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

-------