EPA-600/3-76-023
March 1976
          ENVIRONMENTAL
            PROTECTION
              AGENCY
Ecological

-------
                RESEARCH REPORTING SERIES

Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency,  have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:

     1.    Environmental Health Effects Research
     2.    Environmental Protection Technology
     3.    Ecological Research
     4.    Environmental Monitoring
     5.    Socioeconomic  Environmental Studies

This report has been assigned to the ECOLOGICAL RESEARCH series. This series
describes research  on the effects of pollution on humans, plant and  animal
species, and materials. Problems are assessed for their long- and short-term
influences. Investigations include formation, transport, and pathway studies to
determine the fate of pollutants and their effects This work provides the technical
basis for setting standards to minimize undesirable changes in living organisms
in the aquatic, terrestrial, and atmospheric environments.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

-------
                                          EPA-600/3-76-023
                                          March 1976
   GEOCHEMICAL INTERACTIONS OF HEAVY METALS

   IN SOUTHEASTERN SALT MARSH ENVIRONMENTS
                     by
              Herbert L. Windom
      Skidaway Institute of Oceanography
           Savannah, Georgia 31406
                Grant R-800372
               Project Officer

            Donald J. Baumgartner
     Ecological Effects Research Division
 Corvallis Environmental Research Laboratory
           Corvallis, Oregon 97330
   U. S. ENVIRONMENTAL PROTECTION AGENCY
     OFFICE OF RESEARCH AND DEVELOPMENT
CORVALLIS ENVIRONMENTAL RESEARCH LABORATORY
          CORVALLIS, OREGON 97330

-------
                             DISCLAIMER

This report has been reviewed by the Corvallis Environmental Research
Laboratory, 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 mention of trade names or commercial products consti-
tute endorsement or recommendation for use.
                                  ii

-------
                             ABSTRACT
This report summarizes the results of a three year study of the
transport, fate, and geochemical interactions of mercury, cad-
mium, and other inorganic pollutants in the southeastern coastal
littoral-salt marsh environment.  The general objectives of the
study were to determine:  1) the rate of input of these materials
to salt marsh estuaries, 2) the geochemical interaction they ex-
perience there and, 3) their ultimate fate in coastal littoral
waters.

The results provide a base for future evaluation of the rates of
inputs of the metals studied and their existing concentrations
in the water and sediment column of salt marsh estuaries.  The
interactions of metals with organic matter in rivers and estuaries
and their effect on transport and fate are discussed.  The effects
of processes such as flocculation, precipitation, adsorption, and
desorption from particles in estuaries are evaluated.  The distri-
bution and rate of accumulation of Hg, Cd and other metals in salt
marsh sediments are compared to their inputs to determine the
amount of these metals that ultimately reach coastal littoral
waters.  And finally, the residence time of Hg and Cd in coastal
littoral  waters is estimated from their input rates and concentra-
tions.

This report was submitted in fulfillment of Grant No. R-800372 to
the Skidaway Institute of Oceanography from the United States
Environmental  Protection Agency.
                               m

-------
                         ACKNOWLEDGMENTS
The project was financially supported by the U.S. Environmental
Protection Agency (Grant No. R-800372).  The author of this re-
port wishes to acknowledge the encouragement and assistance of
Dr. Donald Baumgartner and Dr. Milton Feldman who served as
liaison with EPA.

The many collaborators at the Georgia Institute of Technology,
School of Geophysical Sciences and at the Skidaway Institute,
whose names appear in the publications list (Section X) were
essential to the success of this project.  The assistance of
students, technicians and secretaries at the Skidaway Institute
is also greatly appreciated.  I wish to express special thanks
to Ms. Paula Vopelak for typing and preparing this report for
publication.

-------
                             CONTENTS
                                                              Page
Abstract                                                       iii
Acknowledgments                                                 iv
List of Figures                                                 vi
List of Tables                                                 vii
Sections
I           Conclusions                                          1
II          Recommendations                                      2
III         Introduction                                         4
IV          Methods                                              6
V           Data Base Storage                                   12
VI          Heavy Metal Transport by Southeastern Rivers        13
VII         Heavy Metal Geochemical Interactions in Estuaries   19
VIII        Cadmium and Mercury in Coastal Littoral Waters      31
IX          References                                          33
X           Publications                                        35

-------
                             FIGURES
No.                                                            Page
 1   Study area showing the location of major rivers
     and estuaries (in outlined areas).                         5
 2   Schematic diagram of mercury analytical  apparatus.          8
 3   Schematic non-biologic model for pollutant transfer
     through estuaries.                                        20
 4   Iron and manganese concentrations in estuarine waters
     versus salinity.                                           22
 5   Mercury and cadmium concentrations in estuarine  water
     versus salinity.                                           23
 6   Salt marsh core locations.                                26
 7   Mercury concentration in marsh sediments versus  total
     organic carbon (regression line drawn through data).       28
 8   Station location in vicinity of mercury contaminated
     study area.                                               30
                                 vi

-------
                              TABLES
No.                                                            Page
 1   Mean and range in dissolved metal  concentrations in
     southeastern rivers.                                        14
 2   Average metal  concentration in suspended sediment of
     southeastern rivers.                                        15
 3   Metal transport by southeastern rivers.                     17
 4   Average metal  concentrations in marsh cores  (dry
     weight basis).                                             27
 5   Budget for the annual  flux of metals through
     southeastern Atlantic  salt marsh estuaries.                 29
 6   Concentration of mercury in marsh  sediment.                 29
 7   Residence time of mercury and cadmium in coastal
     littoral waters.                                           32
                               vi

-------
                            SECTION I

                           CONCLUSIONS
The conclusions from this study relate specifically to southeastern
United States rivers, salt marsh estuaries and coastal areas.   It is
likely that many of the conclusions apply to other areas as well;
however, this point must be made to insure that unique characteristics
of the environments studied are not assumed to be universally  common.

Generally the conclusions of this report are of the greatest practical
value in evaluating the present and potential impact of man's  activities
on water quality in the southeastern coastal littoral-salt marsh environ-
ment.  The principal conclusions are as follows:

     1.  Southeastern rivers are similar in their concentrations of
     dissolved mercury and cadmium with mean values of 0.04 to 0.07
     and 0.3 and 1.0 yg/1, respectively.  In this regard heavily
     industrialized rivers do not differ from those that are undeveloped.
     This suggests that metal inputs are not reflected in the  water
     column but are probably evident in sediments.  Greatest effects
     are more likely to be observed in the immediate area of the input.

     2.  The high levels of dissolved and particulate  organic  matter
     in southeastern rivers can influence the fate of  metal pollutants.
     This is especially true at the river-estuary interface where floccu-
     lation of organic matter acts to scavenge metals  from the water
     column.

     3.  Industrialized estuaries differ from undeveloped estuaries
     only in the concentration of heavy metals in bottom sediments.
     The relative insolubility of heavy metals and the scavenging
     processes occurring in southeastern estuarine waters are  the major
     reasons for this.

     4.  Estuaries particularly act as sinks for iron  and manganese.
     This is probably true for other transition metals as well.

     5.  The accumulation of mercury in salt marsh sediments may be
     followed by slow release due to methylation.  This results in
     long term, localized environmental problems related to accumu-
     lation in organisms.

     6.  Mercury levels in coastal littoral water vary seasonally due
     to seasonal atmospheric inputs.  Levels vary from 0.02 to 0.3
     yg/1.

     7.  The residence time of mercury and cadmium in  coastal  littoral
     water is estimated to be 17 and 3 weeks, respectively.

-------
                           SECTION II

                         RECOMMENDATIONS
RECOMMENDATIONS CONCERNING CRITERIA FOR THE EVALUATION OF MAN'S IMPACT
ON COASTAL LITTORAL-SALT MARSH ENVIRONMENTS.

The following recommendations relate to criteria for evaluating man's
impact in coastal  areas similar to the ones studied.

     1.  Salt marsh estuaries, because of the processes occurring there,
     are very vulnerable to heavy metal pollutants.   In this  environment,
     processes lead to the accumulation of heavy metals in salt marshes
     and increase  the possibility of their entering  the estuarine food
     web.  Inputs  of heavy metals by man would have  far less  impact if
     discharged to the coastal littoral environment  rather than in
     estuarine areas.  In the latter area, greater dispersion and the
     absence of processes bringing about accumulation, decrease the
     probability of accumulation of metals in the food web.

     2.  In river and estuarine environments  monitoring of the water
     column tells  very little about the input of heavy metals.  The
     determination of heavy metal concentrations in  sediments reveals
     much greater information on the extent and impact of heavy metal
     pollution.

     3.  An estimate of the potential  impact of heavy metal  pollution
     must consider the specific metal  involved.  In  this regard it is
     necessary to  know details about the geochemical interactions of
     the given metal concerned rather than considering all heavy metals
     together as a group.

     4.  In evaluating the impact of a proposed heavy metal  input into
     an estuarine  system, an estimate of the  residence time of the metal
     in the system must be made.  For example, a slow rate of input of
     a pollutant over a long period of time will be  equivalent to a short
     term massive  input if its residence time in the system is long (i.e.,
     the pollutant concentration in the system builds up).

     5.  A sediment monitoring program in estuarine  areas is  recommended,
     especially in industrialized areas.

RECOMMENDED RESEARCH

The following research recommendations are primarily addressed to the
coastal littoral-salt marsh environment.  The recommendations, however,
have some application to other areas as well.

     1.  It is of primary importance in estuarine areas to obtain
     knowledge on  the interrelationship between the  water column and
     sediment column.  Particularly more research is needed on the
     distribution  of heavy metals in various  sediment phases  and on
     processes that affect transfer between these phases as well as

-------
the water column.  This information is particularly important
in evaluating the impact of increased heavy metals in estuarine
environments because it determines residence time of the metal
and the quantity of the metal which enters the estuarine food web.
This information is also particularly important in evaluating the
effects of sediment disturbances due to man's activities in coastal
areas.

2.  Better models are needed for pollutant cycling in estuaries so
that the environmental  impact can be evaluated.  These models must
not be of the general type commonly discussed in the literature,
but must be specific for given pollutants.

3.  A good basis for evaluating analytical data on sediments is
needed so that the physical characteristics of the sediment (e. g.3
grain size) can be taken into consideration.  This may require
development of special  chemical leaching techniques.

4.  An understanding of organism-sediment relationships regarding
heavy metals is needed.  Research related to this is recommended.

-------
                           SECTION III

                          INTRODUCTION
The following report is based on results of research on EPA Project
No. R-800372 for the period from 1 May 1972 to 30 April 1975.  The
purpose of the total project was to study the transport, fate, and
geochemical interaction of heavy metals (primarily mercury and cadmium)
being transported into the coastal littoral-salt marsh environment of
the southeastern United States coast between Georgetown, South Carolina
and Jacksonville, Florida (Figure 1).  The primary objectives of the
project are given below:

     1.  Develop a data base for the concentration of heavy metals in
     the coastal littoral-salt marsh environment of the southeastern
     United States and to elucidate their variation in time and space.

     2.  Determine the flux of metals into and through this system,
     ultimately to continental shelf waters.

     3.  Establish the significance of salt marsh sediments as a sink
     for various heavy metals.

     4.  Elucidate the form in which metals enter the estuarine environ-
     ment and changes they undergo during transport.

     5.  Elucidate processes, both chemical and physical, responsible
     for the distribution, transport and fate of heavy metals in the
     nearshore (coastal littoral) environment.

     6.  To establish the distribution of heavy metals in continental
     shelf waters and their variation with time and to elucidate the
     processes responsible for this.

     7.  Develop a model for the non-biological transport of heavy
     metals through the coastal littoral-salt marsh environment of
     the southeastern United States.

All of the above objectives have been met to some degree.  A complete
understanding of many of the problems listed in the objectives is still
lacking; however, results of this study do serve as a base.

The importance of this work lies in the fact that geochemical processes
control the form and quantity in which heavy metal pollutants enter
estuarine-nearshore food webs subsequent to their input by man.  These
processes must therefore be understood if the impact of the pollution
is to be predicted.  By studying natural transport and geochemical pro-
cesses this prediction need not be based on actual pollution case studies,

-------
The results  of  this  project have been  reported in several journal
articles and student theses.  It is not  feasible to review all  of
these studies in  this report.  It is therefore meant to provide a
summary of the  most  important results.   A complete list of publications,
however, is  given in Section X.
                        South  Carolina
                                                             33'

                                                     INYAH BAY
                                                     PEE DEE RIVER
                                                     BLACK RIVER
                                                SANTEE RIVER
                                       COOPER RIVER
                                   SAVANNAH RIVER


                              CEECHEE RIVER
                         ALTAMAHA RIVER
                         SATILLA  RIVER
                        ST. JOHNS RIVER
Figure 1.  Study  area showing the location  of major rivers and
           estuaries (in outlined areas).

-------
                            SECTION IV

                             METHODS
In this section the sampling and analytical methods used in the
study are described.  Details are only given for cadmium and mer-
cury analytical techniques to allow a critical  evaluation of re-
sults.  For other procedures, including the analysis of other
metals, references are given where details can  be found.  While
some analytical procedures were developed under our grant, most
are slight modifications of existing procedures reported in the
literature.

SAMPLING PROCEDURES

Water

The quality of any analysis of water samples for heavy metals first
depends on the adequacy of collection techniques in providing un-
contaminated samples.   Precautions necessary in sampling depend on
the sampling site, platform from which samples  are collected, and
processing prior to storage.  In general, the approach used by most
water chemists calls for the use of non-metallic samplers.  Non-
metallic sampling bottles lowered on non-metallic hydro!ines have
been successfully used in some cases.  However, it has been our ex-
perience that spurious, contaminated samples are collected from
time to time.  For the present study a non-metallic pumping system
which draws water in without allowing metal surface contact was em-
ployed.  The system is so constructed that water is sucked in from
the sampling site by means of a teflon tube.  The water enters a
teflon head to which a teflon sample container  is connected.  The
outlet to the teflon head is connected to a vacuum pump which creates
the proper suction.  Once the teflon bottle has been flushed and
filled with the water sample it is acidified with subboiling re-
distilled hydrochloric acid to a pH of 2, capped and sealed in plas-
tic bags to be returned to the laboratory.  Teflon bottles are used
for water sampling because of their chemical inertness which allows
them to be thoroughly cleaned prior to use.

Samples for mercury are collected in a similar  way; however, they
are stored in stoppered pyrex flasks and acidified with sulphuric
acid for storage.  The use of pyrex rather than teflon is to assure
that mercury is not gained or lost through sample container walls.
The acid addition here for sample preservation  also is the first
step in the analytical procedure described below.

Sediments

Because of higher concentrations of trace metals in sediments, pre-
cautions during collection are not nearly as severe as they are for
water samples.  Nonetheless it is important to  take as many precautions
as possible to insure that samples are uncontaminated during these
procedures.

-------
In the present study all samples of marsh sediment were collected
from the interiors of marsh areas, access to which was gained by
helicopter.  Samples were extracted from the sediment using a non-
metallic PVC coring device so that the sediment never comes in con-
tact with metal surfaces.  Upon returning to the laboratory cores
are extruded and sampled at intervals placing samples in teflon
beakers for digestion.

To collect bottom samples either Petersen or Eckman dredges were
used.  Prior to sampling these devices were painted with an epoxy
resin to cover all metal surfaces.  Immediately after collection
samples were placed in plastic bags and frozen until their return
to the laboratory for analysis.

ANALYTICAL PROCEDURES (WATER SAMPLES)

Mercury

With the introduction of new long light path length fTameless mer-
cury analyzers (e.g., Laboratory Data Control Mercury Monitor),
sensitivities of as low as 1 ng can be obtained so that samples as
small as 200 ml can be used.  Because of the inexpensive nature of
the equipment required and the sensitivity and ease of the analysis
the following procedure (or one similar) is favored by many marine
chemistry laboratories today.   The apparatus used included a long
path length flameless atomic absorption system having an absorption
tube with quartz end windows,  a peristaltic pump and a sample aera-
tion system such as that shown in Figure 2.  Samples were collected
directly in 300 ml Erlenmeyer flask with ground glass stoppers.
These can then be directly fitted to the closed aeration system.
The initial acid addition in Step 1 of the procedure was carried
out immediately after collection to preserve the sample.  The pre-
cision of analyses is about ±10%.

Reagents

     1.  Concentrated HaSO^.
     2.  Concentrated HNOit.
     3.  5% (W/V) potassium permanganate solution.
     4.  5% (W/V) potassium bisulfate solution.
     5.  12% (W/V) NaCl-hydroxylamine hydrogen sulfate solution.
     6.  10% (W/V) SnCl2 in 3.6 N HzSOi*.

Procedure

     1.  8 ml of cone. H2SQ^ and 4 ml cone. HN03 were added to a
     200 ml sample.
     2.  Sufficient 5% potassium permanganate solution was added
     to maintain color for 15 minutes (approximate 3 ml).
     3.  Sample was allowed to stand 30 minutes and then 2 ml  of
     12% NaCl-hydroxylamine hydrogen sulfate solution were added.
     4.  10 ml of 10% SnClz in 3.6 N HzSOi*, were added and sample
     flask was immediately fitted to aeration apparatus.

-------
     5.  Closed system was allowed to  equilibrate and absorption
     was then recorded.
                           vapor flow
                    PERISTALTIC
                      PUMP
    GROUND GLASS
      JOINT
    300ml
  ERLENMYER,
SAMPLE FLASK
                          OPTICAL
                           CELL
DRYING
 TUBE
(Magnesium
   perchlorate)
                     BUBBLER
Figure 2.  Schematic diagram of mercury  analytical apparatus.
Cadmi urn

Cadmium in natural  waters  occurs  at  levels too  low for direct
measurement by atomic absorption  spectrophotometry.  Atomic absorp-
tion analyses require enrichment  of  the  element from the original
sample in order to  reach concentrations  above its limit of detection.
Riley and Taylor (1968a) have  described  a method using a chelating
ion exchange resin  for the concentration and the subsequent deter-
mination by atomic  absorption.  The  method is also applicable to
other trace metals  (Co,  Mn,  Ni, Zn).   It involves chelating the
trace metals on Chelex 100 resin  and eluting them with a small
volume of acid.  The procedure outlined  below follows that described
by Riley and Taylor (1968b)  with  several modifications.  The eluate
from the column after being evaporated to dryness was dissolved in
HC1 rather than acetone.  It was  found that the addition of acetone
or other miscible organic  solvents produces a precipitate which clogs
the aspiration assembly  of the atomic absorption system.  Also, if
larger than 1 L samples  are used, the ion exchange column should
be backwashed occasionally to  avoid  resin compaction.  A batchwise
modification of this technique similar to that  described by Smith
(1974) alleviates many of  the  problems encountered with the column.
                                8

-------
Standards were prepared by spiking water samples which had been
stripped of trace metals by passage through a column of the chelating
resin.  A blank was prepared by processing stripped water in the
absence of the spike.  The absorption from the blank was subtracted
from the absorption of the samples and standards.  In the case of
sea water this will correct for absorption due to matrix effects
of samples having similar salinity (with 10°/oo).  Samples differing
by more than 10°/oo salinity were treated individually.

The pH of the water sample was adjusted to 7.8 +0.1 prior to analysis.
The ion exchange resin is pH selective, and the pH adjustment is
critical if 100% recovery is to be obtained.   The sample flow rate
through the column was not allowed to exceed 5 ml/minute since the
resin has a very slow exchange rate.

Replicate analyses of one large water sample indicated a precision
of +7% using this technique.  As with other procedures sensitivity
was improved by the application of a  heated graphite furnace (Perkin
Elmer HGA 2000).

Reagents

     1.  Chelex 100 ion exchange resin.  50-100 mesh.
     2.  HN03, (2N) (127 ml cone. HN03/L).
     3.  HC1 (2N) (167 ml/L).

Procedure

     1.  A suitable aliquot of the resin was  washed with excess 2N
     HNOs three times in (1 ml/ml of  resin).   Approximately 10 ml
     of resin were needed per column.
     2.  The resin was washed with double distilled HzO and packed
     on an ion exchange column (1 cm  diameter) to a depth of 9.0 cm.
     To avoid air bubbles in the resin, a few ml of double distilled
     HaO was poured into the column,  then the resin was poured in
     a thick slurry.   The water level in the  column was never allowed
     to drop below the resin bed level.  This was best accomplished
     by connecting a  piece of tygon tubing to the bottom of the
     column and looping it above the  resin bed.
     3.  The ion exchange column was  connected to a reservoir and
     the resin was washed with an additional  50 ml of H20.
     4.  The water sample, pH = 7.8,  was allowed to flow through
     the column (5 ml/min.).  The flow rate was occasionally checked
     with a stop watch and graduate cylinder.  If more than 1 L of
     sample was used  (usually 5L was  used), the column was backwashed
     every two hours  with distilled deionized water.
     5.  The column was washed with 250 ml H20, discarding the wash.
     6.  The metals were eluted with  30 ml of SN HN03 followed by
     20 ml of 2N HC1  and finally with 20 ml of double distilled water.
     The eluants were combined in a vycor flask.
     7.  The solution was evaporated  to dryness at low temperature
     and the residue  was dissolved with 1 ml  of 2N HNOs.
     8.  The solution was diluted to  5 ml.

-------
Other Metals

Procedures for the analysis of other metals reported here (e.  g.3
iron and manganese) are given in either Smith and Windom (1972)  or
Smith (1974).

ANALYTICAL PROCEDURES (SEDIMENT SAMPLES)

Metal concentrations in sediments are much higher than those in
natural waters, therefore, generally eliminating the necessity of
preconcentration prior to analysis.  However, other difficulties
are encountered with sediment samples that are not as severe in
water samples.

The greatest problem in the chemical analysis of sediments is the
lack of uniformity from sample to sample.  As a result samples vary
in matrix composition.

For the present study atomic absorption spectrophotometric techniques
emphasizing the isolation of the metal under consideration by either
separation or matrix correction were used.  When metals other than
Hg were being analyzed, a matrix correction was used on a solution
of digested sample.  The sample digestion procedure for these analyses
is given below followed by the mercury digestion procedure.

Procedure for Metals other than Mercury

Sediment samples are totally digested by the removal of silica.   This
was accomplished with hydrofluoric acid.  Upon gently heating, the
silica was volatilized as silicon tetrafluoride.  After the complete
destruction of organic matter with perchloric and nitric acids,  the
residue was brought into solution with 1:1 hydrochloric acid.

Digestion was carried out in teflon beakers having high acid and heat
resistance.  The beakers and all glassware used were cleaned in  hot
nitric acid and rinsed with redistilled water.

Reagents

     1.  Hydrofluoric acid (48%).
     2.  Nitric acid (cone.).
     3.  Perchloric acid (70%).
     4.  Hydrochloric acid (cone.).

Procedure

     1.  0.50 gm sample was weighed onto weighing paper.
     2.  The bottom of a teflon beaker was covered with redistilled
     water.  The weighed sample was dumped into the water.
     3.  15 ml of HF and 10 ml cone. HNOa were added to each sample.
     4.  Samples were covered with a teflon beaker cover and allowed
     to stand two hours.
     5.  2 ml of HClOtf were added and heated on hot plate at low
     temperature (setting of 3) until dense fumes of perchloric acid
     subsided.

                                10

-------
     6.  Samples were allowed to cool, and the sides of the beaker
     were washed with a minimum of redistilled water.
     7.  Samples were evaporated again to dryness.
     8.  The residue was dissolved with 4 ml of hot 1:1 HC1 and diluted
     to proper volume.

Once the proper dilution of the sample was made so that the metal
concentrations were in the optimum working range for atomic absorption
but prior to analysis, the matrix of the sample solution was established.
In the case of mercury, the problem with matrix interferences is cor-
rected essentially by a separation technique.

Atomic absorption spectrophotometry is relatively free of interference
in comparison to other spectral  methods; however, interferences do
exist, the most serious for heavy metals being molecular absorption.
This interference is due to blocking or absorbing of some of the light
passing through the flame.  Calcium is probably the most serious inter-
fering element; however, other specific element interference in trace
metal analysis do occur.  For cadmium these include Mg, Na, K and Fe
in addition to Ca.  Standard addition is not effective in dealing with
this type of interference.  The most effective means of dealing with
the interference is by simply determining the amount of interfering
elements present in the sample.   When a trace blank and standards are
prepared with the same concentration of interfering elements found in
the sample, the interfering signal is subtracted from the total signal.

Matching sample and standard matrices of every sample  was virtually
impossible and unnecessary.  At concentrations of the interfering ele-
ment where its ratio to the concentration of cadmium was less than
500:1, the interference was almost undetectible.lt was only important
that sample and standard matrices did not differ by more than about
500 ppm in the concentration of the interfering element for acceptable
precision.  This, therefore, allowed for grouping of samples with
similar matrices.  Final absorption was made using a heated graphite
furnace which further decreases  interference.

Procedure for Mercury

A modification of the Hatch and Ott (1968) procedure was used in the
mercury analysis of sediments.   This procedure calls for sample (approxi-
mately 0.5 gm) digestion with a 2:1 sulphuric-nitric acid solution on
a water bath at 58°C overnight.   It was found that shorter digestion
periods are insufficient for the complete destruction of organic matter.
Samples were then transferred quantitatively to BOD bottles and brought
to a volume of 200 ml  with washing from the digestion beaker.  The
procedure at this point followed that given above for mercury in water
samples.

OTHER ANALYTICAL TECHNIQUES USED IN THIS STUDY

For details of specific analytical techniques related to the characteriza-
tion of organic matter and to laboratory studies the reader is directed
to the references cited in subsequent sections.  Because of the nature
of these techniques it is unfeasible to report them here.

                                11

-------
                            SECTION V

                         DATA BASE STORAGE

Many of the results discussed below are in a form useful to other
researchers and agencies as a data base for future reference.  For
this reason those data that are useful in this regard and are amena-
ble to storage are being deposited in the EPA STORET System.  These
are primarily water quality data that are useful  independent of
interpretation.
                                  12

-------
                            SECTION VI

           HEAVY METAL TRANSPORT BY SOUTHEASTERN RIVERS
To evaluate the potential affect of increased heavy metals on the
coastal littoral-salt marsh environment of the southeastern United
States it is first necessary to evaluate the inputs of these metals
to the system due to natural processes.  Since the largest natural
input of materials to salt marsh estuaries is due to river runoff,
it is obvious that the quantity of metal input to this system due
to this process should be ascertained.  Some of the rivers in the
study area are heavily industrialized.  Studies of the form and
concentrations of the metals in the river studied should identify
characteristics of man's input.  Finally the role of organic matter
in the river transport of metals will enable a better understanding
of the impact of this input source.

Prior to the initiation of the present study relatively little
information was available on metal loads of southeastern rivers.
Turekian and Scott (1967) evaluated the concentration of several
heavy metals, not including mercury and cadmium, in suspended
material in several southeastern streams.  Windom et aZ.(1971)
investigated the magnitude of the stream supply of particulate and
soluble zinc, copper, cobalt, nickel and chromium by the Altamaha,
Ogeechee and Satilla Rivers.  The results of this study, however,
were based on one set of data so that seasonal variations could not
be taken into account.   Other than these two studies, little additional
research has been carried out in this area.

To develop a better understanding of the present rate of supply of
heavy metals, especially mercury and cadmium, the major rivers emptying
into this area (Figure 1) were sampled biomonthy.  These rivers repre-
sent about 95% of the total runoff between Cape Remain, South Carolina
and Jacksonville, Florida.   The concentration of iron, manganese, cad-
mium and mercury were determined on each sample.  In addition, other
studies were conducted to elucidate the form in which the metals are
transported and especially the role of organic matter in this transport.

The results of these various studies are summarized below and reference
is made to the pertinent published papers and reports which give the
details of the studies.

CONCENTRATIONS OF HEAVY METALS IN SOUTHEASTERN RIVERS AND THEIR RATE OF
TRANSPORT.

Although relatively large variations in the concentrations of the metals
studied are observed for each river (Table 1) the mean concentration of
cadmium and mercury differ little between rivers.  Iron and manganese,
however show relatively high variability.  This may reflect variations
in the composition of the drainage basins for each river as does the
mineralogical composition of suspended matter (Neiheisel and Weaver, 1967;
Windom, et al.3 1971).
                                 13

-------
     Like  the  dissolved metal  loads the concentration of mercury and cadmium
     in  participate matter also varies little  from river to river with larger
     variations  in iron and manganese (Table 2).  Again this probably reflects
     drainage  basin composition.

                                  Table  1

            MEAN AND  RANGE  IN DISSOLVED METAL  CONCENTRATIONS  IN
                            SOUTHEASTERN  RIVERS1
ppb
River
Pee Dee


Black


Santee


Cooper


Savannah


Ogeechee


Altamaha


Satin a


St. Johns



Mean
Max
Min
Mean
Max
Min
Mean
Max
Min
Mean
Max
Min
Mean
Max
Min
Mean
Max
Min
Mean
Max
Min
Mean
Max
Min
Mean
Max
Min
Fe
210
450
90
240
410
60
170
360
30
80
240
10
170
380
70
310
490
140
150
280
30
370
610
280
70
90
40
Mn
21
40
12
18
53
8
12
53
3
3
8
1
21
34
6
33
63
15
17
39
2
42
85
22
4
8
2
Cd
1.0
1.4
0.4
1.0
3.6
0.1
0.4
0.7
0.1
0.7
2.6
0.1
0.3
0.6
0.2
1.0
2.0
0.4
0.8
2.6
0.2
0.8
1.6
0.3
0.8
1.5
0.1
Hg
0.06
0.10
0.02
0.06
0.14
0.02
0.05
0.08
0.02
0.04
0.08
0.02
0.07
0.13
0.02
0.07
0.14
0.02
0.05
0.06
0.02
0.07
0.14
0.04
0.05
0.08
0.03
!Based on results of a total  of from 6  to 30 samples  collected from each
 river bimonthly, May 1972 to July 1973.

                                     14

-------
                                  Table  2

              AVERAGE  METAL  CONCENTRATION  IN  SUSPENDED  SEDIMENT
                          OF  SOUTHEASTERN RIVERS1
   River
Av. Suspended
Sediment Load
  (mg/1)
Fe
                                                  _PPm_
Mn
Cd
Pee Dee
Black
Santee
Cooper
Savannah
Ogeechee
Altamaha
Satilla
St. Johns
9
11
53
7
22
7
16
26
12
6.6
5.0
5.7
5.7
8.2
5.0
5.2
6.2
4.4
1050
1490
1400
830
1120
1300
1600
330
500
5
7
3
15
8
13
26
10
20
0.5
0.6
0.2
1.0
0.7
0.6
0.7
0.4
0.6
^Based on results  of a  total  of from 6  to  30  samples  collected  from
 each river bimonthly,  May 1972 to  July 1973.
     Three  of  the  rivers  studied  (Savannah, Cooper and St. Johns) are
     highly industrialized and are expected to receive large concentra-
     tions  of  heavy metal pollutants.  The lack of significantly in-
     creased levels of the metals studied in these rivers, however,
     indicates that the increased industrialization is not reflected
     in water  column concentrations.  Because of the short residence
     time of heavy metals in natural waters, it is more likely that they
     accumulate in bottom sediments shortly after their input from in-
     dustrial  effluents.  Subsequent transport in the river is prooably
     then by bedload traction.

     Present rates of input of iron, manganese, cadmium and mercury to
     the southeastern coastal littoral-salt marsh environment by rivers
     can be estimated from the concentration of metals in both dissolved
     and particulate forms and the average annual discharge of each river.
                                      15

-------
These results (Table 3) indicate that most of the iron that is
carried by rivers is in participate form, while a relatively
small percentage of the cadmium and mercury is particulate.
Manganese is equally distributed between the two phases.   A
detailed analysis of particulate matter in the Savannah and
Ogeechee Rivers for trace metal partitioning between phases
(•i.e., absorbed, reduced, oxidized and residual) indicates that
most of the iron is in refractory residual phases.   Manganese
occurs in reduced phases with significant amounts adsorbed and
in residual phases.  Cadmium shows the highest concentration in
residual and reduced phases.

The details of the research described in this section are given
in Windom (1975).  Results of studies on metal partitioning in
particulate matter are given in Nance (1974).

HEAVY METAL-ORGANIC MATTER INTERACTIONS IN SOUTHEASTERN RIVERS

As a part of this project studies of the composition of south-
eastern river water organic matter was determined and its inter-
action with heavy metals was investigated.  The results of these
studies are given in Beck et al. y (1974), Dunn (1974) and Martin
(1973).

The chemical characteristics of southeastern rivers, especially
those whose drainage basin is entirely in the coastal plain, differ
considerably from those of most rivers of the world (Beck et al.,
1974).  The difference is reflected primarily in the dominance of
organic matter over inorganic constituents and the resultant low
pH.  Based on chemical and physical characteristics, southeastern
river water organic matter consists of humic substances,  the bulk
of which resembles fulvic acid.  In areas of high rainfall, and
low relief such as Coastal Plain River Basins, large quantities of
humic substances are produced from the degradation of abundant
vegetative litter (Beck et al. 3 1974).  Of the humic substances
produced, only the more soluble, lower molecular weight material
is carried into the river.  Limited by its solubility in water,
river water organic matter is of relatively low polydispersity when
compared to other humic substances (Martin, 1973).   As river water
organic matter is transported downstream, the higher molecular
weight portion (low in total acidity) resembling humic acid is re-
moved, probably by flocculation.   The dissolved organic matter
transported to the estuary would be characterized by low molecular
weight and high exchange capacity.

The importance of humic materials as complexing agents in soils has
been studied by numerous experimenters (see Schnitzer and Kahn, 1972,
for a review).  Beck et al. 3 (1974) considered the possible role of
river water organic material in mobilizing complexed metals in
natural waters.  These organic compounds can effectively increase
the solubility of metals and retard precipitation (Rashid et al. 3
1973), providing a mechanism for heavy metal transport and enrich-
ment.  As a part of this research program an examination was made


                                  16

-------























oo
o:
UJ
C£
z
UJ
I —
oo
UJ
P*—
— >
O
00 <"
ni >•
£ CQ
-° 1—
•= i
D-
oo
z:
o:
i
i —
i
•_j
^-
UJ






































s-
o
Q_
to
(D ^— N

Q.


O>
31


O


Q-




s- cn
h- l^ -0
ro O
r— O
tO r-H
1 % ^^-
CU
s:
r—
n3
C
<






Q


d.
C
s:

Q






^-> o.
CD
U3 CO
O U_
T-H
•s 	 .x

o

cu
re
u
01 vi ^^
(O Q (N
S- r-l
CU r- O
> (O i— 1
fit 3 ^
c
c
"*•
















OOOOi— IO<— IOO
• ••••••••
ooooooooo



^^ CJ^ CO CM Lf) CTl OO 00 ^O
VO f~ ^ f*^ i^j (,^ f™> LO t~H CM
• ••••••••
OOOOOOOOO

• ••••••••
f"~» c^ CO ^™~i 1"~i CO ^" CO *"""!







^i-crir~.uDCMco<— ILOOO
t-H CO CO 00 CO t~H CT^ *""! CO
I— t


r- li— )^-r-»LOr-(O'— (CO
i-H i— 1 CM CO


O ^" O •"I O VO LO CO O
^^ r-H CM ^" CO VO CTt 00 CM
CM CM r-l





t— 1



LO CM ^i~ CO CTt *«O OO OO CO
• ••••••••
CMOO-— 1 «-« O •— IOO




r-~.°oair^«*ocnoo
f-H CO T""! CM CO CM r— * CM LO
t— 1 t— 1 r— 1 i— 1











(/I
f CU fO C
CU CO CU -C (O _C
CO  CX n3 CO to T-
O)fOCO>CU4J4->
CO i — to O fO O^ i — fO 4-*
D-CQOOOOOOsiOOOO
T— 1
LO
•
o
o
T-H
CO
CTl
o

co
o

r— 1
T— 1

o
• f-H
CM CM
LO

o
1— 1


LO
IQ r— 1
CO LO
r- 1
en
o
cr>





CM
LO
CM
LO
CM 00
vO
LO
•
cr>




^
oo
LO

CO
(0
3
0
1o -£>
-!-> S-
O to
h- 0-
•O -M
c c
p— re co
re s- o
4-> CU S-
O CO
h- Q-




17

-------
of the mechanism and degree of transport of purified river water
organic-cadmium complexes to enable a more complete assessment of
the possible impact of the mobilization of toxic metals in marine
environments (Dunn, 1974).

A sample of dissolved river water organic acid (largely fulvic acid)
was extracted from the Satilla River and purified.  Cd-organic acid
complexing studies were performed using potentiometric titrations,
with and without a background of Cd2+.  Acid dissociation values were
obtained from the acid-base titration data using the assumption of
the similarity of the system to a polyprotic acid, and Cd-organic
stability constants were obtained from titrations with a background
of Cd in which the free cadmium ion was followed with a Cd electrode.
Results indicate that the nonchelating complexing phenomenon can play
an important part at low pH and can explain the change in stability
constants with pH noted by other experimenters.  At pH 6.3 Cd-organic
complexes form in a 1:1 ratio of chelated to non-chelated complexes.
At lower pH levels non-chelated complexes become increasingly import-
ant, a result which is of direct interest to those concerned about
the environmental impact of the mobilization of toxic metals.

The results of these studies have broad implications regarding en-
vironmental problems such as the toxic metal pollution of rivers and
estuaries.   Heavy metal pollutants have the ability to form stable,
water soluble organo-metallic complexes with humic substances
(Schnitzer, 1971).  Precipitation of heavy metals as insoluble salts
is prevented by the presence of soluble humic substances (Rashid and
Leonard, 1973).

In the Satilla River and other rivers, where dissolved organic matter
resembling humic substances is abundant, its presence may play an
important role in the transportation of heavy metals.  Organic matter
and metal-organic complexes are introduced to the estuary as either
dissolved or colloidal material.   In laboratory studies, the addition
of an electrolyte such as NaCl to humic colloids causes coagulation
(Ong and Bisque, 1968).  Thus it is expected that when soluble or-
ganic matter and associated metals transported by river water encoun-
ter increased salinity in the estuary, flocculation occurs resulting
in rapid deposition and accumulation of metal in salt marsh sediments.
                                  18

-------
                          SECTION  VII

        HEAVY METAL  GEOCHEMICAL  INTERACTIONS  IN ESTUARIES

 One  of  the  prime  objectives of  this study was to evaluate geochemical
 interactions involving heavy metals entering estuaries in river runoff.
 This  information  is of primary  importance in evaluating the environ-
 mental  impact of  heavy metal inputs to the coastal littoral-salt marsh
 system.   By understanding these processes it is possible to predict
 whether a given metal will accumulate in salt marsh estuaries or be
 transported directly to the coastal littoral environment.  This
 obviously determines the point  of  impact.

 The  geochemical processes affecting heavy metal fate in estuaries are
 schematically shown in Figure 3.   In this model the river is divided
 into  three  compartments or phases  which contain metals.  As river
 water is  transported into estuaries metals are potentially transferred
 from one  compartment of the river  system to various compartments or
 phases  of the estuarine water column as indicated (processes 1 through
 9).   Once the metals are in the estuarine system in one of the various
 compartments, their fate there  depends on the transfer to other com-
 partments where modifications may  occur to produce further transfer.
 Ultimately  the metals must be transferred out of the estuarine system
 by various  pathways, the importance of which depends upon the coeffi-
 cient of  transfer.  In the present model only the non-biologic part
 of the  system is  considered; therefore, only two pathways of loss
 can  be  identified.  These are (1)  loss to the marsh sediment due
 to sediment accumulation which  is  assigned the transfer coefficient
 Kis  and (2) loss  in solution due to offshore transport (transfer
 coefficient Kie).   It is obvious that the offshore transport of metals
 could be  in the tissues of migrating organisms or rafted marsh grass
 detritus.   For the  present simplified model which considers only geo-
 chemical  processes however, the form in which the metal is transported
 out  of  the  system is not important.

 The model in Figure 3 emphasizes the geochemical processes that are
 of interest.  Generally they can be divided into two categories.  The
first includes those occurring at the salt water-fresh water boundary.
The second  includes processes occurring in the estuary itself.   Within
each of these categories  specific important processes  or process
related phenomena were studied.   These are listed immediately below
with specific references  to  a more complete discussion and are  sum-
marized in the following  paragraphs of this section.

Geochemical  processes  occurring  at the river-estuary  boundary.

     1)   Metal  concentration variations  in solution versus salinity
     (Windom,  1975).
     2)   Adsorption-desorption reactions  (Nance, 1975).
     3)   Variations in metal  partitioning  in particulate matter in
     rivers  versus estuaries (Nance,  1974).
     4)   Flocculation  at  the river-estuary boundary  (Arnone,  1974).

                                19

-------
Rivers
Estuaries
Offshore
                                            Marsh
                                          Sediments
                       Loss to Marsh Sediments
      Figure 3.  Schematic non-biologic model  for pollutant transfer through
               estuaries.
                                  20

-------
Geochemical processes occurring in estuaries.

     1)  Characteristics of heavy metal accumulation in salt marsh
     sediments (Windom, 1974).
     2)  Remobilization of mercury from sediments due to methylation
     (Windom  et al., 1975).

METAL CONCENTRATION VARIATIONS IN SOLUTION VERSUS SALINITY

Concentrations of heavy metals in rivers are generally much higher
than those found in marine waters (Turekian, 1971).  Even coastal
waters of the southeastern United States show levels of metals higher
than those of the open ocean indicating an increasing concentration
gradient from offshore to inshore (Windom and Smith, 1972).  If
metals transported in solution by rivers undergo no reaction that
lead to their precipitation in the estuarine zone (i.  e.3 conserva-
tive mixing), variations in their concentrations would be a linear
function of salinity, showing a decrease due to dilution with metal
poor marine waters.  If, on the other hand, the metal  concentrations
change with salinity in a nonlinear way, the influence of processes
other than dilution is indicated.

The variations in metal concentration with salinity in eight south-
eastern estuaries (Figure 1) were determined and exhibited expected
characteristics.   For example, iron decreases exponentially with
increasing salinity (Figure 4) owing to its precipitation upon entry
into the estuarine zone.  Manganese shows a similar but less pro-
nounced trend.  These decreases in iron and manganese at the river-
estuary boundary may be due to the formation of hydrated iron oxide
which flocculates, precipitates and accumulates in estuarine sedi-
ments.  The increased electrolyte concentration of estuarine waters
can also influence the fate of very fine (<0.45 microns) particulate
iron and manganese associated with organic matter, also leading to
their flocculation.  Variations in Cd and Hg with salinity (Figure
5) appear to follow what is expected for more conservative mixing,
in that significant changes in concentrations from rivers to estuaries
are not observed.

These results imply that estuaries are sinks for dissolved riverborne
iron and manganese but not cadmium and mercury.  The solubility of
the forms of these latter two metals in rivers is apparently not
greatly different than that in estuarine waters.  This is clearly
not the case for iron and manganese.  Since iron precipitation is
an effective means of scavenging other heavy metals, especially
those in the transition series, its accumulation in estuaries may
be accompanied by the accumulation of other metals.  As is the
case in fresh water, cadmium and mercury inputs by man would not
be reflected in the estuarine water either.
                                21

-------
                            600
                            400
                     PPB
                     PPB
                            200
                                                   Iron
                                •  «i  •
                               •*••?•••.:.
                                I .Vl  •     * •
                              0    5/0/3  20   25  30  35
                            /50 r
                            (00
                            50
                                              Manganese
             •
pi   •  *
 • * •




 :•;?•;:?'!.  **';.
                                                     »

                                                     ,.
                                                     Hi
                              0   5/0/5   20  25   30   35
                                         Salinity (%°)



Figure 4.   Iron and manganese concentrations  in estuarine  waters versus

            salinity.
                                     22

-------
                 30
                 20
           PPB
                 10
                                                        Cadmium
                        • •     f  •

                        • T • • *  T**  • t  *   t  " *
                                                       _L
                 03
                 02
          PPB
                 Ol
                               10     15     20     25     30    35

                                   Salinity (%0)


                                    .                    Mercury
                     •*•••
                               10
                                     15     20     25     30
                                                             35
                                   Salinity  (%0)

Figure 5.   Mercury and  cadmium concentrations in estuarine water versus
            salinity.
                                    23

-------
ADSORPTION-DESORPTION REACTIONS

As suspended matter moves across a salinity gradient the increased
ion strength of estuarine waters is probably the major factor in
adsorption-desorption reactions.  Common ions such as Na+, K+, Ca++
and Mg++ compete for adsorption sites on suspended sediment, hence
decreasing its adsorption capacity for trace metals.  This adsorption-
desorption mechanism probably functions for Zn, Cd, and Mn in south-
eastern river estuaries to varying degrees.  It is also apparent
that seasonal changes in the character of the suspended sediment
(primarily organic matter content) affects its adsorption capacity
for Cd, Mn, Zn and Hg.  These changes are especially important in
the uptake and release of Hg.

Radioisotopes (65Zn, 5t|Mn, 109Cd, and 203Hg) were used to trace the
effect of salinity on metal  exchange with suspended sediment in south-
eastern estuaries.  Results  indicate that a substantial release of
Cd, Zn and Mn occurs at salinities between 0°/oo and 5°/oo.  No direct
correlation was observed between salinity and Hg exchange.

VARIATIONS IN METAL PARTITIONING IN PARTICIPATE MATTER

Suspended sediment collected from the estuarine zones of two major
southeast Georgia rivers (Savannah and Ogeechee) were analyzed using
a selective chemical leaching technique to determine the role of
suspended material on metal  transport in southeastern estuarine environ-
ments.  The metal concentrations (Cu, Cd, Pb, Zn, Mn, and Fe) in the
leached fractions, (adsorbed, reduced, oxidized and residual) of
suspended sediments from fresh water and saline environments were
compared.  Results indicate  that suspended sediment can account for
a significant portion of these metals in natural waters.  Generally,
the residual fraction (metals in lattice sites of crystalline detri-
tal material) and the reduced fraction (metals precipitated and copre-
cipitated as metallic coatings) are the phases containing the major
portion of the trace metals  in suspended sediment.  The effect of
salinity changes on the metal concentrations of the leached fractions
appear to be a function of the specific metal, season, and area of
study.  Comparison with the  results of similar studies indicate the
importance of regional differences in the character of suspended
matter on the transport of trace metals.

FLOCCULATION AT THE RIVER-ESTUARY BOUNDARY

At the initial appearance of salt water (3-50 ppm Na) organic-rich
floccules occur which are not observed upstream in coastal plain
rivers such as the Satilla.   The organic carbon content of the floccules
decreases with increasing salinity (>50 ppm Na).  Dissolved organic
carbon (DOC) decreases proportional to the mixing with sea water.
Therefore, the organic floccules do not appear to originate immediately
by precipitation of the DOC, and thus the composition of the floccules
is not quite understood.  The suspended particulate organic carbon
(POC) appears to be incorporated with the resuspended bank material in
the mixing zone and eventually becomes included in the marshes.

                                 24

-------
Dissolved metals show a rapid decrease seaward and are assumed to be
incorporated into the floccules.  However, the increase of metals in
suspended particulate matter observed in the mixing zone is not attri-
buted to this incorporation only.  An increase in the concentration
of the suspended material in the mixing zone due to resuspension of
bank material is related to the increase of metals in the particulate
matter and masks the precipitational feature of the dissolved metals.

CHARACTERISTICS OF HEAVY METAL ACCUMULATION IN SALT MARSH SEDIMENTS

The Fe, Mn, Cu, Cd, Hg, Pb and Zn concentrations were determined
at 10 cm intervals on cores taken from 25 stations in salt marshes
(Figure 6).  Throughout this area the average concentration of the
metals (Table 4) appears to be relatively constant.  The small vari-
ations that are present are probably due to variations in organic
matter as appears to be the case for mercury (Figure 7).  No differ-
ences due to geographical area were indicated in these results.

The overall mean concentrations of heavy metals in salt marsh sediments
can be taken to represent what metal levels unpolluted sediments in
the coastal littoral-salt marsh environment should be.  The concen-
trations of heavy metals in the sediments of three major harbors in
the study area (Charleston, South Carolina, Savannah and Brunswick,
Georgia) are generally higher than these levels.  For example, the
average concentrations of mercury, cadmium, lead and zinc in Charleston
Harbor sediments are over twice that of the average salt marsh.  This
is the worst case observed in the area studied.

By using an average sediment accumulation rate of 1 mm/yr for salt
marsh sediments as explained in Windom (1975), the rate of accumula-
tion of heavy metals can be estimated.  These are given in Table 4.
Applying these values to the total 400 hectare salt marsh system a
budget can be established for Fe, Mn, Cd and Hg since their river
inputs are known (Table 5).  From this budget it can be seen that
salt marshes act as a sink for Fe and Mn while only the particulate
Cd and Hg accumulates there.

REMOBILIZATION OF MERCURY FROM MARSH SEDIMENTS DUE TO METHYLATION

During the course of this study, an area highly contaminated by mercury
inputs from a chloral kali plant was investigated.   The source of the
mercury has been previously identified by EPA as the Allied Chemical
Company near Brunswick, Georgia (Figure 8).  Although the mercury input
occurred over several years, its impact apparently has only been on
a local  area judging from the Hg concentration in sediments (Table 6).
The discharge was discontinued in 1972; however, high levels of mercury
are still present.

The initial input of the inorganic mercury would be followed by
accumulation in adjacent sediments, therefore affecting only a localized
area.   One of the important processes for subsequent remobilization of
the mercury is methylation in the sediment column.  Using data on the
accumulation of methylmercury in lower trophic level  organisms from
this area, Windom et al.  (1975) estimated the annual  production rate

                                 25

-------
                     South  Carolina
Figure 6.   Salt marsh core locations.
                              26

-------
                                Table 4

              AVERAGE METAL CONCENTRATIONS  IN MARSH CORES

                           (Dry Weight Basis)
Station
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
% ppm
Fe
3.9

3.9
5.0


1.2
1.8
4.1
1.9
2.4
4.8
4.6
3.4
4.0
2.9
2.7
3.2
1.1
3.7
3.1
4.7
2.9
0.9
1.5
Mn
103
250
166
142
209
158
139
114
309
174
151
260
353
319
173
366
262
273
175
168
244
132
128
205
74
Cu
30
16
26
25
24
10
4
4
16
3
8
12
20
10
10
8
6
12
4
8
8
7
10
5
2
Hg
0.12
0.09
0.16
0.05
0.08
0.01
0.04
0.02
0.04
0.04
0.06
0.09
0.16
0.10
0.11
0.02
0.05
0.16
0.03
0.06
0.07
0.06
0.08
0.05
0.05
Cd
3.4
5.0
2.9
3.3
4.4
1.0
0.4
0.8
2.1
0.1
0.5
0.6
0.8
0.4
018
0.8
0.2
0.8
1.6
0.8
1.2
0.8
0.7
0.8
0.4
Pb
25
26
20
20
27
16
9
9
13
4
8
16
21
20
19
15
20
24
11
15
12
13
20
15
9
Zn
63
67
67
72
70
36
51
21
51
11
67
70
81
57
64
48
42
61
17
43
37
39
45
21
15
Mean
3.1
202
11
0.07
1.4
17
49
Average
Accumulation
Rate*
(mg/m2-yr)
        4.5x10"
          303
            16
         10
         2.1
           25
          71
*Assuming an average sedimentation rate of 1 mm/yr and an average specific
 gravity of marsh sediments of 1.5 g/cc.

 Analytical techniques similar to those for suspended sediment.
                                    27

-------
 sx
X
                                                                           7.0
       Figure 7.  Mercury concentration in marsh sediments versus total organic
                  carbon (regression line drawn through data).
                                        28

-------
                  Table 5



BUDGET FOR THE ANNUAL FLUX OF METALS THROUGH



 SOUTHEASTERN ATLANTIC SALT MARSH ESTUARIES

Total Input
(% of total input in
particulate phases)
Sedimentation Loss
% Sedimentation Loss
Net Flux through Estuary
Fe
(106Kg)
62
(85)
210
100
0
Mn

1866
(51)
1200
64
666
Cd
(103 Kg)

52
(21)
9
17
43
Hg

3.6
(14)
0.4
11
3.2
                  Table 6



          CONCENTRATION OF MERCURY



              IN MARSH SEDIMENT
Station
Number
1
2
3
4
5
6
7
8
9
10
Total
0 - 5 cm
1.34
0.34
—
1.70
0.48
0.27
0.55
0.41
0.28
0.27
Hg (ppm dry wt.)
5 - 10 cm
0.15
0.08
0.06
0.48
0.33
0.08
0.82
0.30
0.25
0.23
                      29

-------
3fio'
  05'
                               A ALLIED CHEMIC
                                                             St. Simons Sound
                                                                     nautical miles
                                                                    1  1/2  0
                                                          1
                                                  yards

                                             1OOO 0   ~ 2000
           Figure  8.
            81° 30'

Station location in vicinity  of mercury contaminated study
area.
                                                                                   25'
           of this  form  to be approximately 50 yg/g of total mercury in the upper
           5 cm of  the marsh sediment column.  From these results it would appear
           that the loss of mercury from the system might be a slow process.   In
           fact,  the Department of Natural Resources for the State of Georgia has
           reported high levels of mercury in a number of game animals from this
           areas  over the last several years.  Results of the present study suggest
           that this may persist for some time to come.
                                           30

-------
                          SECTION VIII

         CADMIUM AND MERCURY IN COASTAL LITTORAL WATERS

The coastal littoral environment extends offshore to less than 20 km.
The average water depth is about 5 m and the length of the system
between Cape Romain, South Carolina and Jacksonville, Florida is
approximately 300 km.  This environment receives inputs of metals
from the adjacent southeastern United States and under steady state
conditions the concentration of a given metal will depend on its
rate and mode of input and its residence time in the environment.
This latter parameter is primarily influenced by water mass mixing
and exchange.  To predict the impact of increased metal inputs on
coastal littoral waters, information on the above parameters is
critical.  This is especially true if the input occurs directly to
the coastal littoral environment.  The input due to rivers would,
to a large degree, be controlled by processes occurring in estuaries
which would tend to act as buffers.

The following paragraphs summarize data obtained on the rate of
input, concentration and residence time of cadmium and mercury in
the southeastern coastal littoral environment.

MERCURY CONCENTRATION IN, AND INPUT TO, COASTAL LITTORAL WATERS

Windom et al. (1975) found that the mercury concentration in
continental shelf waters, including those of the coastal littoral
environment, vary seasonally, ranging from a minimum of 5 ng/1 to
a maximum of 300 ng/1.  These variations cannot be explained by
river runoff.  Studies of the concentration of mercury in the atmo-
sphere, however, suggest that atmospheric transport may provide a
major input.  This conclusion is based specifically on the observa-
tion that mercury in the atmosphere is highest during periods of
offshore winds and this is correlated to high mercury concentrations
in the water column.  It is estimated that the annual input of
mercury to these waters due to atmospheric transport is greater than
.3 mg/m2.  Added to the net flux through estuaries due to runoff,
the total mercury input to the coastal littoral environment is
approximately 5.4 metric tons per year.

Taken over the entire year the average concentration of mercury in
coastal littoral waters is approximately 60 ng/1.  Judging from the
relatively rapid fluctuation of mercury in these waters, physical
processes of mixing must occur within a relatively short time period.

CADMIUM CONCENTRATION IN, AND INPUT TO, COASTAL LITTORAL WATERS

The average concentration of cadmium in coastal littoral waters has
been reported to be approximately 0.1 yg/1 (Windom and Smith, 1972).
This concentration is somewhat higher than those found offshore and
indicates an effect due to coastal  runoff.
                                 31

-------
The annual input of cadmium to estuaries due to river runoff was
discussed in Section VI.  Only a portion of this, however, ultimately
reaches the coastal littoral environment as is indicated in Table 5.
Using the results shown in this table the annual input of cadmium is
estimated at 52 metric tons.

RESIDENCE TIME OF MERCURY AND CADMIUM IN COASTAL LITTORAL WATERS

The residence time of a metal in coastal littoral waters can be
expressed by the following equation:
Where C is the total content of the metal in coastal littoral waters
and dc/dt represents its rate of input.  If it is assumed that the
rate of mixing of continental shelf waters is relatively constant,
then the concentration of a given heavy metal in the coastal littoral
waters would be a function of input rate (steady state).  Therefore,
if the residence time of the metal in these waters is known then
the concentration of the metal can be predicted for increased rates
of input.

The approximate volume for coastal littoral waters is 3x1010 cubic
meters.  Using this volume and the concentration of mercury in Windom
et al.  (1975), the total content of the these two metals in coastal
littoral waters can be determined (Table 7).  Using the annual inputs
of these two metals, their residence times can be estimated (Table 7).
The residence times calculated for the two metals appears to be similar
to the mixing time expected for the area.

                            Table 7

             RESIDENCE TIME OF MERCURY AND CADMIUM

                   IN COASTAL LITTORAL WATERS


                 Content of              Annual          Residence
           Coastal Littoral Waters        Input            Time
               (in metric tons)       (metric ton/yr)     (weeks)


Mercury             1.8                    54               17


Cadmium             3.0                    52                3
                                 32

-------
                            SECTION IX
                            REFERENCES

Beck, K.C., Reuter, J.H. and Perdue, E.M. (1974) Organic and inorganic
     geochemistry of some coastal plain rivers of the southeastern
     United States.  Geochim. Cosmochim, Acta, 38, 341-364.

Dunn, T.H. (1974) Electrode studies of stability constants of Cd-
     river water organic matter complexes.  M.S. Thesis, The Georgia
     Institute of Technology, Atlanta, 70 pp.

Hatch, W.R. and Welland, L.O. (1968) Determination of sub-microgram
     quantities of mercury by atomic absorption spectrophotometry.
     Analytical Chemistry 40(14): 2085-2087.

Nance, S.W. (1974) The role of suspended matter on trace metal trans-
     port in an estuarine environment.  M.S. Thesis, The Georgia
     Institute of Technology, Atlanta, 36 pp.

Neiheisel, J.  and Weaver, C. (1967) Transport and deposition of clay
     minerals, Southeastern United States, J. Sed. Pet., 3_7, 1084-
     1116.

Ong, H.L. and Bisque, R.E. (1968) Coagulation of humic colloids by
     metal ions.  Soil Sci., 106. 220-224.

Rashid, M.A.  and Leonard, J.D.  (1973) Modifications in the solubility
     and precipitation behavior of various metals as a result of their
     interaction with sedimentary humic acid.  Chem. Geol., 11, 89-97.

Riley, J.P. and Taylor, D. (1968a) The determination of manganese in
     sea water.  Deep-Sea Res., _15_, 629-632.

Riley, J.P. and Taylor, D. (1968b) Chelating resins for the concentra-
     tion of trace elements from sea water and their analytical use
     in conjunction with atomic absorption spectrophotometry.  Anal.
     Chim. Acta, 40, 479-485.

Schnitzer, M.  and Kahn, S.U.  (1972) Humic substances in the environ-
     ment, Marcel Dekker, Inc., New York, 327 pp.

Schnitzer, M.  (1971) Metal-organic matter interactions in soils and
     waters.   Jji Organic Compounds in Aquatic Environments (editors
     S.J. Faust and O.V.  Hunter), Marcel  Dekker, Inc., New York,
     297-315.

Smith, R.G. and Windom, H.L.  (1972) Analytical  Handbook for the
     Determination of As, Cd, Co, Cu, Fe, Pb, Mn, Hg,  Ni, Ag, and
     Zn in the Marine and Estuarine Environments, Ga.  Mar.  Sci. Cen.
     Tech. Rpt. No.  72-6, 62 pp.
                                 33

-------
Smith, R.G. (1974) Improved ion-exchange techniques for the concen-
     tration of Mn from sea water.   Anal. Chem., 46, 607-608.

Turekian, K.K. and Scott, M.R.  (1967) Concentrations for Cr, Ag, Mo,
     Ni, Co, and Mn in suspended material in streams.  Environ.  Sci.
     Tech. I, 940-942.

Turekian, K.K. (1971) Rivers, tributaries and estuaries.  In
     Impingement of Man on the Oceans.   (Editor, D.W. Hoody Wiley
     Interscience, New York, 9-74.

Windom, H.L., Beck, K.C. and Smith, R.  (1971) Transport of trace
     metals to the Atlantic Ocean by three southeastern rivers.
     Southeast Geol.  _12, 169-181.

Windom, H.L. and Smith, R.G. (1972) Distribution of cadmium, cobalt,
     nickel and zinc in southeastern United States Continental Shelf
     waters.  Deep-Sea Res., 19_, 727-730.

Windom, H.L. (1975) Heavy metal fluxes  through salt marsh estuaries.
     In  Proc. 2nd Int. Estuarine Conf., Academic Press, in press
     Tsee author for complete reference).

Windom, H.L., Gardner, W., Stephens, O.A., and Taylor, F. (1975) The
     role of methylmercury production in the transfer of mercury in
     a salt marsh ecosystem.  Estuarine and Coastal Mar. Sci.  (sub-
     mitted, see author for complete reference).

Windom, H.L., Taylor, F.E., and Waiters, E.M. (1975) Possible in-
     fluence of atmospheric transport on the total mercury content
     of southeastern Atlantic Continental Shelf surface waters.
     Deep-Sea Res. (in press, contact author for complete reference).
                                 34

-------
                             SECTION X
                           PUBLICATIONS

 The following is a list of publications which cite the support of
 EPA Grant No. R800372.

 1973     Bloomer, D.R.,  A hydrographic investigation of Winyah Bay,
         South Carolina and the adjacent coastal waters, M.S. Thesis,
         Georgia Institute of Technology, Atlanta, 57 pp.

 1973     Martin, S.J.,  Chemical characteristics of dissolved organic
         matter in river water, M.S. Thesis, Georgia Institute of
         Technology, Atlanta, 45 pp.

 1974     Arnone, R., Floccule characterization of the Satilla Estuary,
         M.S. Thesis, Georgia Institute of Technology, 56 pp.

 1974     Beck, K.C., Renter, J.H. and Perdue, E.M.,  Organic and in-
         organic geochemistry of some coastal plain rivers of the
         Southeastern United States, Geochim. Cosmochim, Acta, 38.
         341-364.

 1974     Dunn, T.H.,  Electrode studies of stability constants of
         Cd-river water organic matter complexes, M.S. Thesis,
         Georgia Institute of Technology, Atlanta, 70 pp.

 1974     Nance, S.W.,  The role of suspended matter on trace metal
         transport in an estuarine environment, M.S. Thesis, Georgia
         Institute of Technology, Atlanta, 36 pp.

 1974     Smith, R.G.,  Improved ion-exchange technique for the concen-
         tration of manganese from sea water, Anal. Chem., 46. 607-608.

 1975     Nance, S.W.,  The role of suspended matter in transporting
         trace metals through the southeastern Atlantic estuarine
         environment, Jj^ Proc. Mineral Cycling Symp., ERDA, Augusta,
         Ga., In press.

 1975     Windom, H.L.,  Heavy metal fluxes through salt marsh estuaries.
         Jjl Proc. 2nd Int. Estuarine Conf., Academic Press, In press.

 1975     Windom, H.L., Taylor, F.E. and Waiters, E.M,,  Possible in-
         fluence of atmospheric transport on the total mercury content
         of  southeastern Atlantic Continental Shelf surface waters,
         Deep-Sea Res., In press.

 1975     Windom, H.L., Dunstan, W.M. and Gardner, W.S.,  River input
         of  inorganic phosphorus and nitrogen to the Southeastern salt
         marsh estuarine environment, Jjn Proc. Mineral Cycling.

1975      Windom,  H.L...  Gardner,  W.S., Stephens,  J.A.  and Taylor,  F.,
         The role of methylmercury production in the transfer of  mercury
         in a salt marsh ecosystem, Estuarine and Coastal  Mar.  Res.,
         Submitted.
                                  35

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
  REPORT NO.
    , EPA-600/3-76-023
                             2.
                                                           3. RECIPIENT'S ACCESSION-NO.
4. TITLE
GEOCHEMICAL INTERACTIONS  OF HEAVY METALS IN SOUTH-
EASTERN SALT MARSH  ENVIRONMENTS
                                                           5. REPORT DATE
                                                            March 1976
                                                           6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)

 Herbert L. Windom
                                                           8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS

 Skidaway  Institute of Oceanography
 P.O. Box  13687
 Savannah, Georgia  31406
                                                           10. PROGRAM ELEMENT NO.
                                                          11. CONTRACT/GRANT NO.

                                                          EPA R-800372
12. SPONSORING AGENCY NAME AND ADDRESS
 United States  Environmental Protection Agency
 Office of Research and Development
 Corvallis Environmental  Research Laboratory
 Con/all is,  Oregon   97330
                                                          13. TYPE OF REPORT AND PERIOD COVERED
                                                          Final. 5/72 - 4/75
                                                          14. SPONSORING AGENCY CODE

                                                          EPA-ORD
15. SUPPLEMENTARY NOTES
16. ABSTRACT
      This  report summarizes the results of  a  three year study of the  transport, fate,
 and geochemical  interactions of mercury,  cadmium and other inorganic  pollutants in
 the southeastern coastal littoral-salt marsh  environment.  The general  objectives of
 the study  were  to determine: 1) the rate  of input of these materials  to salt marsh
 estuaries,  2) the geochemical interaction they experience there and,  3) their ulti-
 mate fate  in  coastal  littoral waters.
      The results provide a base for future  evaluation of the rates  of inputs of the
 metals  studied  and their existing concentrations in the water and sediment column of
 salt marsh estuaries.   The interactions of  metals with organic matter in rivers and
 estuaries  and their effect on transport and fate are discussed.  The  effects of pro-
 cesses  such as  flocculation, precipitation, adsorption, and desorption  from particles
 in estuaries  are evaluated.  The distribution and rate of accumulation  of Hg, Cd and
 other metals  in salt marsh sediments are  compared to their inputs to  determine the
 amount  of  these metals that ultimately reach  coastal littoral waters.   And finally,
 the residence time of Hg and Cd in coastal  littoral waters is estimated from their
 input rates and concentrations.
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                            b.lDENTIFIERS/OPEN ENDED TERMS  C.  COS AT I Field/Group
 Estuaries,  flocculating, geochemistry,
 metal  complexes, metals, rivers, saltwater
 sediments,  water, water analysis, water
 pollution.
                                              Heavy metal geochemistry
                                              Heavy metal transport
                                              Estuarine geochemistry
13. DISTRIBUTION STATEMENT

 Release  to public
                                             19. SECURITY CLASS (ThisReport)
                                              Unclassified
                                                                         21. NO. OF PAGES
                                              20. SECURITY CLASS (Thispage}

                                               Unclassified
.13.
                                                                         22. PRICE
EPA Form 2220-1 (9-73)
                                            36
                            AU S GOVERNMENT PRINTING OFFICE 1976-696-884171 REGION 10

-------