.58
10.46
5.18
3.47
4.00
3.86
3.69
6.50
10.21
3.16
3.99
3.47
3.53
4.72
3.53
T=16°C
12.02
4.16
7.12
13.34
6.61
4.42
5.09
4.93
4.70
8.29
13.03
4.03
4.73
4.11
4.18
5.60
4.18
Table 10. Effective diffusion coefficients used in indirect flux
calculations. (Field estimates=8°C; flux experiment
estimates = 16°C). Data from Li and Gregory (1974) and
Wollast and Garrels (1972). Coefficients listed above are
adjusted for a porosity to tortuosity squared ratio
(/62) = 0.75.
104
-------
o
U1
Core
LCE-6
LCB-7
LCE-8
LCB-9
LCE-12
Mean
Variance
C.V.
MIC-R
HHC-N
HMC-B
HHC-H
HHC-J '
HHC-T
Mean
Variance
C.V.
HKC-G*
HMC-A*
HHC-C*
HHC-0*
HMC-C*
HMC-B*
Mean
Variance
C.V.
1
Direct
n.d
1050
883
1050
1276
1065
• 161
15.1
n.d
341
564
677
741
684
601
159
26.5
n.d
1200
180
1402
1128
410
864
535
61.9
°C
Indirect
' H/m2/day
561
2306
1807
1371
3552
1919
1115
58.1
6085
1788
1797
2349
1044
1410
2412
1851
76.7
3713
3933
5020
5630
347
2751
3566
1874
52.6
Direct
810
n.d
581
342
51
71
261
251
96.2
n.d
836
1336
592
371
769
781
359
46.0
n.d
0
79
736
293
280
347
277
79.8
Fe+2
Indirect
'6 M/n2/day
22
79
30
126
685
188
281
149.5
462
766
514
80S
437 -
464
575
166
28.9
292
712
661
875
70
1356
661
451
68.2
SRP
Direct
X10"6 H/«
n.d
597
86
169
160
253
232
91.7
0**
581
1710
388
192
96
593
651
109.8
0**
54
68
204
120
70
103
62
60.2
Indirect
«2/day
30
343
51
30
509
193
221
114.5
71
690
707
537
165
194
394
284
72.1
670
447
914
433
10
221
449
319
71.0
Direct
XlO-3
n.d
a
4
9
6
7
2
28.6
23
11
24
6
6
5
13
9
69.2
22
24
8
7
13
5
13
8
61.5
HCO
3 Indirect
M/ra2/day
4
9
2
5
14
7
S
71.4
20
23
16
36
24
14
22
8
36.4
16
18
20
29
7
33
21
9
42.9
Direct
xlO"
n.d
3868
4286
3520
2265
3485
871
25.0
2362
1097
1232
588 -
1168
847
1216
610
50.2
814
1042
1072
1129
958
932
825
384
46.5
H4S10
Indirect
6 M/m2/day
1414
2752
748
1076
5597
2317
1985
85.7
3342
3697
5007
3473
3123
2151
3466
926
26.7
2053
2659
3446
3145
1332
2566
2534
761
30.0
Table 11
Direct and indirect flux estimates obtained from the lake core experiment and the homogenized mud core experiment.
SRP flux is reported as moles P/in /day. * = tubificid population present. ** zero values not considered in
calculation of mean and variance (before anoxia), n.d - not determined
-------
core variability is higher in both the homogenized mud core experiment and the
lake core experiment. On the whole, direct flux estimates tend to be less
variable (overall mean C.V.~58%) than the indirect flux estimates (overall
mean C.V. ~70%). The paramenter showing the least overall variability in
estimated direct flux was NH* (C.V.=34.5%) followed by SRS = (C.V.=40.6%),
HCO" (C.V.=53.1%), Fe+2 (C.V.=74%), and SRP (C.V.=87.2%). The parameter
exhibiting the least overall variation in estimated indirect flux was SRS
(C.V. =47.5%) followed by HCO~ (C.V.=50.2%), NH* (C.V. = 62.4%), Fe+2
(C.V.=82.2%) and SRP (C.V.=85.9%).
Homogenized Mud Core Experiment
Because of the high degree of variation in estimated flux between cores,
no statistically significant difference in flux (either direct or indirect)
between with and without worm cases was found. If, however, only the means
are considered the following pattern emerges: The presence of tubificids
appears to have no effect on bicarbonate flux. Ammonia flux (both direct and
indirect estimates) is higher by a factor of ~1.44 (direct) to 1.48 (indirect)
in the presence of tubificids. Silica flux (both direct and indirect esti-
mates) is higher in the absence of tubificids by a factor of 1.47 (direct) to
1.37 indirect. Direct flux estimates for ferrous iron (x 2.25) and SRP (x
5.76) are higher in the absence of worms while indirect flux estimates for
+2
these paramenters are elevated in the presence of worms (x 1.15 for Fe ,- x
1.14 for SRP). From this information, we can reach the following tentative
conclusions: The presence of tubificid oligochaetes (at least at higher
population densities) appears to enhance ammonia flux by a factor of ~1.4 and
suppress silica flux by approximately the same factor. Examination of the
ammonia concentration profiles shows that this is probably the result of
enhanced near-surface ammonia production in the presence of worms. The sup-
106
-------
pression of silica flux in the presence of worms appears to be the result of
their reduction of the silica concentration gradient at the sediment-water inter-
face. No reason for this phenomenon is immediately apparent. The presence of
tubificids has no effect on bicarbonate flux. Tubificid inhabited sediments
exhibited suppressed ferrous iron flux (factor of ~ 2) and SRP flux (factor of
~ 6) even though indirect flux estimates based on the same cores suggest that
the presence of worms weakly enhances the flux of these two materials (factors of
+2
1.15 and 1.14 respectively). This result indicates that the movement of Fe
and SRP from sediments to overlying water is controlled by processes at the
sediment-water interface that are modified by the presence of worms. Tubificid
+2
activities have little effect on pore water gradients of either SRP or Fe , but
clearly affect the flux of both materials. The most likely reason for this is as
+2
follows: In the absence of tubificids both SRP and Fe diffuse toward the sediment-
water interface. Ferrous iron precipitates there as ferric exyhydroxide which
adsorbs SRP. This process results in the formation of a surface layer rich in
both iron and phosphorous. In the presence of worms such a layer is prevented from
forming by the worms' particle reworking activities (see Fisher et al., 1980).
Lake Core Experiments
Comparison of flux estimates for silica, ammonia, ferrous iron, and SRP
between the lake core and homogenized mud experiments can be made since these
materials were either not present or present at very low levels in the filtered
lake water overlying sediments in the lake core experiment. Lake cores ex-
hibited a higher direct ammonia flux estimate than either the with or without
worm homogenized sediments, but indirect flux estimates place the lake core
sediments midway between the homogenized sediments with worms (highest) and
the without worm homogenized sediments (lowest). Ferrous iron flux (both
direct and indirect estimates) was lower in the lake core experiments than in
107
-------
either of the homogenized mud core experiments. This may reflect mobilization of
iron by the homogenization process (e.g. oxidation of stoichiometric iron sulfides).
The flux estimates show the lake core sediments to be roughly comparable to the
homogenized sediments with worms, but the latter case showed the greatest difference
between the direct and indirect flux estimate. With respect to silica flux, the
lake core experiment sediment possessed the highest direct flux, but estimates of
indirect silica flux place the lake core sediments midway between the with and with-
out work homogenized mud sediments. This result is not surprising since the lake
cores have diatom-rich surface layer that is lacking in the homogenized mud cores.
Iron and Phosphorous. Examining flux data from all experiments, it was found
that direct flux estimates of ferrous iron were modestly correlated (r=.794) with
direct SRP flux estimates. A regression line calculated for this data had a
slope .962. A very weak correlation, however, was found to exist between the
indirect flux estimates for ferrous iron and SRP (r=.36). Flux estimates (direct
and indirect) of ferrous iron and SRP are compared in Fig. 40.
Field Cores
Indirect flux estimates from field measurements at stations 83 and Al are
given in Tables 12 and 13 (temperature assumed to be equal to 10°C).
Load and Loss Calculations
Metals
Industrialization has increased the loading of toxic metals to the en-
vironment. Most toxic metals do not occur in high concentrations in soils and
groundwater, and do not form colloids that are easily transported. Instead,
they form soluble complexes that remain dissolved or adsorbed to surfaces.
These metals are specifically adsorbed to ferric hydroxides which have a
positive surface charge at the pH of natural waters (Stumm and Morgan, 1970),
or adsorb to negatively charged surfaces such as the surfaces of clay plate-
lets. Iron forms a highly insoluble ferric hydroxide in oxygenated waters and
108
-------
2000-
a
•a
CM
E
V,
C/>
"5
E
to
b
X
u.
+
1000-
0-LCE DIRECT
• -LCE .INDIRECT
0
1000 2000
SRP FLUX (xlO~6 moles/mVday)
Figure 40 . Phosphorous flux estimates versus iron flux estimates (direct and indirect) for
the homogenized mud core experiment (with and without worms) and the lake core experiment.
-------
STATION 83
STATION Al
SPECIES
HC03
Cl"
HP04
»4Si°4
Ca+*
Mg"
Na+
K+
Fe*+
K
Mn++
31-V-78 17-V1-78 21-V11-79
-1109 1061 1109
371 578
175 6 5.5
801 785
500 834 417
-481 76.6
351
187
54.9 0 1.8
403 649 719
58.4
16-V111-78
-675
1536
18.2
973
168
22.3
130
20.3
37.8
602
49.4
MEAN
* SD
97*1001
829*508
51.1*71.5
853*85
479*238
-128*291
240*111
104*84
23.7*23.5
594*118
54.1*4.2
MEAN
CORE PEEPER * SD
867 410 639*229
-741 -56 -399*342
25 10.2 17.6*7.4
976 200 588*389
387 179 283*105
1228 49 638*589
Table 12. Indirect flux estimates determined from field measurements at Station 83 and Station Al. Units
are xio"6 Moles/m2/day.
-------
Species
Flux (10~6 tnoles/m2/day)
HC03
Cl"
SRP
SRS
Ca+2
Mg+2
Na+
K+
Fe+2
NH*
Mn+2
Zn+2
Cd+2
Pb+2
Cu+2
Station 83
97±1000
829±508
51±72
853±85
480±239
128±251
2401111
140+84
24+24
5941118
5414
0.634+3
0.024+0.028
0.046+0.205
0.20810.188
Station Al
6391229
-399+342
1817
588+389
-
-
-
-
283+105
6391589
-
-
-
«
Table 13. Summary of the flux calculations for Stations 83 and Al,
Negative fluxes indicate a flux to the sediments.
Ill
-------
under low oxygen conditions reduces to ferrous iron. Iron occurs in high
concentrations in terrigenous soils and groundwater, and is carried to lakes
as a colloid or adsorbed to mineral surfaces (Carrol, 1957).
Ferric hydroxide coatings are unstable under anoxic conditions (e.g.
buried lake sediments); they dissolve and release their adsorbed and copreci-
pitated trace metals. Further, sorption equilibria may change under different
alkalinities and pH's in the interstitial water system to release metals. The
released metals would be free to diffuse and react to form aiathigenic mineral
phases, and like phosphorous, would be expected to be concentrated in the
oxidized surface layer of sediments whch is rich in iron hydroxides. If the
hypolimnion becomes anoxic, the adsorbed metals might be expected to be re-
leased into the water column when the ferric hydroxides dissolve.
Factors governing the selectivity of clays for different cations are
valance, hydrated ionic radius, electronegatively, and free energy of forma-
tion (Leland e_t al. , 1974). Based on the frequently used tool for predicting
adsorption behavior, ionic potential, the order of ionic displacement on clays
is approximately:
Cu>Pb>Ni>Co>Zn>Ba>Rb>Sr>Ca>Mg>Na>Li
(Mitchell, 1964) where copper is the strongest displacer. The order is only
approximate and depends on relative concentrations in solution and the type of
adsorption surface: clays, hydroxides, oxides, or quartz grains. In his
study of limonitic concretions, Hirst (1962) found higher concentrations of
lead associated with the ferric hydroxides than with the clay fraction, and
recent literature has stressed the importance of trace metal association with
organic matter and ferric hydroxides rather than with clays (Lerman and Childs,
1973). This would suggest that burial by sedimentation of toxic metal con-
taining sediment would remove metals from the system (i.e. overlying water)
and lessen their potential pollution hazard.
112
-------
Metals can be incorporated into the crystal lattice of clays and other
minerals. For example, due to its similar ionic radius, lead replaces pot-
assium in its structural position in clays (Pinta and Ollat, 1961). Once the
metal ion is incorporated into the inner structure of a particle, it is no
longer available to water systems (Matson, 1968).
Organic complexing may tie up significant amounts of trace metals in both
particulate and soluble phases. Due to their similarities in chemical pro-
perties, heavy metals can substitute for other trace and alkali metals. The
degree of complexation is dependent on concentrations available in the en-
vironment, properties of each ion, organic ligands available for chelation and
complexation, and stability of the organic-metal complexes formed. No gen-
eralizations concerning relative affinities can be made, since the order of
stability is different for each metal-organic complex and depends on the
specific conditions of pH, Eh and metal and ligand concentration at the time
of formation. Algae show an affinity for trace metals according to the fol-
lowing order:
Zn>(Br), Cu>As>Cr>Co (Leland et al_._, 1974)
and considering that most organic matter in Lake Michigan sediments is of
autochthonous origin, a relationship between Lake Michigan trace element
concentrations in phytoplankton and organic carbon in the surficial sediments
is expected (Leland et al., 1974). Leland found that surficial sediments
showed an affinity for trace metals according to the following order:
Zn>Rb>Cr>(Br)>Cu>As>Ni>Co.
With the exception of chromium, organic affinity for cations is the same in
Lake Michigan phytoplankton and surficial sediments.
The above result suggests that a large percentage of the metal delivered
to the lake can be expected to end up in the sediments. Several investigators
113
-------
have determined metal contents in surficial sediments in order to identify the
anthropogenic and natural inputs of metals to sediments (Walters, e_t al.,
1974; Kemp et al., 1976; Nriagu et al., 1979). Knowledge of sedimentation
rate, water content, and metals concentrations is used to calculate the flux
of metals to sediments. The age of the sediment at depth, z, may be calcu-
lated as (Berner, 1980):
where is porosity at depth in the sediment, and
» PS (i - V •
the sedimentation rate at depth in the sediment where R is the mass flux of
sediment to the sediment-water interface (g/cm /yr) (usually determined by
radioisotope dating tecniques) and p is the density of the sediment solids.
s
For Station 83, this relationship is approximately:
T(years) = 14.857 * z(on) + 139.14 * (e~°-08615 * z
-------
At station 83, toxic metal concentrations typically decreased rapidly
immediately below the sediment-water interface and reach approximately con-
stant values below 5-10 cm. From equation 17 it can be determined that the
material at 10 cm represents the year -1910. This supports the interpretation
of Nriagu e_t al. (1979) that the constant concentration at depth reflects
'natural1 or background flux of metal and the increased concentration near the
surface is the result of anthropogenic metal input to the lake. It should be
noted, however, that the alternative interpretation that the metals are under-
going vertical transport (via molecular diffusion of dissolved metals) and are
deposited in the surficial oxidized sediment cannot be ruled out on the basis
of the above findings. Pore water concentration gradients indicate that
diffusive transport is not occurring and these data lend further support to
the anthropogenic loading interpretation.
We have adopted the mass balance of Nriagu e_t al. (1979), with the excep-
tion that we have included a flux of dissolved metal from the sediments to the
lake water as indicated by the interstitial water concentration data. The
amount of metal retained in the sediments was calculated from the difference
between total inputs and total outputs and not from indirect flux calcula-
tions. The calculated mass balance does not include a component for the
accumulation of metal within the lake water (dissolved + particulate). Our
simplified two compartment model is given in Fig. 41 . F represents the
natural flux of metal into the lake. F._ represents the flux of metal from
compartment 1, the lake water (dissolved + particulate) to the sediments. F91
£ X
is the flux of metal in the opposite direction. FI„ is predominantly particu-
late while F?1 is transport in the dissolved state. F, represents export from
the lake to the Niagara River and the Welland Canal. F_ is the anthropogenic
flux, and is considered to be zero before 1850 and to increase exponentially
with time since then (i.e. mirror population growth in the drainage basin).
115
-------
F
2
LAKE
WATER
R1
F
12
SEDIMENTS
R
2
Figure 41. Schematic diagram of the simplified two compartment model.
Fj represents natural flux to the lake. F12 represents the flux of metal
from compartment 1, the lake water (dissolved and particulate) to the
sediments. ^2^ is tne flux of metal in the opposite direction. F3 re-
presents export from the lake to the Niagara River and the Welland Canal.
F2 is the anthropogenic flux to the lake.
116
-------
The modified mass balance is given in Table 14. The basic difference
between the mass balances in Table 14 and Nriagu et al. (1979) is the in-
clusion of a flux from the sediments in our balance. Fluxes were calculated
from Pick's first law,
dc.
F. = D. ~ (22)
i i dz ..
z=0
Values of D. were calculated from data given in Li and Gregory (1974) and are
presented in Table 10. Concentration gradients at the sediment-water inter-
face (z = 0) were taken as the linear change between the overlying water and
the first sediment sample. Results of the flux calculations are given in
Table 12. Data from all cores at a single station were averaged and are
reported as the average ± the standard deviation (Table 13).
The toxic metals data display a large variability, but like iron and
manganese they are fluxed from the sediment to the water column. The mean
values were used in determining F? in Fig. 41. Table 14 is incomplete for
Cd. Consequently, only fluxes for Cu, Pb, and Zn were determined.
The molar ratio of Cu:Pb:Zn in the metals fluxing from the sediment is
4.5:1:13.8, and the molar ratio for the same metals in the accumulating sedi-
ments is 4.0:1:8.6. Thus, the flux of these metals from the sediment, F?1, is
proportional to the flux to the sediment, F „,- F01 = c*F „. Values for a can
A ^ O J. JL.&.
be estimated from data in the mass balance and can be assumed to be constant
in time and independent of metal loading:
Cu: F21 = 1.19 x 106 M/yr
F12 " F21 = 22'23 X 1Q6 M/yr
F12 = 23.42 x 106 M/yr
a = .0508
117
-------
Flux (10 moles/yr)
Input:
Detroit River
Tributaries, USA
Tributaries, Ontario
Sewage
Dredge Spoils
Atomosphere
Shoreline Erosion
Flux from Sediment*
Total*
OUTPUT:
Niagra River and Welland Canal
Retained by Sediment (mass balance)*
Cd
0.049
0.037
0.3A7
0.070
0.138
Cu
25.81
1.57
0.488
7.05
0.661
3.24
2.99
1.19
43.00
20.77
22.34
Pb
3.04
0.251
0.092
1.37
0.270
3.11
1.07
0.261
9.46
3.19
6.27
Zn
79.85
4.15
2.14
11.61
2.68
13.81
4.71
3.62
122.57
67.31
55.26
Table 14.
Inventory of sources and sinks of heavy metals in Lake Erie.
* All data except "Flux from sediment1 'Total1 and 'Retained by Sediment' from
Nriagu et al. (1979).
118
-------
Pb: F21 = .261 x 106 M/yr
F12 - F21 = 6.27 x 106 M/yr
F12 = 6.531 x 106 M/yr
a = .0400
Zn: F21 = 3.62 x 106 M/yr
F12 - F21 = 55.26 x 106 M/yr
F12 = 58.88 x 106 M/yr
a = .0615
From Table 3 of Nriagu et al. (1979), the fraction of the total flux of
metals to the sediments that can be assumed to be the "natural" or "pre-cul-
tural" component is as follows:
Cu: 0.527 ± 0.153
Pb: 0.339 ± 0.194
Zn: 0.327 ± 0.117
Calculation of F.. :
Assume the amount of metal accumulating in the sediment is proportional
to the total metal input to the lake. The current amount accumulating is
F 2 - F21 and the total input to the lake is ¥^ + Fy + F .. The ratios are:
cu: iri = -5170
Pb: |^|| = .6628
Zn: rif^JI = .4508
The natural accumulation in the sediment is F, _/.. . , x - F01,., . 1N and is
12(Natural) 21(Natural)
119
-------
calculated as the product of the fraction that is natural times the current
total flux to the sediment:
Cu: .527 x 22.23 x 106 = 11.72 x 106 M/yr
Pb: .339 x 6.27 x 106 = 2.13 x 106 M/yr
Zn: .327 x 55.26 x 106 = 18.07 x 106 M/yr
The total natural metal input is F + F_ , and may be calculated as
F12(Natural) " F21(Natural)
F - F
12 21
p + F + F
12 21
Cu-. 11.72 x 106/.5170 = 22.67 x 106 M/yr
Pb: 2.13 x 106/.6628 = 3.21 x 1Q6 M/yr
Zn: 18.07 x 106/.4508 = 40.08 x 106 M/yr
From the previous proportionality factors, F21(Natural) and F12(Natural) are
related.
Cu: F21(Natural) = °-°508 F12(Natural)
Pb: F21(Natural) = °-040° F12(Natural)
Zn: F21(Natural) = °-°615 F12(Natural)
Accumulation in the sediment is FI? - F?1
Cu: Accumulation = 11.72 x 106 M/yr = (1 - 0.508)F
F12(Natural) = 12'35 * ^
F21 (Natural) = °'627 x 10&
Pb: Accumulation = 2.13 x 106 M/yr = (1 - .0400)F12
F12(Natural) = 2'22 X
F21 (Natural) = °-088 X l
120
-------
Zn: Accumulation = 18.07 x 106 M/yr = (1 - .0615)F12
F12(Natural) = 19'25 X 10& M/*r
F21(Natural) = 1-184 x 106 M/yr
Therefore, F_ = Total natural influx - F?1/ , x
Cu: 22.67 x 106 M/yr - .627 x 106 M/yr = 22.04 x 106 M/yr
Pb: 3.21 x 106 M/yr - .0888 x 106 M/yr = 3.12 x 106 M/yr
Zn: 40.08 x 106 M/yr - 1.184 x 106 M/yr = 38.90 x 106 M/yr
Calculation of F_, the anthropogenic input:
This flux is set equal to zero for times prior to 1850, and is assumed to
increase exponentially until present (1975, Nriagu ejt al. (1979) mass balance
year). Further, for the purposes of demonstration, the exponential rate of
increase in the anthropogenic loading is assumed to continue for the next 100
years (2075). At any time t > 1850, F2(t) = FTotal input - FI - F21(t) F£ = 0
in 1850 and may be calculated for t = 125 yr from the mass balance data.
Cu: F2(125) = 43.00 - 22.04 - 1.19 = 19.77 x 106 M/yr
Pb: F2(125) = 9.46 - 3.12 - .261 = 6.079 x 106 M/yr
Zn: F9(125) = 122.57 - 38.90 - 3.62 = 80.05 x 106 M/yr
£t
Assuming exponential growth between t = 0 (1850) and t = 125 (1975), F2(t) may
Ki"
be formulated as F2(t) = A(e - 1)
Cu: F2(t) = 22.04 (e-0051t - l)x 106 M/yr
Pb: F2(t) = 3.12 (e-00865t - l)x 106 M/yr
Zn: F2(t) = 38.90 (e-0089t - l)x 106 M/yr
Calculation of FI_ and F?1:
Because F_ is time dependent, FI?, F?1, and F_ are also time dependent.
The time dependencies of FI ,, and F?1 may be determined using the previous
assumption that the amount of metal accumulating in the sediment (F1? - F?1)
is proportional to the total input to the lake (F + Fo + F21^' F12 ~ F21 =
-------
Therefore,
F12(t)(l - a) = 3(F: + F2(t) + F12(t)a)
where p has also been previously determined. Thus,
(F + F (t))
(1 - a - ap)
.. Cu: F12(t) = 12.35 e-°051t x 106 M/yr
F21(t) = .627 e-°°51t x 106 M/yr
Pb: F12(t) = 2.22 e-°0865t x 106 M/yr
F21(t) = 0.088 e-°0865t x 106 M/yr
Zn: F12(t) = 19.25 e'00891 x 106 M/yr
F21(t) =1.184 e-0089t x 106 M/yr
Calculation of F (t):
F- is assumed to be proportion to the mass of metal contained in reser-
voir 1:
F3(t) =
F3(125)
K3 = /12S) ^rom tne Present mass balance.
20.77 x 10* M/yr = 1 ^ yf-l
11.07 x 10 Moles
Pb: 3.19 x 10 M/yr = Q ?Q4 yf-l
4.53 x 10 Moles
zn: 67.31 x 10 M/yr = l ^ yr.-l
43.05 x 10 Moles
122
-------
where R. (125) is the present reservoir size = concentration of the metal in
the lake water (dissolved plus particulate) times the volume of the lake (469
Km3).
Calculation Procedure :
Because of the set-up of the problem, the knowns are for the present
time. The accumulation of the metal in the sediment represents deposition at
previous times. R (t) and F (t) do not have explicit formulations and must
be solved simultaneously in terms of the other parameters. Thus, the calcu-
lation was run backwards in time from t = 125 yrs (1975) until t = 0 (1850) .
Then the future was extrapolated by extending the calculations for 100 yrs
to 5 = 225. The calculations were performed step-wise with small At increments.
This resulted in a small discontinuity at -1975 (not shown). The two re-
lationships are :
Backward :
F21(t2) - F12(t2))At]
(1 + K3 At)
Forward :
R1(t2) = ^(t.^ + [F1 + F2(t1) + F21(t1) - F12(t1) - K^ (tj)] At (23)
The results of the calculations are shown in Figures 42 and 43. The effect
of the exponential loading is seen clearly in Fig. 42 where the concentrations
of the toxic metals in the lake water (dissolved plus particulate) also increases
exponentially. However, even with an exponential increase in toxic metal loading
in Lake Erie for the next 100 years , the concentration of these metals in the
lake water increase by about a factor of 3-5. Most of this increase would be
123
-------
JD
Q.
M
35
o>
Z
LU
O
z
o
o
, 30
o
25
O
20
I
O
c
N
10
5
oo 0
i
o
1850 1900 1950
2000 2050 YEAR
Figure 42 . Model predictions of Zn, Pb, and Cu concentrations in Lake Erie water
to the year 2075.
-------
CONCENTRATION (pg/g dry sed)
10
Ol
o
0
5
10
100 200 300 400 500 600
x15
a. 20
m
D25
30
2065
2050
2020
1975 m
1900
1850
1760
Figure 43. Model predictions of Zn, Pb, and Cu concentrations in Lake Erie sediments
(Station 83) to the year 2075.
-------
in the particulate component, but there is insufficient data to partition the
concentraton between dissolved and particulate. The effect of the ex-
ponentially increasing load to the lake is recorded in the hypothetical sedi-
ment profile in the year 2075 seen in Fig. 43. The accumulation of metal in
the sediment since 1850 is apparent, but an exponentially increasing load is
not. If loading rates continue to incease exponentially, the concentration of
lead in the sediments will surpass that of copper in about the year 2000. The
extent of mans influence on the toxic metals is Zn »Pb >Cu, while the natural
abundances is Zn > Cu > Pb.
Nutrients
The cultural eutrophication of Lake Erie and its effects on the lake's
environment and biota has been will documented (e.g. Beeton, 1965; 1969;
I.J.C., 1969). Even so, evidence for changes in Lake Erie over the past
century has depended mainly on the water quality data (from municipal water
intakes) and records of commercial fish catches. Few scientific studies
(Beeton, 1969) are available, and the detailed limnologic history of man's
effects on Lake Erie is not known. Studies of sediment cores have revealed
much about man's influence on toxic metal's budget (e.g. Nriagu, et al. 1979),
but knowledge of historic nutrient budgets for Lake Erie is scant. Historic
phosphorous loadings have been estimated from population growth data for the
basin (I.J.C., 1969), and attempts have been made to reconstruct nutrient
loadings from sediment stratigraphic data (Kemp e_t al., 1967). As noted by
Sly (1976), however, the use of sediment stratigraphy to reconstruct nutrient
loading must be carefully assesed with respect to the effects of in situ
degradation, bioturbation, and physical reworking. Lake Erie's sediments do
contain a record of the chemical, physical, and biological history of the
126
-------
lake. The question is whether or not we can interpret the record. This
section represents a first attempt to provide an answer to this question.
Here an attempt will be made to deconvolve the history of nitrogen loading
to Lake Erie using data on the present day concentration (solid phase) of
organic nitrogen in a core from Lake Erie (Station 83) and the present day
rate of organic nitrogen destruction (i.e. ammonia production) at the same
locale. Bioturbation and physical reworking are not considered.
In anoxic lake sediments, particulate organic nitrogen is transformed
to ammonia via bacterial action. The change in ammonia concentration with
time is assumed to be dependent on the amount of particulate organic nitrogen
present. At any depth, z,
ftfi
~ (z) = K-L (z) N (z) (24)
where dA/dt is the time rate of change in ammonia concentration, N is the
concentration of particulate nitrogen, and K is the rate constant. The
change in particulate organic nitrogen can be related to the change in
ammonia concentration by:
dN
f <*> - K2 -df (Z) (25)
where K is a constant used to convert units. Combining equations 24 and 25
results in
dN K (Z)
(z} = —yz) . (26)
127
-------
A solution for this equation is
N (z) = N (t - —)eb (27)
px ' po v o u)' v '
where
i z Ki(z)
b • - - ' - dz <28'
N is the concentration of particulate organic nitrogen at the time of de-
position, (t - z/w), u) is the sedimentation rate, and t is the present time.
Compaction is accounted for by considering iu as a mass flux (mass of dry
2
sediment /length /unit time) and z as overlying dry sediment mass (mass/
length2).
Equation 27 was used to calculate the history of particulate nitrogen
flux to the sediment-water interface at Station 83. Empirical deter-
minations of the rate of ammonia production as a function of depth (10°C data,
Production Experiment), particulate organic nitrogen (measured as total
kjeldhal nitrogen) as a function of depth, sediment porosity (ratio of pore
water volume to total volume) as a function of depth, and sedimentation rate
(by Pb-210, J.A. Robbins, pars, comm). were used in the calculation. (see
Figs. 7, 17, and 44). The dependence of K on depth (i.e. time, taken here as
comulative mass of dry sediment) was determined from the empirical data using
equation 27. The form of K.. (z) is well approximated as an exponential (r =
.997) (see Fig. 45 ). If dA/dt and dN /dt are expressed in equivalent units
P
(e.g. moles/volume wet sediment/time) K_ is dimensionless and has a numerical
value one. Once K and K are known, N (t -z/ui) is calculated using
128
-------
0
10
20
-P 30
.c
"Q.
Q 40
50
60
7n
% H20 (by weight)
40 60 80 1C
i i i
"
/
1
o
o
0
o
o
o
o
0 83-4-11-77
shell layer
O
Figure 44. Water content (%H O) versus depth at
Station 83.
129
-------
0
1977
CC
<
UJ
1900
Figure 45. K^ as a function of time of sediment deposition at Station 83.
130
-------
equation 27. The results of this calculation are shown in Fig. 46. which
illustrates the calculated history of particulate organic nitrogen loading,
N , and the amount of particulate organic nitrogen remaining at present, N .
Examination of Fig 46 reveals that the loading of particulate organic nitrogen
to sediments at station 83 has increased since -^1950. Scatter in the data,
preclude a detailed examination of particulate organic nitrogen flux increase
prior to this date. The above result agrees well with phosphorous loading for
Lake Erie's drainage basin (Sly, 1976). In the early nineteenth century, the
flux of particulate organic nitrogen to the sediment water interface, uu * N
(1820, 1860), was ~ .165 moles/M2/year, while u>* N (1974, 1977) was * .285
2
moles/m /year. This result indicates that the present ratio of anthropogenic
to natural nitrogen loadings at station 83 is ~ 0.73. This result is sub-
stantially lower than the estimates of this ratio made by Kemp e_t al. (1976).
These workers calculated the average ratio of anthropogenic to natural nitro-
gen loading to Lake Erie sediments to be ~ 1.79±0.7S. Kemp e_t al. (1976) did
not consider decomposition of nitrogen containing organic matter. This omis-
sion will increase the apparent ratio of anthropogenic to natural nitrogen
loading. For example, application of the methodology of Kemp e_t al. (1976) to
our organic nitrogen data for station 83 results in a anthropogenic to natural
nitrogen loading ratio of 1.99 for the present day. The neglect of organic
decompositon in estimates of man's effects on the flux of decomposable mater-
ials to lake sediments results in a significant overestimate of anthropogenic
inputs. The method outlined above for taking decomposition into account should
be tested at other locales and in other environments.
131
-------
0
1950
1900
o
o>
1850
moles/g x 10"
100 200
O
O
O
o
o
o
o
o
o
o
o
o
0
0
300
0
-|0
0
0
10
Q.
o>
Q
15
20
Figure 46. Present day particulate organic nitrogen concentration
(Np) and calculated organic nitrogen concentrations at the time of
deposition (N ) plotted versus depth and time of deposition at
Station 83.
132
-------
Prediction of Pore Water Chemistry
Equlibrium modeling of sediment pore water chemistry provides useful
informtion concerning potential solid phase sinks for materials of interest
(nutrients, toxic metals, etc). Further, such modeling can be used to check
consistency in understanding and interpretation of diagenetic processes in
sediments (Aller, 1977). Equlibrium modeling, however, cannot predict pore
water chemistry. Predictive modeling of pore water chemistry requires a
transport-reaction approach.
In Lake Erie reactive materials fall onto the sediment surface, move into
the sediments at the net rate of sedimentation, tu, and react. Solutes derived
from these reactions either move across the sediment-water interface via
molecular diffusion or precipitate as anthigenic minerals. Using this
scenario (which ignores bioturbation and physical mixing) , changes in the
chemistry of interstitial water can be described by one -dimensional
transport-reaction models. Formally, such a model can be written (ignoring
compaction) (Aller, 1977);
9£ - _1_ JL /n §£N _ ,„ ^ +
3t " 1+K 3z (D 3z' * 3z *
where:
C = concentration of pore water constituent being considered
z = vertical dimension, origin (z=0) at sediment-water interface,-
postive downwards
\u = sedimentation rate
D = effective diffusion coefficient modified for porostiy and
tortuosity
133
-------
R = reaction; function of z, and temperature
K = Langmuir adsorption coefficient
t = time
Using data from the production and flux experiments, we have attempted
transport-reaction models for the time evolution of ammonia and bicarbonate
profiles for the simplest case - the homogenized mud core experiment without
worms. In these experiments there is no sedimentation, temperature is con-
stant, there is no depth variation in the vertical distribution of reactive
material, and K is constant in both depth and time. In this situation, the
transport-reaction model is
_ . R
3t ~ + R'
and appropriate boundary and initial conditions are
C(Z > 0, 0) = C.
and
C(0, t) = CQ
Where C. is the • initial concentration at all depths and C is the concen-
tration at z=0. The solution of equation (30) under these conditions for the
semi- infinite case is given by Carslaw and Jaeaer (1959) :
C (z,t) = [AC + 7-fjL + 5f-J erf [ Z- ]
Vl+K; ^U _ /
2V/ DtN
Where AC = C. - C , -.
i o 134
-------
This equation was used to predict the evolution of ammonia and bicarbonate
profiles in the homogenized mud core experiment without worms. Values for R
were obtained from the production experiment data by fitting regression lines
to R vs TKN and R vs organic carbon data. For the ammonia calculation, K was
taken as 1.2 (the value obtained in the production experiment). Values for D
were obtained by fitting the models calculations to the observed profiles of
ammonia and bicarbonate. Values for C. and C were obtained fromt the flux
i o
experiment data. The numerical values for D,R,K, C , and C., used for the
calculations are given in Table 15.
The results of these calculations are compared with the experimental data
in Fig. 47. The agreement between the model results and the observed pore
water profiles is good for both ammonia and bicarbonate. The ammonia profiles
do not vary strongly with time while the bicarbonate profiles show strong time
variation. The model results track these trends well. For this case, homo-
genized sediment with no worms, a one-dimensional transport-reaction model
provides an adequate representation of the behavior of ammonia and bicar-
bonate. Even for this simple case, however, more complex models which con-
sider mineral equlibria and reactions near the sediment- water interface are
required to describe the behavior of iron and phosphorous. The addition of
tubificids to homogenized sediments confounds the model presented above since
ammonia is sourced near the sediment-water interface. The situation presented
by the real lake cores in even more complex as the distribution of decom-
posable organic matter is not constant with depth and animals are present.
135
-------
103
Ammonia Model Bicarbonate Model
D (cm2/yr) 3.8xl02 2.25xl02
K 1.2
R (mM/yr) 0.315 7.60
C (mM) .100 2.5
C. (mM) .160 2.5
Table 15. Values for D,K, R, C , and C. used in the time dependent
ammonia and bicarbonate transport-reaction calculations.
136
-------
— Ammonia (mM)
. 10
Q.
&
20
05 0
—i r-
t
\~2-doto
\~*-model results
0.5 0
0.5 0
05 O
05 O
0.5
— Alkalinity (mM)
0 10 O 10 0 10 0 IO O 10 0 K
o
J 10
-C
"a.
Q
20
" \
\
Y
t
V
• I
-
\~
1
-
}
r »\
-
.
1 ~ 1 1
' X*
\«
\*
\ •
\ ^
. \
•
\
f
I
- \
•
r x«
\»
• \
\ i •
\6
\ *
\\
•
1
Time (yrs) = .197
.271
.329
.364
.402
.463
Figure 47. Observed and calculated concentration profiles of ammonia and alkalinity (HCO )
for the homogenized mud core experiment (without worms).
-------
References
1. Agemian, H. and Chau, A.S.Y., 1975, An atomic adsorption method for the
determination of 20 elements in lake sediments after acid digestion.
Anal Chim. Acta 80: 61-66.
2. Aller, R.C., 1977, The influence of macrobenthos on chemical diagenesis
in marine sediments. Unpublished Ph.D. Dissertation, Yale University New
Haven.
3. Atlas, E. and Pytkowicz, R.M., 1977, Solubility behavior of apatites in
seawater. Limnol. Oceanogr. 22: 290-300.
4. Bachraann, R.W., 1963, Zinc-65 in studies of the freshwater cycle. In
Radioecology (eds. V. Schultz and A.W. Klement), pp. 485-496, Rheinhold.
5. Bancroft, K.,-Paul, E.A., And Wiebe, W.J., 1976, The extraction and
measurement of adenosine triphosphate from marine sediments. Limnol.
Oceanogr. 21: 473-479.
6. Beeton, A.M., 1965, Eutrophication of the St. Lawrence Great Lakes
Limnol. Oceanogr. 10: 240-254.
7. Beeton, A.M., 1969, Changes in the environment and biota of the Great
Lakes. In Eutrophication: Causes, Consequences, Correctives, Nat. Acad.
Sci.
8. Berner, R.A., 1970, Sedimentary pyrite formation. Amer. J. Sci. 268:
1-23.
9. Berner, R.A., 1971, Principles of Chemical Sedimentology. McGraw Hill,
240pp.
10. Berner, R.A., 1974, Kinetic models for the early diagenesis of nitrogen,
sulfur, phosphorous, and silicon in anoxic marine sediments. In: The Sea,
Vol. 5, Marine Chemistry (ed. E.D. Goldberg) pp. 427-450, John Wiley and
Sons.
11. Berner, R.A., 1976, The inclusion of adsorption in the modeling of early
diagenesis. Earth Planet. Sci. Lett. 29: 333-340.
12. Berner, R.A., 1980, Early Diagenesis: A Theoretical Approach. Princeton
University Press, 241pp.
13. Bradford, W.L., 1972, A study of the chemical behavior of zinc in Chesapeake
Bay waters using anodic stripping voltammetry. Chesapeake Bay institute
Technical Report, 76, Ref. 72-7.
14. Bray, J.T., 1973, The behavior of phosphate in the interstitial waters of
Chesapeake Bay sediments. Unpublished Ph.D. Dissertation, The Johns
Hopkins University, Baltimore.
138
-------
15. Bremner, J.M, 1965, Total nitrogen. In Methods of Soil Analysis, Part 2.
Agronomy (ed. C.A. Black), pp. 1149-1178, Amer. Soc. Agron.
16. Bricker, O.P., and Troup, B.N., 1975, Sediment-Water exchange in Chesapeake
Bay. In Estuarine Research (ed I.E. Cronin), Vol. 1, pp 3-27, Academic
Press.
17. Bricker, O.P., Matisoff, G. , and Holdren, G.R. Jr., 1977, Interstitial
water chemistry of Chesapeake Bay sediments. Basic Data Report No. 9,
Maryland Geological Survey, 67 pp.
18. Brooks, R.R., Presley, B.J., And Kaplan, I.R., 1968, Trace elements in
the interstitial waters of marine sediments. Geochim Cosmochim. Acta 32:
397-414.
19. Burns, N.M., 1976, Nutrient budgets for Lake Erie, 1970, J. Fish. Res.
Bd. Can. 33: 520-536.
20. Burns, N.M. and Ross, C. , 1972, Oxygen-nutrient relationships within the
central basin of Lake Erie. In Project Hypo (eds. N.M. Burns and C.
Ross), Can. Cent. Inland Waters Pap. No. 6, pp. 85-119.
21. Burns, N.M., Williams, J.D. H., Jaquet, J.-M., Kemp, A. L. W., and Lam,
D.C.L., 1976, A phosphorous budget for Lake Erie. J. Fish Res. Bd. Can.
33: 564-573.
22. Carroll, D., 1957, Role of clay minerals in the transportation of 'iron.
Geochim. Cosmochim. Acta 14: 1-27.
23. Carslaw, H.S. and Jaeger, J.C., 1959, Conduction of Heat in Solids.
Clarendon Press, 510 pp.
24. Chave, K.E., 1965, Carbonates: Association with organic matter in sur-
face seawater. Science 148: 1723-1724.
25. Chawla, V.K. and Chau, Y.K., 1969, Trace elements in Lake Erie. Proc.
12th Conf. Great Lakes Res: 175-184.
26. Christ, C.L., Hostetler, P.B., and Siebert, R.M., 1974, Stability of
calcite and aragonite. J. of Res., U.S. Geol. Surv. 2: 175-184.
27. Cline, J.T. and Upchurch, S.B., 1973, Mode of heavy metal migration in
the upper strata of lake sediments. Proc. 16th Conf. Great Lakes Res:
349-356.
28. Duchart, P., Calvert, S.E., and Price, N.B., 1973, Distribution of trace
metals in the pore waters of shallow marine sediments. Limnol. Oceanogr.
18: 605-610.
29. Duff, E.J., 1971, Orthophosphates. Part IV. J. Chem Soc. A. 1971:
921-926.
30. Emerson, S., 1976, Early diagenesis in anaerobic lake sediments: chemical
equlibria in interstitial waters. Geochim. Cosmochim Acta 42: 925-934.
139
-------
31. Emerson, S., and Widmar, G. , 1978, Early diagenesis in anaerobic Lake
sediments - II. Thermodynamic and kinetic factors controlling the for-
mation of iron phosphate. Geochim Cosmochim Acta 42: 1307-1316.
32. Fisher, J.B., and Matisoff, G. , in press, High resolution vertical pro-
files of pH in recent sediments. Hydrobiol.
33. Fisher/ J.B., and Lick, W.J., McCall, P.L., and Robbins, J.A., 1980.
Vertical mixing of lake sediments by tubificid oligocheates. J_._
Geophys. Res. 85: 3997-4006.
34. Gamble, D.S. and Schnitzer, M. , 1973, The chemistry of fluvic acids and
its reactions with metal ions. In Trace Metals and Metal-Organic
Interactions in Natural Waters (ed. P.C. Singer) pp. 265-302, Ann Arbor
Science.
35. Gardiner, J. , 1974a, the chemistry of cadmium in natural water - I. A
study of cadmium complex formaton using the cadmium specific electrode.
Wat. Res. 8: 23-30.
36. Gardiner, J. , 1974b, The chemistry of cadmium in .natural water - II.
The adsorption of cadmium on river muds and naturally occurring solids
Wat. Res. 8: 157-164.
37. Gardner, L.R., 1974, Organic versus inorganic trace metal complexes in
sulfidic marine waters. Geochim. Cosmochiro. Acta 38: 1297-1302.
38. Garrels, R.M. and Thompson, M.E., 1962, A chemical model for sea water at
25°C and one atmosphere total pressure. Amer. J. Sci. 260: 57-66.
39. Goldhaber, M.B., Aller, R.C., Cochran, J.K., Rosenfeld, J.K., Martens,
C.S., and Berner, R.A., 1977, Sulfate reduction, diffusion, and bio-
turbation in Long Island Sound sediments: Report of the FOAM group.
Am. J. Sci . 277: 193-237.
40. Gunkel, W. And Oppenheimer, C.H., 1963, Experiments regarding the sulfide
formation in sediments of the Texas Gulf Coast. In Symposium on
Marine Microbiology (eds. C.H. Oppenheimer and C. Thomas) pp. 674-683.
41. Harlow, G. L., 1966, Major sources of nutrients for algal growths in
western Lake Erie. St. Johns Rev. Div. , Univ. of Mich., Pub. No. 15,
389 pp.'
42. Hem, J. D., 1972, Chemistry and occurrence of cadiumm and zinc in surface
water and ground water. Wat. Resourc. Res. 8: 661-679.
43. Hesslein, R.H., 1976, The fluxes of CH., 2CO~, and NH--N from sediments
and their consequent distribution in a small lake. ""IJnpublished Ph.D.
Dissertation, Columbia University, New York.
44. Hirst, D.M., 1962, Geochemistry of modern sediments from the Gulf of
Paria II. The location and distribution of trace elements. Geocfaem.
Cosmochim. Acta 26: 1147-1187.
140
-------
45. HoLdren, G.R., Jr., Bricker, O.P., and Matisoff, G. , 1975, A model for
the control of dissolved manganese in the interstitial waters of Chesapeake
Bay. In Marine Chemistry in the Coastal Environment (ACS Symposium
Series 18), (ed. T.M. Church), pp. 364-381, American Chemical Society.
46. International Joint Commission, 1969, Pollution of Lake Erie, Lake Ontario,
and the international section of the St. Lawrence River. Vol. 2 -- Lake
Erie. Int. Joint Comm. Kept. : 316 pp.
47. Jacobson, R.L. and Langmuir, D. , 1974, Dissociation constants of calcite
and CaHCO from 0 to 50 C. Geochim. Cosmochim. Acta 38: 301-318.
48. Jenne, E.A., 1968, Controls on Mn, Fe, Co, Cu, Zn concentrations in soils
and water: The significant role of hydrous Mn and Fe oxides. In Trace
Inorganics in Water, Advances in Chemistry Series 73, pp. 337-387.
49. Kee, N.S. and Bloomfield, C., 1961, the solution of some minor element
oxides by decomposing plant materials. Geochim. Cosmochim. Acta 24:
206-226.
50. Kemp, A. L., 1969 Organic matter in sediments of Lake Ontario and Erie.
Proc. 12th Conf. Great Lakes Res. : pp. 237-249.
51. Kemp, A.L. W. , Thomas, R.L., Dell, C. I., and Jaquet, J.-M, 1976, Cul-
tural impact on the geochemistry of sediments in Lake Erie J. Fish
Res. Bd Can. 33: 440-462.
52. Leland, H.V., Shulka, S.S., and Shimp, N.F., 1973, Factors affecting the
distribution of lead and other trace metals in sediments of southern Lake
Michigan. In Trace Metals and Metal-Organic Interactions in
Natural Waters (ed. P.C. Singer) pp. 89-130, Ann Arbor Science.
53. Lerman, A., 1975, Maintenance of steady state in ocean sediments. Amer.
J. Sci. 275: 609-635.
54. Lerman, A. and Childs, C.W., 1973, Metal-organic complexes in natural
waters: Control of distribution by thermodynamic, kinetic, and physical
factors. In Trace Metals and Metal-Organic Interactions in
Natural Waters (ed. P.C. Singer) pp. 201-236, Ann Arbor Science.
55.' Li, Y-H. and Gregory, S., 1974, Diffusion of ions in seawater and in deep
sea sediments. Geochim. Cosmoshim. Acta 38: 703-714.
56. Lu, J.C.S. and Chen, K. Y. , 1977, Migration of trace metals in interfaces
• of seawater and polluted surficial sediments. Environ. Sci. Technol. 11:
174-182.
57. Martens, C.S., and Berner, R.A., 1974, Methane production in the inter-
stitial waters of sulfate depleted marine sediments. Science 185:
1167-1169.
58. Martens, C.S. and Harriss, R.C., 1970, Inhibition of apatite precip-
itation in the marine environment by magnesium ions. Geochim
Cosmoshim Acta 34: 621-629.
141
-------
59. Matisoff, G. , Harris, A., Matis, S., and Soster, F. , in press, Trace
metals mineral equlibria in Lake Erie sediments. J. Great Lakes Res.
60. Matson, W.R. , 1968, Trace metals, equlibrium and kinetics of trace metal
complexes in natural media. Unpublished Ph.D. Dissertation, Massachuesetts
Institute of Technology, Cambridge.
61. McCall, P.L. and Fisher, J.B., 1980, Effects of tubificid oligochaetes on
physical and chemical properties of Lake Erie sediments. In Aquatic
Qligochaete Biology (eds. R.O. Brinkhurst, And D.G. Cook), pp. 253-317,
Plenum.
62. Mitchell, R.L., 1964, Trace elements in soils. In Chemistry of the Soil
(ed. E.F. Bear) pp. 57-78, Reinhold.
63. Morgan, J.J. 1967, Chemical equlibria and kinetic properties of manganese
in natural waters. In Principles and Applications of Water Chemistry
(eds. S.D. Faust and J.V. Hunter) pp. 561-624, John Wiley and Sons.
63. Mortimer, C.H., 1941, The exchange of dissolved substances between mud
and water in lakes J. Ecol. 30: 280-329.
64. Mortimer, C.H., 1942, The exchange of dissolved substances between mud
and water in lakes. J. Ecol. 30: 147-201
65. Morton, S.D. and Lee, G.F., 1968, Calcium carbonate equlibria in lakes.
J. Chem Educ. 45: 511-513.
66. Muller, G. , 1969, Diagenetic changes in interstitial waters'of Holocene
Lake Constance sediments. Nature 224: 258-259.
67. Murphy, J. and Riley, J.P., 1962, A modified single solution method for
the determination of phosphate in natural waters. Anal Chim. Acta 27:
31-36.
68. Murry, J.W., Grundmanis, V., and Smethie, W. M. Jr., 1978, Interstitial
water chemistry in sediments of Saanich Inlet. Geochim. Cosmoshim. Acta.
42: 1011-1026.
69. Nelson, D.W. and Sommers, L.E., 1972, A simple digestion procedure for
the estimation of total nitrogen in soils and sediments. J.Environ. Qual.
1: 423-425.
70. Nissenhaum, A., Baebecker, M.S., and Kaplan, I.R., 1971, Studies on
dissolved organic matter from interstitial water of a reducing marine
fjord. In Advances in Organic Geochemistry (eds. H.R. Von Gaertner and
H. Wehner) pp. 427-440, Pergamon.
71. Nordstorm, O.K., Plummer, L.N., Wigley, T.M.L., Wolrey, T.J., Ball, J.W.,
Jenne, E.A., Bassett, R.L., Crerar, D.A., Florence, T.M., Fritz, B. ,
Hoffman, M. , Holdren, G.R. Jr., Lafon, G.M., Mattigod, S.V., McDuff,
R.E., Morel, F. , Reddy, M.M., Sposito, G. , and Thrailkill, J. , 1979, A
comparison of computerized chemical models for equlibrium calculations in
aqueous systems. In Chemical Modeling in Aqueous Systems (ed. E.A.
Jenne) pp. 857-892, Amer. Chem. Soc. Symp. Ser. No. 93.
142
-------
72. Nriagu, J.O., 1972a, Stability of vivianite and ion-pair formation in the
system Fe (PO.) -H PO.-H.O. Geochim. Cosmochim. Acta 35: 459-479.
73. Nriagu, J.O., 1972b, Lead orthophosphates I. Solubility and hydrolysis
of secondary lead orthophosphate. Inorg. Chem. 11: 2499-2503.
74. Nriagu, J.O., 1973, Solubility equlibrium constant of a-hopeite. Geochim.
Cosmochim. Acta 37: 2357-2362.
75. Nriagu, J.O., 1974, Lead orthophosphates - IV. Formation and stability
in the environment. Geochim. Cosmochim. Acta 38: 887-898.
76. Nriagu, J.O., 1975, Sulfur isotopic variations in relation to sulfur
pollution of Lake Erie. In Isotope Ratios as Pollutant Source and
Behavior Indicators, pp. - 77-93, IAEA.
77. Nriagu, J.O., and Dell, C.I., 1974, Diagenetic formation of iron phos-
phates in recent lake sediments. Am. Miner. 59: 934-946.
78. Nriagu, J.O., Kemp, A.L.W., Wong, H.K.T., and Harper, N. 1979, Sedi-
mentary record of heavy metal pollution in Lake Erie. Geochim.
Cosmocfaim. Acta 43: 247-258.
79. O'Connor, T.T. and Renn, C.E., 1964, Soluble-desorbed zinc equlibrium in
natural waters. J. Am. Waterworks Assoc. 56: 1055-1061.
80. Otsuki, A. and Wetzel, R.G., 1974, Calcium and total alkalinity budgets
and calcium carbonate precipitation in a small hard-water lake.
Arch. Hydrobiol. 73: 14-30.
81. Pankow, J.F., and Morgan, J.J., 1980, Dissolution and tetragonal ferrous
sulfide (mackinawite) in anoxic aqueous systems. 2. Implications for
the cycling of iron, sulfur, and trace metals. Environ. Sci. Technol.
14: 183-186.
82. Pinta, M. and Ollat, C. , 1961, Recherches physico-chemiques des elements
-traces dans les sols tropicaux - I. Etude de quelques du Dahomey
Geochim Cosmochim. Acta 25: 14-23.
83. Presley, B.J., Koloday, Y., Nissembaum, A., and Kaplan, I.R., 1972, Early
diagenesis in a reducing fjord, Sannich Inlet, British Columbia - II,
Trace element distribution in interstitial water and sediment. Geochim.
Cosmochim. Acta 36: 1073-1090.
84. Rashid, M.A. and Leonard, J.D., 1973, Modification in the solubility and
precipitation behavior for various metals as a result of their interaction
with sedimentary humic acid. Chem. Geol. 11: 89-97.
85. Reeburgh, W. S., 1967, An improved interstitial water sampler
Limnol. Oceanogr. 12: 163-165.
86. Reeburgh, W.S., and Heggie, D.T., 1977, Microbial methane consumption
reactions and their effect on methane distribution in freshwater and
marine environments. Limnol. Oceanogr. 22: 1-9.
143
-------
87. Richards, R.A., 1965, Anoxic basins and fjords. In Chemical Oceanography
(eds. J.P. Riley and G. Skirrow), Vol. 1., pp. 611-645, Academic Press.
88. Riley, J.P., and Taylor, D., 1972, The concentrations of Cd, Cu, Fe, Mn,
Mo, Ni, V, and Zn in part of the tropical north-east Atlantic Ocean.
Deep Sea Res. 19: 207-318.
89. Robbins, J.A., and Callender, E., 1975, Diagenesis of manganese in Lake
Michigan sediments. Amer. J. Sci. 275: 512-533.
90. Roberson, C.E., 1969, Solubility implication of apatite in seawater.
U.S. Geol. Sur. Prof. Pap. 550-D, pp. D178-D185.
91. Rosenfeld, J.K., 1977, Nitrogen diagenesis in nearshore anoxic sediments.
Unpublished Ph.D. Dissertation, Yale University, New Haven.
92. Sandell, E.B., 1959, Colorimetric Determination of Traces of Metals,
Interscience, 1032 pp.
93. Semkin, R.G. and Kramer, J.R., 1976, Sediment geochemistry of Sudbury-area
lakes. Can. Miner. 14: 73-90.
94. Sholkovitz, E., 1973, Interstitial water chemistry of Santa Barbara Basin
sediments. Geochim. Cosmochim. Acta 37: 2043-2073
95. Siever, R. , Beck, K.C., and Berner, R.A., 1965, Composition of inter-
stitial waters of modern sediments. J. Geol. 73: 39-73.
96. Sillen, L.G., 1971, Stability constants of metal-ion complexes.. Suppl.
1. Spec. Pub. 25. The Chera Soc., London, 865 pp.
97. Sillen, L.G. and Martell, A.E., 1964, Stability constants of metal-ion
complexes. Spec. Pub. 17. The Chem. Soc., London, 754 pp.
98. Sly, P. G., 1976, Lake Erie and its basin. J. Fish Res. Bd. Can. 33:
355-370.
99. Smith, R.M. and Martell, A.E., 1976, Critical Stability Constants.
Vol. 4: Inorganic Complexes. Plenum, 257 pp.
100. Solorzano, L. , 1969, Determination of ammonia in natural waters by the
phenol hypochlorite method. Limnol. Oceanogr. 14: 799-801.
101. Sommers, L.E., and Nelson, D.W., 1972, Determination of total phosphorous
in soils: A rapid perchloric acid digestion procedure. Soil Sci.Soc.
Amer. Proc. 36: 902-904.
102. Standard Methods For the Examination of Water and Wastewaters, 1975, 14th
Edition, American Public Health Association, 1193 pp.
103. Strickland, J.D.H. and Parsons, T.R., 1969, A practical handbook of
seawater analysis. Can. Fish Res. Bd. Bull. 167: 311 pp.
144
-------
105. Stumm, W. and Leckie, J.O., 1971, Phosphate exchange with sediments; its
role in the productivity of surface waters. In Proc. of the 5th
Int. Conf. of Water Pollution Res. 1970 (ed. S.H. Jenkins), Paper
111-26 pp. 1-16, Pergamon.
106. Stumm, W., and Morgan J.J., 1970, Aquatic Chemistry, John Wiley and Sons,
583 pp.
107. Suess, E., 1970, Interaction of organic compounds with calcuim carbonate.
I. Association phenomena and geochemical implications. Geochim.
Cosmochim. Acta 34: 157-168.
108. Suess, E., 1979, Mineral phases formed in anoxic sediments by microbial
decomposition of organic matter. Geochim Cosmochim. Acta. 43: 339-352.
109. Taylor, A.W., Frazier, A.W., and Gurney, E.L., 1963, Solubility products
of magnesium ammonium and magnesium potassium phosphates. Trans. Faraday
Soc. 59: 1580-1584.
110. Tisue, T., 1980, Sulfate reduction in the interstitial fluids of Great
Lakes sediments. 23rd Conference on Great Lakes Research, Abs., Kingston,
Ontario, May 19-22, 1980.
111. Tobin, R.S. and Anthony, D.H.J., 1978, Tritiated thymidine incorporation
as a measure of microbial activity in lake sediments. Limnol. Oceanogr.
23: 161-165.
112. Troup, B.N., 1974, The interaction of iron with phosphate, carbonate, and
sulfide in Chesapeake Bay interstitial waters: a thermodynamic inter-
pretation. Unpublished Ph.D. Dissertation, the Johns Hopkins University,
Baltimore.
113. Vosjan, J.H., 1974, Sulfate in water and sediment of the Dutch Wadden
Sea. Neth. J. Sea Res.. 8: 208-213.
114. Wagman, D.D., Evans, W.H. Parker, V.B, Halow, I., Bailey, S.M. and.Schumm,
R.H. , 1968, Selected values of chemical thermodynamic properties. National
Bureau of Standards Technical Note 270-3.
115. Walters, L.J. and Wolrey, T.J., 1974, Transfer of heavy metal pollutants
from Lake Erie bottom sediments to the overlying water. Water Resources
Center, Engineering Experiment Station. Ohio State University.
116. Walters, L.T., Wolrey, T.J., and Myser, R.D., 1974, Occurrence of As, Cd,
Co, Cr, Cu, Fe, Hg, Ni, Sb, and Zn in Lake Erie sediments. Proc.
17th Conf. Great Lakes Res.: pp. 219-234.
117. Watson, S.W., Novitsky, T.J., Quinby, H.L., and Valois, F.W., 1977,
Determination of bacterial number and biomass in the marine environment.
Appl. and Environ. Microbio. 34: 940-946.
118. Weiler, R.R., 1973, The interstitial water composition in the sediments
of the Great Lakes - I, Western Lake Ontario. Limnol. Oceanogr. 18:
918-931.
145
-------
118. Weiler, R.R., And Chawla, V.K., 1968, The chemical composition of Lake
Erie. Proc. llth Conf. Great Lakes Res. : 583-608.
119. Wetzel, R.G., 1975, Limnology. W.B. Saunders Company, 743 pp.
120. Wheeler, R., and Duriming, C., 1976, Trace element geochemistry of piston
cores from western Michigan coastal lakes. The Can. Miner. 14: 23-31.
121. Wollast, R. and Garrels, 1971, Diffusion coefficient of silica in sea-
water. Nature, Physical Science 229: 94.
122. Zajic, J.E., 1969, Microbial Biogeochemistry, Academic Press, 345 pp.
123. Zirino, H. and Healy, M.L., 1970, Inorganic zinc complexes in seawater.
Limnol. Oceanogr. 15: 956-958.
146
-------
APPENDICES
I. CHEMICAL METHODS
Aqueous Samples
Interstitital water samples were apportioned for several analyses:
1. First 8 mis were collected in an electrode cell or a vial to measure pH
and sulfide.
2. 3 mis of sample were drained into graduated test tubes containing re-
agents for complexing Fe(H)-
3. The remainder was collected in acid-washed sample bottles to be ap-
portioned as below:
a. 1 ml for analyses of NH , NO - NO., soluble reactive phosphate
(SRP), and soluble reactive silica (SRS).
b. 1 ml for alkalinity titration.
c. 1 ml for a chlorinity titration.
d. The remainder was saved for later trace metal and major cation
analyses.
NH , NO-NO,, SRP, SRS, carbonate alkalinity, pH and sulfide analyses
were all performed within six hours of sample collection.
147
-------
Sulfide and pH
These measurements were made on the interstitial water samples in a glove
box under a nitrogen atmosphere since (1) oxygen affects the sulfide concen-
trations in the pore water, and (2) equilibration of dissolved CO with CO in
the atmosphere affects the pH of the sample. Samples were drained from a
squeezer directly into small vials. A Fisher miniature glass universal elec-
trode, a calomel or silver/silver chloride double junction reference electrode
and an Orion Ag+/AgS sulfide electrode were used for these measurements. All
sulfide values were found to be below the detection limit of the electrode
(•v-10 M at pH 7) and are not recorded in this report.
pH was measured with a Fisher miniature microprobe micro combination
electrode on a Fisher Accumet 420 pH/ion meter.
Chlorinity
Chlorinity was determined in one of two ways. The shipboard method is
described in Standard Methods (1975). One milliliter of sample was placed in
a vial and titrated with standard mercuric nitrate delivered via microburet.
An acidified mixed indicator composed of diphenylcarbazone, xylene cyanol FF,
and nitric acid gives a purple endpoint at a pH of about 2.5. For some labor-
148
-------
atory work, chlorinity was determined on a Technicon Auto Analyzer II.
Mercuric thiocyanate and ferric nitrate are mixed with the sample where chlor-
ide can displace the thiocyanate ion which then reacts with ferric ion to form
a highly colored complex whose absorbance, read colorimetrically at 480 run, is
proportional to the original chloride concentration.
Alkalinity
Total carbonate alkalinity was determined as described in Standard
Methods (1975). One milliliter of sample was placed in a vial and titrated
with standard (0.02N) hydrochloric acid delivered via microburet to a mixed
bromcresol green - methyl red endpoint (pH = 4.7.)
Ammonia
The method of Solorzano (1969) was used for the determination of ammonia.
An appropriate aliquot of the pore water was diluted to five milliliters. To
this were added sequentially ethanolic phenol, sodium nitroprusside and a
/
mixture of alkaline sodium citrate and sodium hypochlorite. The amount of the
blue indophenol formed was determined spectrophotometrically at 640 nm after
one hour. Shipboard analyses for ammonia utilized the same reaction, however,
the analysis was performed with the aid of a techncion Auto Analyzer II.
149
-------
Nitrate-Nitrite
Nitrate plus nitrite nitrogen was determined with a Technicon Auto Ana-
lyzer II on a suitably diluted aliquot of the pore water sample. Nitrate is
reduced to nitrite by a copper-cadmium reduction column. The nitrite ion then
reacts with sulfanilamide under acidic conditions to form a diazo compound.
This compound then couples with N-l naphthylethylene-diamine dihydrochloride
to form a reddish-purple azo dye whose color is read at 550 nm.
Soluble Reactive Phosphate (SRP).
SRP was determined with a Technicon Auto Analyzer II on a suitably di-
luted aliquot of the pore water sample. A single reagent solution, when added
to the sample, forms a phosphomolybdenum complex which is reduced by ascorbic
acid to a blue compound. The intensity of absorbance, measured color-
imetrically at 880 nm, is proportional to the original phosphate con-
centration.
Soluble Reactive Silicate (SRS)
The procedure for SRS given by Strickland and Parsons (1969) was modified
for small water volumes. The sample is reacted in a test tube with acidified
ammonium molybdate for fifteen minutes. A combined reagent containing p-
methylaminophenol sulfate reduces the silicomolybdate to a blue compound while
oxalic acid removes phosphorus interference. The absorbance was read spec-
trophotometrically at 810 run between three and six hours after reagent ad-
dition. A Technicon Auto-Analyzer II provided an alternate method of ana-
lysis. The chemistry involved is similar to the manual determination des-
cribed above except that ascorbic acid is used as the reducing agent. Ab-
sorbance is read 660 nm.
150
-------
Ferrous Iron
Fe(II) concentrations were determined using 1, 10-o-phenathroline as the
colorimetric complexing reagent, as descrived by Sandell (1959). In a glove
box under a nitrogen atmosphere, 3 mis of pore water sample were drained
directly into graduate test tubes containing phenanthroline reagent buffered
at pH 3.6 (Troup, 1974). After 30 minutes, the 76(11) complex is stable with
respect to oxygen and can be removed from the glove box and read spetro-
photometrically at 510 nm.
Calcium, Magnesium, and Mananese
Calcium, magnesium and manganese were determined using a Perkin Elmer 360
atomic absorption spectrophotomater in an air-acetylene flame for pore water
samples which had been acidified upon collection with Ultrex nitric acid to pH
1 and refrigerated in polyethylene bottles. All samples were diluted before
being run (calcium and magnesium with 5% HCL - 0.2% LaO; manganese with de-
ionized water).
Zinc, cadmium and lead
Zinc, cadmium and lead present in trace amounts in pore waters were
determined by differential pulse anodic stripping voltammetry using a
Princeton Applied Research Model 174 polarographic analyzer. Pore waters were
squeezed directly into polyethylene bottles, acidified with Ultrex nitric acid
to pH 1, and refrigerated until analyzed. Prior to analysis, an aliquot was
pipetted into a beaker and acidified at the rate of one milliliter of Ultrex
Nitric acid per ten milliliters of sample. The sample was covered by a watch-
glass and taken to dryness in an 80°C oven. The sample was reconstituted by
adding ten millilters of 0.1N Ultrex nitric acid and lOOpl of 10% hydro-
xylamine hydrochloride and warming at 80°C for fifteen minutes. Addition of
15 milliliters of 0.2N ammonium citrate buffer of pH 3.5 brought the sample pH
to 3.1 ± 0.2. 151
-------
An aliquot of the diluted sample was pipetted into a polarographic cell.
Electrodes were inserted into the cell, and the cell was purged with oxygen
-free nitrogen. Following purging, a fresh mercury droplet was dialed on the
hanging mercury drop electrode. Metals were concentrated onto the mercury
drop from the stirred solution for two minutes at -1.2 volts vs. S.C.E.
Deposition was continued for an additional fifteen seconds with no stirring.
A voltage ramp was then applied at a rate of five millivolts per second in a
positive direction for approximately 1.4 volts. Superimposed on this ramp are
voltage pulses. Low detection limits were achieved by measuring current flow
differentially immediately before and immediately after pulse application.
The resulting polarogram records differential current per change in voltage
vs. voltage. As a result of increased current flow at the metals' charac-
teristic oxidation potentials, a peak-shaped waveform results whose height is
proportional to the amount of metal originally present. Zinc produced a peak
at about -1 volt, cadmium, at -0.6 volts and lead near -0.4 volts. After the
polarogram was run on the diluted sample, a known quantity of zinc, cadmium
and lead was added to the cell, and the cell was nitrogen purged. The pre-
vious mercury drop was dislodged, a new one was dialed, and another polarogram
was obtained with increased peak heights. Usually, three standard additions
were made for a total of four polarograms. The original metal concentrations
were determined from the change in peak height based on the known additions.
Blanks comprised of diluent nitric acid, hydroxylamine and citrate buffer were
run daily. The contamination measured was subtracted from the sample values.
Several tests were performed to evaluate the precision and accuracy of
the entire procedure. The concentrations of 37 sample blanks over a period of
several months yielded Zn = 1.27 ± .57 ppb; Cd = .03 ± .06 ppb; Pb = .59 ± .22
ppb. A similar study on a synthetic sample was conducted. A sample con-
152
-------
sisting of (approximately) 30 ppb Zn, 3 ppb Cd, and 4 ppb Pb, was made.
Twelve aliquots of this sample were withdrawn over a period of about 7 months
and analyzed. The results were Zn = 24.64 ± 2.95 ppb; Cd = 3.93 ± 2.39 ppb;
and Pb = 4.33 ± 2.80 ppb.
This test gives a realistic estimate of the accuracy and precision of the
entire procedure. The precision is good, considering the concentration levels
of these particular metals. It might be improved somewhat by using a higher
purity nitric acid in the digesting procedure and even stricter controls
against contamination (using a laminar flow hood, for example). The accuracy
is difficult to evaluate. -The difference between the measured concentrations
and the 'true1 concentrations in the sample could have resulted from 1) a
small error in the 'true1 concentrations magnified by successive dilutions
involved in making the standards, 2) loss (or gain) of metal by adsorption (or
desorption) on the sides of the volumetric and storage bottles during the
sample preparation and storage, and/or 3) a statistically significant in-
accuracy in the analytical procedure and method. Even so, the data are reli-
able enough to permit their interpretation and use in models of the system,
because the errors associated with these models are often larger than the
analytical errors.
Sediment Solids
Sediments that had previously been squeezed for pore waters and then
frozen, were removed from their polyethylene bags, placed in polystyrene
weighing boats, and dried under heat lamps at ~60°C. The station 83 core
collected on 4/11/77 was carefully sectioned, weighed wet and weighed dried
for water content determination. The dried sediment was ground in a Spex
Mixer Mill for ten minutes using a hardened steel container and tungsten
carbide balls. Glass vials were used for storage of the sediment until it was
needed for analysis.
153
-------
Carbon
Total carbon was determined with a LECO induction furnace connected to a
hydroxide absorption pipette determinator. Ground sediment was oven-dried at
65°C and dessicated prior to weighing 0.05-0.15 grams into a LECO crucible.
Iron and copper chip accelerations were added to the crucible and gently mixed
with the sample. The prepared sample was combusted in a stream of oxygen at
>1600°C. Liberated gases were collected in a gas burette. When combusion was
complete, the gases were passed to a potassium hydroxide absorption pipette
where they were allowed to react for one minute. The remaining gases were
returned to the gas buret. Percent carbon was determined by noting the volume
change after CO adsorption.
Organic carbon was determined as total carbon remaining after acid pre-
treatment. The sample containing crucible was placed under heat lamps at
~60°C. The sample was wetted with a few drops of deionized water and then 1 N
HC1 was added dropwise to drive off inorganic carbon. The crucible was swirled
and acid was added as needed until fizzing stopped. Then, a few more drops of
acid were added. The sample was left under the heat lamps overnight to dry.
These samples were covered with aluminum foil and stored in air until run.
Inorganic carbon was determined by difference between total carbon and organic
carbon.
Sulfur
Total sulfur was determined with a LECO induction furnace connected to an
automatic buret. Ground sediment was oven-dried at 65°C and dessicated prior
to weighing 0.2 to 0.5 grams into a LECO cruicble. Copper chip accelerator
was added to the crucible an<3 gently mixed with the sample. The prepared
sample was combusted in a stream of oxygen. Combustion gases were passed to
the automatic buret where sulfur was determined idiometrically.
154
-------
Acid volatile sulfur was determined idiometricially. Freshly collected
sections of wet sediment (10 to 20 g wet weight) were placed in 300 ml erlen-
meyer flasks and kept under N_. So that the content of acid volatile sulfur
could be expressed in moles/mass dry sediment, an aliquot of each section was
analyzed for water content. The wet sediment sections were reacted with
concentrated HCl (10 ml, slow addition) in the presence of SnCl (0.1 g SnCl_/
g wet sediment). The acid-sediment mixture gently stirred.
Evolved gases were passed to a trap containing a saturated zinc acetate
solution where H-S contained in the reaction gases was precipitated as ZnS.
The reaction was allowed to proceed for 30 min. A flow of N_ was maintained
at all times. At the end of the reaction period, the zinc acetate trap was
removed for analysis. The amount of ZnS precipitated was determined idio-
metrically.
Phosphorous
The perchloric acid digestion of Sommers and Nelson (1972) was used for
the determination of total phosphorus. Aliquots of 0.1-0.3 ground dessicated
sediments were placed 50-ml graduated Folin-Wu N.P.N tubes. Perchloric acid
was added and the digestion was effected by heating the tubes for 75 minutes
at 205°C in an aluminum heating block. The tubes were cooled, diluted, vortex
mixed, and left to stand while the particles settled out. An aliquot of the
supernatant liquid was withdrawn, neutralized with 5N NaOH to a p-nitrophenol
endpoint, and partially diulted in a volumetric flask. The single solution
reagent of Murphy and Riley (1962) was added and the contents of the flask
were brought to volume. Absorbance of the phosphomolybdenum blue complex was
read with a Beckman DU spectrophotometer at 880 nm after one hour.
Total Kjeldahl Nitrogen
155
-------
A modified digestion procedure proposed by Nelson and Sommers (1972)
based on the work of Bremner (1965) was employed for the determination of
total Kjeldahl nitrogen. Ground sediments were oven-dried at 65°C and des-
sicated before weighing 0.05-0.25 grams into 50-ml graduated Folin-Wu N.P.N.
tubes. The tubes were placed into an aluminum heating block and the sediment
was digested at >350°C by a mixture of sulfuric acid and potassium sulfate.
Copper sulfate and selenium acted as catalysts. Heating was continued for
three hours after clearing. The samples were cooled and neutralized with 50%
NaOH to a phenolphthalein endpoint. Sulfuric acid (5N) was then added dropwise
to just clear the pink color. The tubes were diluted to volume, vortex mixed,
and particulates allowed to settle. An aliquot was withdrawn and diluted to
five ml. Nitrogen was determined as ammonia using the method of Solorzano
(1969).
Metals
Approximately 0.1 g of dessicated ground sediment was weighed into the
Teflon cup of a PARR 4745 acid digestion bomb. The hydrofluoric, nitric and
perchloric acid digestion mixture used was that suggested by Agemian and Chau,
(1975). The liquid resulting after heating the bomb at 140°C for 3.5 hours
was mixed with boric acid and made to volume in a volumetric flask. Samples
thus prepared were stored in polyethylene bottles until needed for analysis.
Metals were determined on a Perkin Elmer model 360 atomic adsorption spectro-
photometer.
Bacterial Abundance
Bacterial numbers were determined after the methods of Watson et. al
(1977). Approximately one gram aliquots of wet sediment were placed in pre-
weighed bottles containing 10 ml of 2% gluteraldehyde and shaken. A seperate
determination of sediment water content was also made. The gluteraldehyde
156
-------
treated samples may be stored or counted immediately. The sample was diluted
by a factor of 1,000 to 10,000 (2,000 x was found to be the optimum dilution).
A 10 ml aliquot of diluted sample was stained with acridine orange (three
minute contact time). A 2ml aliquot of stained sample was drawn through a
0.2um Nucleopore filter (previously stained with Irgalan black) using a maxi-
mum pressure differential of 0.5 atm. the filter was then placed on a micro-
scope slide with a drop of immersion oil and covered with a coverslip. The
slides were examined with a Leitz orthoplan microscope fitted with a Ploemopak
2.2 fluorescence vertical illuminator and 150 W high pressure xenon lamp.
Fluorescing bacteria in twenty fields were counted and the data averaged.
Bacterial abundance was reported as number of bacteria per gram dry weight of
sediment.
157
-------
II. FIELD CORE DATA
158
-------
STATION: GASP XXX1-83
DATE: 4-1X-77
U)
Sampling Water Total
Interval Content C
(era) % %
0-1
1-2
2-3
3-4
4-5
5-6
6-7
7-8
8-9
9-10
10-11
11-12
12-13
13-14
14-5
15-16
16-17
17-18
18-19
19-20
24-25
29-30
34-35
39-40
44-45
49-50
54-55
59-60
64-65
69-70
86.0
77.8
75.1
71.1
67.0
65.8
64.6
63.3
63.5
63.4
62.4
62.4
61.5
60.5
58.8
58.6
58.8
59.3
59.3
56.2
56.2
56.1
55.3
54.0
52.0
52.7
52.1
53.7
44.8
3.58
2.42
1.68
1.14
1.47
1.50
1.20
1.35
1.34
1.28
1.35
1.34
1.28
.36
.51
.49
.47
.57
.14
2.32
2.12
1.91
1.98
Organic
C
3.83
2.27
1.74
1.52
1.41
1.18
1.32
1.29
1.29
1.19
1.13
1.36
1.21
1.29
1.05
1.13
0.92
1.23
1.06
1.03
0.86
1.12
0.80
Kjeldahl Total
N P
MM/g MM/g
326
220
182
155
127
130
126
112
118
124
110
132
112
122
109
111
101
124
99
108
106
102
103
97
91
76
88
91
72
30.1
22.6
21.2
22.8
23.0
22.6
22.2
22.9
21.8
22.6
22.0
21.0
21.8
22.4
21.6
21.0
20.9
21.2
19.8
22.2
20.8
21.6
20.3
21.4
21.4
20.1
19.4
18.6
20.6
Total Acid
S Volatile
MM/g S
MM/g
47.32
29.42
22.60
17.15
31.19
55.41
53.95
40.23
15.28
14.66
32.44
29.00
13.09
8.42
16.21
14.88
9.62
10.11
10.49
5.65
1.73
2.27
1.27
1.06
0.58
0.29
0.27
0.29
0.28
0.26
0.25
0.23
0.27
0.34
0.25
1.20
0.79
0.85
1.70
2.99
1.65
0.38
1.57
0.62
Ca
pM/g
228
300
150
687
166
3.06
159
225
180
142
137
132
167
141
129
176
223
208
200
162
305
167
311
173
430
178
575
747
599
Mg
612
70S
602
1011
560
731
565
653
542
553
559
578
556
604
584
601
586
619
609
556
682
545
730
622
876
572
940
900
780
Fe
MM/g
650
580
580
550
640
620
640
550
640
620
630
640
610
620
640
620
580
640
600
570
590
580
630
640
580
540
580
550
620
Mn
MM/g
9.80
6.01
7.00
8.36
6.23
7.71
5.55
6.93
7.56
4.95
6.03
7.41
6.71
7.51
4.75
6.82
6.12
6.00
7.34
6.95
6.78
7.09
6.81
7.05
8.08
5.92
7.20
7.34
6.92
Zn Cd
MM/g MM/g
5.13
1.56
2.68
1.78
2.20
2.03
1.88
1.97
1.62
1.98
1.81
1.87
2.01
1.37
1.83
1.49
1.48
1.90
1.54
1.78
1.23
1.57
1.41
2.18
1.64
2.89
1.54
1.46
1.86
Pb Cu
pM/g |jM/g
1.10
0.43
0.54
0.42
0.37
0.39
0.42
0.36
0.63
0.55
0.36
0.45
0.42
0.36
0.54
0.53
0.57
0.51
0.53
0.42
0.60
0.37
0.61
0.63
0.62
0.54
0.54
0.46
0.54
Table 2. Field Core Data.
-------
STATION: GASP XXX1-83
DATE: 17-V1-78
Sampling Water
Interval Content
(cm) %
0-2
2-3
3-5
5-6.5
6.5-8.5
8.5-11.5
11.5-16
16-21
21-25
25-29.5
Total Organic Kjeldahl
C
%
3.14
2.19
1.80
1.68
1.82
1.89
2.32
2.44
2.64
2.78
C
%
2.11
1.32
1.00
0.84
0.76
1.09
0.75
0.81
0.94
0.79
N
|Jhl/g
1.70
1.04
0.80
0.59
0.55
0.50
0.46
0.46
0.41
0.48
Total
P
pM/g
26.6
21.7
10.0
18.4
19.8
20.2
19.2
18.7
18.9
18.0
Total Acid
S Volatile Ca
pM/g S pM/g
MH/g
513
494
531
548
658
697
763
1002
1218
1364
Mg
MM/g
725
670
672
649
744
822
900
1003
1031
921
Fe
MM/g
510
480
500
450
470
470
460
460
430
450
Mn
pM/g
6.87
6.59
4.39
5.14
5.10
7.04
6.79
7.14
7.30
7.27
Zn
MM/g
2.61
1.71
1.28
0.85
1.24
1.07
0.83
1.22
1.10
0.98
Cd Pb
MM/g MM/g
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
Cu
MM/g
69
54
43
12
26
26
10
25
00
13
STATION A-l
DATE: 6- IX- 7 9
0-2
2-4
4-6
6-8
8-10
10-13
17-21
29.33
37-41
45-49
4.21
4.42
4.65
3.68
4.00
3.56
3.03
1.39
1.43
3.68
3.89
3.79
3.48
3.43
3.17
2.71
1.41
1. 11
3.71
4.17
4.50
3.16
3.38
2.94
2.28
1.39
1.36
42.2
46.0
36.0
40.6
36.2
38.7
30.8
23.7
22.2
367
245
185
236
294
247
178
127
143
594
582
595
589
607
619
632
599
591
810
600
640
650
670
660
640
610
660
9.73
7.48
7.76
9.47
10.5
7.82
9.15
9.14
8.58
8.21
7.30
5.15
6.85
6.30
5.62
4.49
2.16
1.95
0.
0.
0.
1.
0.
1.
1.
0.
0.
84
78
92
04
96
01
01
56
52
Table 2. Field Core Data (cont'd).
-------
STATION: GASP
DATE:
Samp 1 e
No.
0
1
2
3
4
5
6
7
8
9
10
XXX 1-83
3I-V-78
I.W.
Depth
(cm)
I.W
0-2
2-3.5
3.5-5.5
5.5-7
7-9.-S
9.5-12
12-17
17-21
21-26
Carb.
Alk.
pH meq/l
1.87
8.05 1.64
7.61 2.13
7.68 2.30
7.68 2.68
7.89 2.80
8.01 2.85
7.95 3.01
7.82 3.08
7.87 3.11
32-36.5 7.97 3.08
STATION: GASP
DATE:
0
I
2
3
4
5
6
7
8
9
10
XXXI 1-83
neq/l
.544
.585
.605
.599
.597
.603
.605
.612
.625
.630
.632
Fed 1)
ug-at/l
18.0
50.6
28.2
60.9
55.9
61.5
52.7
49.9
48.5
33.6
SRP
og-at/l
O.I
54
12
17
22
25
20
17
25
26
10
pg-at/l
67
47
43
47
62
28
92
21
100
107
85
ug-at/l
13
59
90
104
124
102
179
97
246
185
151
SI02
ug-at/l
243
462
525
545
603
603
507
482
518
462
Zn
ng-9t/l
629
1874
272
414
292
1536
186
143
1 1835
507
230
Cd
ng-at/l
16
22
4
0
-4
-1
-5
-10
4
-4
56
Pb
ng-at/l
14
20
6
16
8
72
4
6
125
74
5
Cu Mn
ng-at/l ug-at/l
78
253
161
92
47
72
17
-4
270
57
42
Ca
Viq-ot/l
900
1050
1300
1250
.
_
1400
1150
1800
1200
1 150
Mg Na K
ug-at/l ug-at/l uq-at/l
518
367
377
467
356
-
509
567
464
493
~
I7-VI-78
O.W.
0-2
2-3
3-5
5-6.5
6.5-8.5
8.5-11.5
II. 5-16
16-21
21-25
25-29.5
1.74
7.69 1.96
7.74 1.99
- 2.29
7.80 2.47
7.72 2.54
7.84 2.73
7.86 3.00
7.94 3.00
7.97 3.07
8.01 3.05
.586
.650
.679
.614
.530
.645
.601
.628
.614
.650
.613
0
0.7
18.8
48.5
32.8
87.1
43.4
32.0
32.6
22.5
0.03
1.9
1.9
4.3
14
30
16
12
13
15
13
67
21
54
10
68
50
97
64
50
54
61
40
114
57
132
123
125
154
127
107
III
126
61.5
257
287
658
695
710
712
582
545
535
503
1957
548
927
365
631
386
164
325
825
186
588
2
13
232
19
25
26
2
9
3
8
3
103
6
25
17
39
12
38
7
3
-II
3
37
60
107
113
115
16
45
-25
24
71
13
950
1200
_
_
_
_
_
1100
1250
1250
1100
Table 2. Field Core Data (Cont'd)
-------
N>
STATION;
DATE:
Sample
No.
0
1
2
3
4
5
6
7
8
9
10
STATION:
DATE:
0
1
2
3
4
5
6
7
8 32
9 41
10
GASP
XXXI II -83
2I-VII-78
I.W.
Depth
(cm)
O.W.
0-4
4-7
7-10
10-12
12-14
14-16
16-19
19-23
23-26
26-30
GASP
16- VI
O.W
0-2
2-5
5-7
7-9
9-11
11-13
19-25
.5-38
.5-47
47-52
PH
XXXIV-83
11-78
7.24
7.24
7.30
7.25
7.25
7.29
7.28
7.39
7.48
7.56
Carb.
Alk.
meq/l
1.87
2.33
2.20
2.42
2.56
2.51
2.62
2.78
2.93
2.76
2.91
1.82
1.68
1.73
2.10
2.28
2.32
2.44
2.66
3.01
3.06
2.97
Cl"
meq/l
.485
.655
.611
.580
.598
.593
.581
.859
.593
.575
.657
Fe< 1 1 >
wg-at/l
1.2
37.6
42.5
36.7
24.5
24.1
46.8
33.8
7.0
29.9
12.4
41.6
43.4
-
41.4
58.1
58.4
25.0
10.7
30.2
SRP
pg-at/l
0.18
3.6
21.7
28.1
31.9
29.9
13.7
19.7
15.7
8.9
16.1
0.85
6.5
21.9
7.0
12.9
9.9
11.8
18.0
9.8
6.0
11.8
N02-N03
ug-at/l
52
290
13
191
9
161
2
151
0
151
0
30.7
18.8
0
47.8
0
0
2.6
141.4
0
0
25.2
NHj
pg-at/l
9.1
173
135
252
186
146
145
187
125
138
142
5.3
74
61
109
71
110
130
191
121
94
166
SI02
lig-at/l
27.5
412
688
728
680
610
578
615
592
588
565
31.8
270
490
393
569
604
613
718
661
639
600
Zn
ng-at/l
285
815
373
801
996
1158
749
466
461
326
293
197
678
767
692
592
566
748
33344
562
70
1020
Cd
ng-at/l
16
II
II
21
55
20
10
10
3
3
4
2
22
18
17
10
15
15
15
10
0
8
Pb
ng-at/l
7
22
a
24
19
35
19
28
3
3
5
5
36
3
8
5
10
12
52
II
4
5
Cu
ng-at/l
106
74
-28
13
93
349
36
-7
-39
-54
-44
II
40
-7
10
2
9
-32
122
22
20
13
Mn
wg-at/l
0.2
43.0
33.1
28.2
25.9
25.7
3.1
21.4
21.7
22.3
25.8
25.5
25.1
43.2
31.9
29.3
Ca
pg-at/l
900
1150
1 150
1050
1050
1050
950
1000
1100
1050
1000
1000
1000
1500
1100
1050
Mg
wg-at/l
206
254
302
262
260
266
228
235
226
214
223
219
216
349
293
304
Na
wg-at/l
489
614
508
498
469
470
480
503
541
540
606
560
511
501
491
511
K
wg-at/l
14.7
57.2
34.9
30.5
19.2
23.7
46.7
49.0
51.3
55.1
55.8
55.8
55.8
75.2
59.6
41.5
Table 2. Field Core Data, (cont'd)
-------
STATION: GASP XLI-AI
DATE: 6-IX-79
~TTW
Depth
Carb.
Alk.
Cl~ Fed I) SRP N02-N0j NH^ SIO, Zn Cd Pb Cu Mn Ca Mg Na
meq/l gg-at/l ug-at/l ug-at/T tig-at/l gg-at/l ng-at/l ng-at/l ng-at/l ng-at/l pg-at/1 ug-at/l pg-at/l pg-at/l
.796 .36 5.4 52.4
Samp Ie
No.
(cm) pH meq/l
K
ug-at/l
0
I
2
3
4
5
6
7
8
9
10
O.W.
0-2
2-4
4-6
6-8
8-10
10-13
17-21
29-33
37-41
45-49
1.95
31
76
2.92
2.49
2.92
3.01
3.28
3.67
3.44
.796
.632
.581
.556
.594
.606
.568
.606
.606
.581
127
156
197
226
189
190
195
228
164
124
.36
8.1
31.6
47.8
59.4
41.6
44.4
39.2
44. I
30.8
29.7
5.4
145
145
207
145
187
232
235
250
281
276
52.4
291
333
466
541
707
749
841
915
874
832
Table 2. Field Core Data (cont'd).
-------
STATION: GASP-XL 1 -A 1
Peeper
DATE: 6- IX- 79
Samp 1 e
No.
1
2
3
4
5
6
7
e
9
10
II
12
13
14
15
18
19
22
24
25
26
27
28
29
1 .W.
Depth
(cm)
-10
-5
-2
0
1
2
3 '
4
5
6
7
8
9
10
12
18
20
32
40
50
60
70
80
90
Carb.
Alk.
pH meq/ 1
2.08
2.60
2.90
2.94
3.01
2.98
2.99
2.88
2.90
2.94
3.08
3.19
3.51
3.58
3.58
3.24
3.67
Cl'
meq/l
606
581
556
556
530
543
467
518
505
556
518
505
480
430
Fe(ll)
lig-at/l
58.7
48.0
.52.8
38.3
52.1
54.0
52.0
55.9
63.1
62.2
58.5
58.7
75.7
101
82.7
89.2
SRP
pg-at/l
1.9
30.5
30.7
49.2
29.2
44.2
42.8
40.4
37.5
37.8
33.4
24.9
36.8
33.9
35.5
30.7
52.9
N02-N0j NHj
pg-at/T pg-at/l
55
79
91
no
119
lie
115
122
130
157
289
118
250
221
146
198
206
SIO, Zn Cd Pb Cu Mn Ca Mg Na K
pg-at/l ng-at/l ng-at/l ng-at/l ng-at/l pg-at/l pg-at/l pg-at/l pg-at/l pg-at/l
91.5
358
566
599
599
599
599
599
583
649
674
658
583
574
558
458
541
Table 2. Field Core Data (cont'd).
-------
III. PRODUCTION EXPERIMENT DATA
165
-------
I—1
CTi
O"*
JARS t
DATE:
Sample
No.
1
2
3
4
5
6
7
= 2 days
20-IX-78
I.W.
Depth
(cm)
0-2
2-4
4-6
6-10
10-14
14-13
35-39
Cart.
Alk.
pH meq/l
3.84
3.22
2.74
2.89
2.36
2.16
2.47
Cl" Fe(ll) SRP N02-N0j
meq/l pg-at/I iig-at/l pg-at/I
0.566 179.8 120.4
0.567 85.0 «I20
0.574 19.2 15.5
0.583 34.4 5.1
0.599 39.4 7.7
0.610 40.8 15.5
0.630 -1.6 1.7
•f
NH*
pg-at/I
39.0
0
0
0
0
117.0
0
SIO,
pg-at/I
980
1025
846
655
733
733
756
Zn
ng-at/l
107
174
312
751
188
158
423
Cd
ng-at/l
10
19
13
1 1
5
1
17
Pb
ng-at/l
6
8
-1
8
-3
6
9
Cu
ng-at/l
-1 13
-2
15
17
-42
0
47
Mn
pg-at/I
69
41
25
24
20
21
19
Ca
pg-at/I
1400
1300
1050
1 100
900
800
850
Mg
pg-at/I
529
461
446
431
348
328
407
Na K
pg-at/I pg-at/I
Table 3. Jar Experiment Data.
-------
JARS t = 22 days
Low Temp.
DATE: IQ-X-78
Zn Cd Pb Cu Mn Ca Mg Na K
I ng-at/l ng-at/l ng-at/l ng-at/l pg-at/l vig-at/l pg-at/l pg-at/l pg-at/l
711 7 I -74 71 nnn Knn
Samp Ie Depth
'No. (cm) pH
Carb. .
Alk. Cl" Fed I) SRP N02-N03 NH^ SIO,
meq/l roaq/l pg-at/l pg-at/l pg-at/l pg-at/l pg-at/
0-2
2-4
4-6
6-10
10-14
14-18
35-39
4.89
4.55
75
20
98
60
2.57
0.555
0.560
0.557
0.599
0.657
0.644
0.746
154.1
137.2
54.3
131.6
114.1
61.5
38.0
46.5
138.9
12.0
38.5
29.8
21.3
11.5
204
154
213
196
217
131
171
163
637
452
494
502
534
540
211
54
316
199
177
231
1170
7
10
9
4
4
-I
0
-74
-15
44
-23
-7
-24
-2
71
56
39
47
37
26
30
1700
1600
1500
1650
1400
950
1000
608
609
593
656
546
352
502
JARS t = 24 days
Mod I urn Temp.
DATE: I2-X-78
1
2
3
4
5
6
7
JARS t
0-2
2-4
4-6
6-10
10-14
14-18
35-39
= 23 days
3.14
5.19
4.54
3.82
3.40
2.88
2.92
0.348 265.5
0.363 190.0
0.412 102.0
0.384 86.3
0.427 85.3
0.420 64.6
0.448 28.6
35.9
55.1
66.8
26.0
24.9
23.7
15.3
307
352
248
243
142
167
444
810
759
680
582
584
647
584
46
-3
10
18
46
47
26
5
2
2
2
2
1
3
4
0
0
3
1
3
3
-46
-27
-37
-22
-19
-1 1
7
79
59
47
36
30
28
26
1900
1850
1750
1600
1250
950
950
701
719
679
596
487
382
454
High Temp.
DATE:
Samp 1 e
No.
1
2
3
4
5
6
7
ll-X-78
I.W.
Depth
(cm) pH
0-2
2-4
4-6
6-10
10-14
14-18
35-39
Carb.
Alk.
meq/l
-6.60
6.16
5.48
4.25
3.82
2.97
2.83
'Cl~ Fe(ll)
meq/l pg-at/l
0.345 360.1
0.368 217.8
0.384 165.9
0.376 87.0
0.507 77.6
0.453 54.3
0.484 3.1
SRP
ug-at/l
54.1
54.9
69.9
25.1
28.9
23.4
8.5
N02-N0j* NHa
pg-at.l pg-at/l
506
378
249
250
323
192
212
SIO,
pg-at/l
894
835
825
601
553
639
528
Zn
ng-at/l
105
72
150
70
93
95
167
Cd
ng-at/l
2
3
2
3
3
4
4
Pb
ng-at/l
4
0
9
5
15
12
19
Cu
ng-at/l
-127
-87
-26
4
-II
9
15
Mn
wg-at/l
99
66
57
43
34
26
23
Ca
pg-at/l
2250
2050
2100
1700
1350
1000
950
Mg Na K
pg-at/l pg-at/l pg-at/l
838
782
857
707
545
396
448
Table 3. Jar Experiment Data (cont'd)
-------
OO
JARS t = 71 days
Low Temp.
DATE:
Sampl
No.
1
2
3
4
5
6
7
JARS
: 28-XI-78
I.W.
e Depth
(cm) pH
0-2
2-4
4-6
6-10
10-14
14-18 •
35-39
t = 73 days
Carb.
Alk.
meq/l
6.03
5.67
5.12
4.13
3.73
2.88
3.08
cr
meq/l
0.509
0.548
0.576
0.576
0.597
0.624
0.613
Fe(ll)
ug-at/l
377.1
235.9
153.2
122.0
97.7
85.2
60.7
SRP
ug-at/l
37.4
39.7
88.9
21.8
26.2
19.3
22.4
N02-N0
ug-at/l
14.4
13.0
21.2
17.5
20.8
-
55.2
» +
ug-at/l
424
461
347
273
243
228
243
SIO,
ug-at/l
698
678
570
451
480
524
525
Zn
ng-at/l
66
3
15
174
172
304
304
Cd
ng-at/l
5
5
6
5
12
14
4
Pb
ng-at/l
5
4
6
3
5
4
10
Cu
ng-at/l
-18
5
-5
12
33
28
48
Mn
ug-at/l
98
63
49
33
30
26
24
Ca
ug-at/l
2250
2100
1850
1750
1400
1150
1200
Mg Na K
ug-at/l ug-at/l pg-at/l
872
809
781
742
572
497
582
Medium Temp.
DATE:
1
2
3
4
5
6
7
JARS
High
DATE:
1
2
3
4
5
6
7
30-XI-78
0-2
2-4
4-6
6-10
10-14
14-18
35-39
t = 72 days
Temp.
29-XI-78
0-2
2-4
4-6
6-10
10-14
14-18
35-39
5.73
5.54
5.25
5.00
3.40
3.02
3.53
8.22
7.78
7.22
6.09
5.67
3.95
3.67
0.566
0.588
0.632
0.828
0.721
0.666
0.718
0.607
0.618
0.638
0.639
0.622
0.666
0.637
431.9
243. B
140.3
182.6
87.4
70.4
62.2
631.6
404.0
315.8
206.3
42.9
109.7
19.7
25.1
32.9
75.5
47.5
21.9
18.6
19.3
56.2
55.4
76.4
77.7
43.8
24.5
3.7
it
9.77
13.2
6.8
6.1
9.7
8.3
II .9
«
22.6
14.6
20.5
35.2
22.6
19.9
35.7
579
458
349
328
257
227
233
1092
953
579
335
450
334
413
868
828
740
634
541
647
596
1087
1072
1028
894
749
804
659
266
-65
31
40
148
146
104
138
86
187
170
143
237
133
6
6
8
3
10
7
7
46
69
24
40
31
51
133
3
-1
-2
-4
-1
0
0
-1
a
-4
0
-8
-1
-9
-37
-110
-25
-56
-36
-28
-47
-72
-14
-34
20
-15
2
-53
91
69
54
46
24
24
29
137
96
88
57
50
34
27
2250
2200
2150
2000
1350
1150
1250
2900
2850
2800
2500
2100
1500
1300
863
889
924
840
603
537
682
1143
1185
1123
996
967
685
720
"Frozen and run I month later.
Table 3. Jar Experiment Data (cont'd).
-------
ON
vO
JARS t = 143 days
Low Temp.
DATE:
Sarnp 1 e
No.
1
2
3
4
5
6
7
JARS t
Me d 1 um
DATE.
1
2
3
4
5
6
7
JARS t
8-JX-79
I.W.
Depth
(cm)
0-2
2-4
4-6
6-10
10-14
14-18
35-39
= 142 days
Temp.
7-JJ-79
0-2
2-4
4-6
6-10
10-14
14-18
35-39
= 143 days
Carb.
Alk.
pH meq/ 1
7.35 7.31
7.33 8.90
7.42 7.35
7.40 6.46
7.42 5.06
7.39 3.99
7.46 4.25
7.25 9.90
7.34 8.87
7.39 6.94
7.43 6.90
7.24 6.45
7.26 5.12
7.51 5.49
Cl~
meq/l
0.720
0.888
0.790
0.684
0.690
0.684
0.672
0.716
0.629
0.673
0.662
0.646
0.716
0.678
Fe( 1 1 )
wg-at/l
364.5
340.4
156.6
156.8
82.5
68.1
48.4
543.2
379.3
452.7
394.0
398.5
281.0
161.4
SRP
iig-at/l
31.8
23.6
59.5
46.4
20.5
14.7
15.2
42.7
47.7
68.8
33.2
34.0
23.1
23.8
N02-N05
vg-at/T
9.9
13.2
12.9
14.3
20.0
21.4
26.5
6.5
9.6
7.0
19.0
12.4
24.0
24.0
NHJ"
wg-at/l
429
722
563
297
285
238
260
1011
725
175
298
310
306
488
SIO,
wg-at/l
847
784
647
549
578
573
532
916
875
804
735
896
714
553
Zn
ng-at/l
60
121
254
130
112
259
201
50
98
80
127
322
77
-22
Cd
ng-at/l
-5
4
-6
1
-2
-4
6
6
7
1
4
7
-2
0
Pb
ng-at/l
-2
1
-4
5
1
10
-3
7
6
1
3
9
-5
-4
Cu
ng-at/l
-68
16
-88
25
II
51
5
-116
-88
-50
-15
29
-59
-40
Mn
wg-at/l
91
76
57
45
32
26
22
123
93
64
61
57
42
45
Ca
ng-at/l
2350
2500
2300
2050
1600
1350
1350
2750
2500
2250
2300
2050
1600
1600
Mg Na K
wg-at/l wg-at/l wg-at/l
917
1057
997
872
686
630
627
1100
1094
935
911
858
674
804
High Temp.
DATE:
1
2
3
4
5
6
7
8-TJ-79
0-2
2-4
4-6
6-10
10-14
14-18
35-39
6.98 14.18
7.06 11.56
7.15 10.65
7.16 9.95
7.10 7.66
7.47 5.13
7.50 5.72
0.733
0.672
0.695
0.697
0.700
0.710
0.697
832.3
553.8
351.7
350.0
269.4
31.38
73.47
74.3
73.3
72.3
88.2
39.0
14.5
19.4
12.4
15.6
17.1
18.6
19.4
31.9
20.6
2483
1139
740
419
349
389
416
1289
1667
1157
1215
1376
916
927
0
-76
-45
73
108
245
33
6
6
8
6
6
14
-6
-2
0
3
-3
3
4
1
-71
-57
-51
-72
-40
97
29
172
117
95
87
61
45
45
3750
3300
3100
3050
2400
1700
1700
1351
1292
1258
1244
971
760
841
Table 3. Jar Experiment Data (cont'd).
-------
JARS t = 197 days
Low Temp.
DYTE:
Sample
No.
1
2
3
4
5
6
7
JARS t
Med 1 urn
DATE:
1
2
3
4
5
6
7
JARS t
3-IV-79
I.W.
Depth
(cm)
0-2
2-4
4-6
6- 10
10-14
14-18
35-39
= 198 days
Temp.
4-IV-79
0-2
2-4
4-6
6-10
10-14
14-18
35-39
= 197 days
Carb.
Alk.
pH meq/l
7.30 8.26
7.38 8.60
7.54 7.87
7.54 5.85
7.51 5.55
7.45 4.67
7.58 5.01
7.13 11.01
7.30 9.36
7.29 8.75
7.32 7.58
7.36 8.31
7.24 3.58
7.38 5.54
Cl~
meq/1
0.714
0.711
0.678
0.629
0.685
0.733
0.665
0.708
0.675
0.678
0.650
0.667
0.672
0.656
Fe(ll)
pg-at/l
488.3
390.5
239.9
129.4
148.1
123.3
98.7
740.4
429.3
294.3
290.0
184.9
170.2
99.5
SRP
pg-at/l
41.3
32.8
46.9
30.9
31.9
24.9
21.1
45.9
44.3
59.3
74.7
32.2
21.1
28.1
N02-N0
M9-at/l
13.5
13.8
16.0
16.5
29.5
24.5
23.1
40.4
39.2
40.0
39.7
56.9
57.1
44.1
3 NH4
pg-at/l
772
864
565
304
269
335
307
1510
1060
692
360
321
378
443
SIO?
pg-at/l
832
822
649
497
564
537
490
785
793
766
694
650
723
629
Zn
ng-at/l
-80
96
246
268
240
176
155
222
-10
161
13
154
300
255
Cd
ng-at/l
-5
4
1
0
7
14
6
5
13
9
2
3
5
3
Pb
ng-at/l
3
5
3
-1
6
0
0
5
5
5
-1
16
3
5
Cu
ng-at/l
-32
5
-20
-8
69
6
-35
-47
-49
34
73
106
189
355
Mn
pg-at/l
III
85
70
55
34
42
46
142
96
83
71
44
45
39
Ca
(ig-at/l
2350
2550
2450
2100
1850
1500
1600
3000
2750
2800
2600
1900
1700
1650
Mg Na K
pg-at/l wg-at/l pg-at/l
917
1059
956
875
710
590
688
1156
1040
1064
1053
737
666
727
High Temp.
DATE:
1
2
3
4
5
6
7
3-IV-79
0-2
2-4
4-6
6-10
10-14
14-18
35-39
6.92 14.06
7.09 13.12
7.30- 11.01
7.29 8.77
7.04 9.08
7.28 6.09
7.41 6.19
0.623
0.815
0.703
0.677
0.685
0.707
0.707
683.3
849.9
453.6
316.2
384.2
183.0
94.0
70.8
73.5
77.6
83.6
29.3
23.7
19.0
7.9
22.5
20.6
21.6
15.9
19.6
36.8
2540
1630
952
397
441
377
421
1198
1185
1160
1020
914
932
872
145
214
276
43
152
199
175
5
6
7
3
5
5
6
3
7
16
2
10
12
14
-29
-16
23
-90
-101
44
1
169
137
103
78
83
60
44
3650
3950
3650
2900
2950
2050
1850
1385
1468
1322
1 133
1127
773
831
Table 3. Jar Experiment Data (cont'd).
-------
IV. FLUX EXPERIMENT DATA
171
-------
CORE R
FLUX EXPERIMENT
DATE: 20-X11-78
Sampling
Interval
(era)
0-1
1-2
2-3
3-4
4-5
5-6
8-10
14-18
22-26
Water
Content
59.9
58.3
56.0
56.3
55.5
54.6
52.7
51.8
Total
C
2.17
2.16
2.30
2.08
2.05
2.11
2.12
1.87
Organic
C
1.12
1.14
1.11
1.37
0.99
1.31
1.41
1.36
Kjeldahl
N
pM/g
0.92
0.93
1.01
1.02
0.94
1.00
1.02
0.95
Total
P
pM/g
22.8
22.2
21.9
21.2
21.1
21.7
22.8
23.1
Total Acid
S Volatile Ca
MM/g S pM/g
MM/g.
r
476
495
511
399
472
455
433
506
Mg
PM/g
657
676
682
687
669
672
647
683
Fe
pM/g
480
510
480
460
460
520
490
510
Mn
pM/g
6.75
5.42
5.95
5.78
5.34
5.05
5.46
5.43
Zn Cd Pb
pM/g pM/g pM/g
1.29
1.83
1.47
1.47
1.42
1.36
1.67
1.36
Cu
pM/g
0.17
0.41
0.60
0.31
0.31
0.39
0.48
0.19
CORE 6
FLUX EXPERIMENT
DATE: 23-V11-79
0-1
1-2
2-3
3-4
4-5
5-6
6-8
8-10
14-18
22-26
3.53
3.41
3.25
2.64
2.69
2.70
2.35
2.18
1.59
1.51
2.70
2.47
2.35
1.84
1.81
2.06
1.64
1.58
1.02
0.82
2.49
2.29
2.01
1.45
1.25
1.62
1.22
0.96
0.66
0.61
31
28
27
25
23
25
24
22
21
20
.0
.6
.8
.6
.0
.7
.4
.9
.0
.1
956
692
1027
939
570
753
477
427
496
514
716
704
735
775
684
727
708
647
660
631
600
530
550
470
470
480
500
520
510
480
16.1
9.27
6.62
4.77
5.46
7.01
6.03
5.44
5.16
5.25
3.41
3.02
2.90
2.48
2.40
2.80
2.72
1.59
0.88
1.29
0.52
0.52
0.71
0.51
0.39
0.68
0.93
0.34
0.12
0.13
Table 4. Flux Experiment Data.
-------
COKE T
FLUX EXPERIMENT
DATE: 27-111-79
Sampling
Interval
(cm)
0-1
1-2
2-3
3-4
4-5
5-6
6-8
8-10
14-18
22-26
Water
Content
59.8
57.5
56.8
56.2
55.6
54.4
54.4
53.2
52.1
51.3
Total
C
1.84
2.05
1.94
1.99
2.01
1.78
1.88
2.02
1.83
1.85
Organic
C
0.88
1.09
1.15
1.11
1.20
1.17
0.95
1.25
1.20
1.12
Kjeldahl
N
1.07
1.10
1.08
1.04
1.04
1.06
1.04
1.04
1.03
0.95
Total
P
pM/g
20.7
18.2
19.8
21.7
20.84
18.8
20.5
21.3
19.6
21.8
Total Acid
S 'Volatile Ca
pM/g S pM/g
MM/8
511
510
480
462
462
461
498
500
487
510
Mg
681
660
668
667
666
670
676
647
658
659
Fe
470
480
500
480
480
500
480
490
460
510
Mn
pM/g
6.71
6.23
6.26
6.96
5.28
5.91
5.88
5.67
5.00
6.04
Zn Cd Pb
pM/g pM/g pM/g
1.30
1.52
1.71
1.50
1.37
1.30
1.36
1.23
1.20
1.53
Cu
0.20
0.36
0.25
0.03
0.37
0.37
0.53
0.34
0.12
0.31
Table 4. Flux Experiment Data (cont'd)
-------
Core: HJ1C-G (Worms present)
Days after preparation: 71
Sample
Number
0
1
2
3
4
5
6
7
8
9
10
I.W.
Depth
(cm) pH
O.W.
0-1
1-2
2-3
3-4
4-5
5-6
6-8
8-10
10-14
14-18
Carb
Alk.
Meq/1
0.61
1.95
2.27
2.58
2.81
3.02
3.28
3.53
5.77
2.10
4.11
Cl"
Meq/1
.102
.202
.236
.276
.316
.376
.395
.447
.511
.570
.651
Fe(II)
pg-at/1
3.5
40.5
74.1
78.4
96.0
82.0
99.3
96.7
98.1
130.3
SRP
pg-at/1
0.1
81.6
114.8
71.2
61.5
64.0
63.6
69.8
60.0
62.0
59.0
NwT
pg-at/1
3
182
194
168
209
214
202
174
167
236
259
SiO
Mg-aE/1
137
513
627
730
820
886
924
951
957
976
959
-4-0
Mn*2
pg-at/1
0
27
31
33
38
35
39
44
48
48
48
Ca
pg-at/1
200
800
850
900
1200
1050
1250
1350
1500
1600
1750
Porosity
%
Core: HrtC-R (no worms)
Days after preparation: 72
0
1
2
3
4
5
6
7
8
9
10
O.W.
o-i
1-2
2-3
3-4
4-5
5-6
6-8
8-10
10-14
14-18
0.63
2.25
2.67
3.00
3.21
3.30
3.40
3.59
3.91
4.01
.097
.419
.376
.462
.477
.550
.595
.581
.823
.693
0
64.3
96.1
83.3
79.8
88.1
77.3
122.4
113.2
122.3
133.2
0.4
9.0
39.1
51.7
54.3
70.0
70.6
70.0
74.5
81.3
73.2
0
293
239
265
274
294
302
293
300
308
105
717
741
778
943
968
989
1010
1029
1111
1220
1
28
33
38
37
39
40
42
44
46
200
1100
1100
1400
1300
1350
1400
1450
1650
1550
Table 4. Flux Experiment Data (cont'd)
-------
Core: 11MC-A (worms present)
Days alter preparation: 100
Sample
Number
0
1
2
3
4
5
6
7
8
9
10
l.W.
Depth
(cm)
O.W
0-1
1-2
2-3
3-4
4-5
5-6
6-8
8-10
14-18
22-26
Carb
Alk.
Meq/1
1.33
2.81
3.15
3.42
.68
,00
.14
.18
.32
.43
4.68
Cl
Meq/1
Fe(II)
pg-at/1
0
99.1
137.7
132.5
145.3
155.9
149.3
157.4
169.3
156.5
191.0
SRP
pg-at/1
3.0
57.4
63.4
71.2
73.2
82.9
65.0
65.9
65.8
60.0
70.5
Mg-aE/1
34
224
213
187
207
199
203
200
199
194
217
SiO
pg-at/1
116
603
740
824
904
933
963
991
1004
952
984
Mn+2
pg-at/1
Ca+2
pg-at/1
Porosity
%
Core: HMC-N (no worms)
Days after preparation: 99
0
1
2
3
4
5
6
7
8
9
10
O.W.
0-1
1-2
2-3
3-4
4-5
5-6
6-8
8-10
14-18
22-26
7.14
7.23
7.37
7.36
7.34
7.40
7.31
7.25
7.26
7.52
7.40
0.58
2.49
2.79
3.06
3.21
3.37 •
3.46
3.46
3.75
3.96
4.00
.144
.400
.499
.434
.468
.480
.525
.539
.590
.722
.784
8.8
115.3
98.1
100.7
125.4
119.0
112.5
143.9
138.7
166.4
155.1
6.2
90.2
93.9
87.2
81.1
79.8
73.4
75.6
76.3
71.3
69.7
8.4
94.5
109.5
122.1
127.5
136.5
141.3
155.7
164.1
188.1
183.9
114
791
817
877
906
915
954
950
970
968
959
Table 4. Flux Experiment Data (cont'd)
-------
Core: HMC-S (no worms)
Days after preparation: 113
Sample
Number
0
1
2
3
4
5
6
7
8
9
10
I.W.
Depth
(cm)
O.W.
0-1
1-2
2-3
3-4
4-5
5-6
6-8
8-10
14-18
22-26
PH
7.20
7.36
7.10
7.13
7.15
7.19
7.23
7.23
7.26
7.16
Garb.
Alk.
Meq/1
0.72
1.42
1.67
2.07
2.42
2.72
2.98
3.33
4.39
4.60
Cl
Meq/1
.070
.179
.373
.418
.440
.493
.508
.552
.552
.764
.874
Fe(II)
pg-at/1
2.9
19.3
35.8
91.0
83.6
126.7
133.2
140.1
189.4
180.2
SRP
pg-at/1
1.1
9.0
13.3
28.8
32.2
49.4
55.6
60.4
62.5
51.5
NH
4
pg-at/1
7
52
59
84
99
116
133
138
190
194
SiO +2
2 Mn
pg-at/1 pg-at/1
198
1515
466
601
742
817
875
934
986
966
984
Core: IIMC-B (no worms)
Days after preparation: 120
0
1
2
3
4
5
6
7
8
9
10
O.W
0-1
1-2
2-3
3-4
4-5
4-6
6-8
8-10
14-18
22-26
7.14
7.14
7.27
7.24
7.28
7.53
7.30
7.22
7.29
7.20
7.22
1.0
2.26
2.78
3.22
3.54
3.44
3.93
4.20
4.48
5.00
5.15
4.8
76.3
78.7
79.4
71.7
83.2
106.9
108.3
118.0
115.1
110.1
11.7
97.8
109.0
113.4
106.3
54.4
105.7
98.5
105.9
94.2
93.9
18
104
117
133
144
150
163
176
186
212
217
133
1050
881
985
1029
501
1051
1046
1072
1466
1064
Table 4. Flux Experiment Data (cont'd)
-------
Core: HMC-Q (worms present)
Days after preparation: 120
Sample
Number
0
1
2
3
4
5
6
7
8
9
10
I.W.
Depth
(en.)
O.W.
0-1
1-2
2-3
3-4
4-5
5-6
6-8
8-10
14-18
22-26
PH
7.06
7.06
7.23
7.22
7.25
7.31
7.22
7.23
7.23
7.24
7.25
Carb
Alk.
Meq/1
0.78
2.39
2.69
2.96
3.17
3.30
3.46
3.89
4.22
4.64
4.97
Cl"
Meq/1
Fe(II)
pg-at/1
7.8
99.8
53.7
78.3
76.2
71.1
73.9
103.7
107.6
112.5
119.7
SRP
pg-at/1
17.7
128.9
116.3
109.5
92.7
102.1
105.9
102.1
106.9
100.2
104.2
NH*
Mg-at/1
14
256
259
201
179
201
154
179
221
193
218
SiO
Mg-at/1
188
819
749
828
906
968
1011
1035
1043
1048
1064
M *2 r +2
tin Ca
jjg-at/1 pg-at/1
Porosity
78.7
77.3
76.8
76.0
76.4
76.0
75.2
74.7
74.2
73.8
Core: HMC-0 (worms present)
Days after preparation: 133
0
1
2
3
4
5
6
7
8
9
10
O.W.
0-1
1-2
2-3
3-4
4-5
5-6
6-8
8-10
14-18
22-26
6.83
7.14
7.18
7.17
7.18
7.19
7.18
7.23
7.25
0.85
3.24
3.48
4.28
4.83
4.89
5.33
5.74
6.02
6.43
6.39
28.6
150.3
44.9
188.2
186.6
147.7
191.6
190.6
239.2
209.8
222.0
6.1
58.8
33.4
86.5
84.8
91.5
80.3
80.6
72.8
50.3
69.7
118
389
357
364
362
352
340
312
293
287
271
161
737
878
933
1002
1034
1036
1069
1058
1065
1078
79.0
76.8
76.8
75.7
75.7
75.5
75.0
74.8
73.6
73.8
Table 4. Flux Experiment Data (cont'd)
-------
Core: HMC-H (no worms)
Days after preparation: 134
00
Sample
Number
0
1
2
3
4
5
6
7
8
9
10
I.W.
Depth
(cm)
O.W.
0-1
1-2
2-3
3-4
4-5
5-6
6-8
8-10
14-18
22-26
PH
6.85
7.34
7.33
7.31
7.30
7.43
7.19
7.21
7.22
7.14
7.17
Carb
Alk.
Meq/1
0.83
2.96
3.61
4.07
4.43
4.52
4.87
5.26
5.39
5.89
6.02
Cl
Meq/1
Fe(II)
pg-at/1
27.5
139.5
149.3
166.2
159.9
162.8
183.7
186.4
197.7
187.6
216.0
SRP
(Jg-at/1
12.4
77.7
80.9
87.4
94.6
80.9
88.1
85.7
98.9
63.3
80.0
KH
pg-aZ/1
10
123
122
131
118
145
160
168
175
189
197
SiO Mn
(jg-at/l Mg-at/1
175
811
956
1016
1015
1067
1078
1156
1100
1123
1123
C.~
pg-at/1
Porosity
79.5
76.8
76.8
75.7
75.7
75.5
75.0
74.8
73.8
73.8
Core: HMC-C (worms present)
Days after preparation: 148
0
1
2
3
4
5
6
7
8
9
10
O.W.
0-1
1-2
2-3
3-4
4-5
5-6
6-8
8-10
14-18
22-26
7.16
7.11
7.17
7.13
7.20
7.14
7.16
7.22
7.18
7.13
7.13
1.46
2.00
2.43
2.82
3.14
3.44
3.81
4.13
4.41
5.12
5.31
13.1
22.8
15.6
13.9
14.2
40.5
53.6
73.1
89.2
88.1
60.6
4.2
6.7
4.3
3.8
8.2
22.9
29.4
42.5
53.4
52.2
49.6
77
110
141
136
147
149
162
170
185
180
194
173
417
504
564
642
742
854
961
1031
1104
1118
11
16
32
23
27
29
32
40
43
49
51
550
650
750
900
1000
1100
1100
1350
1400
1650
1700
82.0
81.4
82.3
81.0
76.8
66.7
75.2
7.53
7.53
7.41
Table 4. Flux Experiment Data (cont'd)
-------
Core: HMC-J (no worms)
Days after preparation:
Sample
Number
0
1
2
3
4
5
6
7
8
9
10
Core:
I.W.
Depth
(cm)
O.W.
0-1
1-2
2-3
3-4
4-5
5-6
6-8
8-0
14-18
22-26
PH
6.74
7.19
7.34
7.38
7.34
7.35
7.30
7.60
7.44
7.54
7.33
Carb
Alk. Cl"
Meq/1 Meq/1
0.60
2.58
2.92
3.31
3.53
3.87
4.13
4.26
4.58
5.12
5.42
Fe(II)
pg-at/1
2.2
63.0
57.8
64.9
74.1
' 77.0
74.8
75.3
90.4
104.2
108.8
SRP
pg-at/1
3.2
43.3
43.5
42.2
48.4
53.7
56.7
48.5
61.4
61.4
38.4
HMC-E (worms present)
Days after prepration:
0
1
2
3
4
5
6
7
8
9
10
O.W.
0-1
1-2
2-3
3-4
4-5
5-6
6-8
8-10
14-18
22-26
7.00
7.10
7.33
7.39
7.24
7.36
7.35
7.12
7.14
7.19
7.24
169
0.85
3.54
4.07
4.52
5.11
5.28
5.76
6.24
6.59
7.26
7.28
11.0
199.7
239.9
236.2
252.2
238.6
237.6
264.9
301.8
276.6
301.4
6.3
60.0
72.1
70.5
67.3
70.9
79.5
75.7
91.9
69.1
70.8
Mg
-a£/]
9
110
135
142
142
149
167
174
184
213
206
SiO
pg-at/l
139
711
857
961
951
995
1042
1042
1056
1064
1082
22
287
319
330
334
340
325
304
296
255
309
156
626
727
800
851
875
879
885
898
892
890
pg-at/1
0
20
26
32
33
38
39
43
45
50
54
0
28
34
43
47
47
54
57
59
64
66
Ca*2
Hg-at/1
150
700
950
900
1100
1250
1250
1300
1400
1700
1750
350
1100
1300
1450
1600
1850
1800
2000
2100
2350
2400
Porosity
78.8
77.0
76.8
74.9
75.9
77.0
76.3
75.2
74.3
73.7
78.6
76.3
76.2
75
75
74
74
73.9
73.8
73.9
Table 4. Flux Experiment Data (cont'd)
-------
Core: I1MC-T (no worms)
Days after preparation: 169
00
O
Sample
Niuuber
0
1
2
3
4
5
6
7
8
9
10
I.W.
Depth
(on)
O.W.
0-1
1-2
2-3
3-4
4-5
5-6
6-8
8-10
14-18
22-26
PH
7.36
7.16
7.25
7.26
7.25
7.24
7.18
7.28
7.28
7.29
7.28
Carb
Alk.
Meg/1
0.74
3.00
3.43
3.78
1.85
6.52
4.63
7.15
3.11
6.13
5.98
Cl" Fe(II)
Meg/1 pg-at/1
3.6
132.6
144.9
152.4
141.7
152.1
175.1
172.9
179.7
182.9
208.6
SRP
jjg-at/1
3,8
51.1
69.9
61.2
51.1
51.1
52.0
45.3
49.4
52.6
45.3
8/1
5-at/l
9
145
179
166
168
171
194
220
234
271
290
SiO
jjg-at/1
136
530
705
759
793
823
845
877
855
866
866
pg-at/1
1
25
32
33
33
41
41
46
47
51
54
+2
Ca Porosity
Mg-at/1 %
300
1000
1200
1300
1400
1500
1500
1700
1750
2000
2150
79.4
77.8
77.3
76.9
76.4
75.6
75.5
74.6
73-8
73.2
Table 4. Flux Experiment Data (cont'd)
-------
Core: LCE-8
Days after collection: 66
Sample
Number
0
1
2
3
4
5
6
7
8
9
10
I.W.
Depth
(cm) pH
O.W.
0-1
1-2
2-3
3-4
4-5
5-6
6-8
8-10
14-18
22-26
Carb.
Alk.
meq/1
2.17
2.34
2.87
3.36
3.72
4.38
4.34
4.30
4.49
3.83
3.38
Cl"
meq/1
.539
.702
.727
.677
.689
.652
.664
.689
.664
.677
.702
Fe (II)
pg-at/1
1.1
5.3
107.9
139.4
128.8
143.2
74.1
60.3
23.8
40.6
39.3
SRP
pg-at/1
1.3
7.5
50.2
77.7
86.8
93.8
59.3
39.0
12.9
14.3
22.5
NH+
Mg-at/1
2
89
121
143
160
172
154
154
174
175
320
sio2
|jg-at/l
471
684
897
975
975
858
747
607
665
742
Core: LCE-9
Days after collection: 84
0
1
2
3
4
5
6
7
8
9
10
O.W.
0-1
1-2
2-3
3-4
-4-5
5-6
6-8
8-10
14-18
22-26
2.04
2.42
3.03
3.58
4.00
4.24
4.55
4.81
4.84
4.00
3.54
.548
.536
.536
.512
.524
.500
.524
.489
.536
.500
.548
0.4
17.9
136.2
179.8
183.6
163.7
167.6
115.2
103.5
92.0
75.6
1.4
5.0
33.9
69.0
76.9
66.3
73.2
50.5
24.3
16.6
15.8
2
68
115
152
166
173
155
149
123
155
153
316
513
743
874
907
874
841
792
710
792
841
Table 4. Flux Experiment Data (cont'd)
181
-------
Core: LCE-12
Days after collection: 92
0 O.W.
1 0-1
2 1-2
3 2-3
4 3-4
5 4-5
6 5-6
1 6-8
8 8-10
9 14-18
10 22-26
Table 4. Flux Experiment Data (cont'd)
Garb.
Alk.
meq/1
2.22
3.33
3.87
4.31
4.62
4.89
5.20
5.29
4.82
3.87
3.00
Cl"
meq/1
.614
.476
.476
.501
.489
.501
.514
.501
.526
.501
.501
Fe (II)
Mg-at/1
0.1
190.8
279.0
189.3
141.3
94.8
70.7
59.6
60.9
42.2
36.1
SRP
Mg-at/1
3.0
64.9
65.7
63.0
52.1
50.9
43.2
21.5
14.1
7.9
7.9
NH*
(Jg-at/1
1
172
223
274
243
240
217
178
178
194
232
SiO
Mg-at/1
263
1288
1513
1513
1457
1288
1232
1176
1064
1232
1176
182
-------
Core: LCE-6
Days after collection: 13
0
1
2
3
4
5
6
7
8
9
10
O.W.
0-1
1-2
2-3
3-4
4-5
5-6
6-8
8-10
14-18
22-26
Carb.
Alk.
meq/1
1.96
2.27
2.73
3.07
3.33
3.44
3.22
3.53
3.51
2.67
2.53
Cl"
meq/1
.489
.575
.546
.546
.575
.561
.574
.546
.575
.561
.546
Fe (II)
pg-at/1
0.0
3.0
69.6
57.8
32.2
20.2
11.6
8.6
12.7
27.2
19.2
SRP
JJg-at/1
0.8
4.5
25.8
36.5
26.7
13.1
7.4
22.1
7.0
8.8
10.1
NH,
Mg-al/1
8
35
73
96
104
75
63
59
63
97
117
SiO
Mg-al/1
127
386
674
770
770
706
626
546
530
679
785
Core: LCE-7
Days after collection: 29
0
1
2
3
4
5
6
7
8
9
10
O.W.
0-1
1-2
2-3
3-4
4-5
5-6
6-8
8-10
14-18
22-26
7.60
7.56
7.48
7.65
7.94
7.40
7.49
7.38
7.41
7.45
7.61
2.19
2.96
3.59
3.96
4.18
4.69
4.66
4.74
4.34
3.53
2.91
.547
.547
.560
.521
.573
.508
.547
.521
.521
.534
.508
14.6
25.6
34.3
157.9
120.9
132.9
101.3
68.8
101.0
90.0
74.4
15.8
57.6
79.0
76.1
60.1
53.3
28.5
10.2
20.1
17.3
25.1
39
150
149
154
167
118
124
89
115
105
115
203
707
919
952
883
818
740
609
661
694
772
Table 4. Flux Experiment Data (cont'd)
183
-------
|