vvEPA
            United States
            Environmental Protection
            Agency
            Region V
             Great Lakes National
             Program Office
             536 South Clark Street, Room 932
             Chicago, IL 60605
EPA-905/9-79-005-B
March, 1979
Maumee River
Pilot Watershed Study

Sediment,
Phosphates, and
Heavy Metal Transport
Defiance Area, Ohio

-------
The United States Environmental Protection Agency was created because of
increasing public and governmental concern about the dangers of pollution
to the health and welfare of the American people. Noxious air, foul water,
and spoiled land are tragic testimony to the deterioration of our natural
environment.
The Great Lakes National Program Office (GLNPO) of the U.S. EPA, was
established in Region V, Chicago to provide a specific focus on the water
quality concerns of the Great Lakes. GLNPO provides  funding and personnel
support to the International Joint Commission activities under the US-
Canada Great Lakes Water Quality Agreement.
Several land use water quality studies have been funded to support the
Pollution  from Land Use Activities Reference Group (PLUARG) under the
Agreement to  address specific objectives related to land use pollution to the
Great Lakes. This report describes some of the work supported by this Office
to carry out PLUARG study objectives.
We hope that the information and data contained herein will help planners
and managers of pollution control agencies make better decisions for
carrying forward their pollution control responsibilities.
                                   Dr. Edith J. Tebo
                                   Director
                                   Great Lakes National Program Office

-------
                                               EPA-905/9-79-005-B
                                               March 1979
          THE MAUMEE RIVER BASIN PILOT WATERSHED STUDY
                            Volume II
         Sediment,  Phosphates, and Heavy Metal Transport
                               by

                         Terry J.  Logan
                     Principal Investigator
                       Agronomy Department
           Ohio  State University,  Columbus, Ohio 43210
       Ohio Agricultural Research and Development Center
                       Wooster, Ohio 44691

                               for

             U. S. Environmental  Protection Agency
                       Chicago,  Illinois
                      EPA Grant No.  R005145

                         Project Officer
                       Eugene Pinkstaff
              Great  Lakes National  Program Office
This study, funded by a Great Lakes Program  grant  from the U.S.  EPA,
was conducted as part of the Task C-Pilot Watershed  Program for  the
International Joint Commission's Reference Group on  Pollution from
Land Use Activities.
              GREAT LAKES NATIONAL PROGRAM OFFICE
        U.S. ENVIRONMENTAL PROTECTION AGENCY, REGION V
               536 SOUTH CLARK STREET, ROOM 932
                    CHICAGO, ILLINOIS 60605
                          U.S. Fr.vimircntai Protection Agency
                          GLr-T-O Library Collection (PL-12J)
                          77 West Jackson Boulevard.
                          Chicago, IL 60604-3590

-------
                          DISCLAIMER
     This report has been reviewed by the Great Lakes National  Program
Office, Region V, 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  constitute
endorsement or recommendation for use.

-------
                            ACKNOWLEDGMENTS
     Work on this project was funded by a grant from the U. S. Environmental
Protection Agency, Region V, Chicago, Gene Pinkstaff, Project Officer, and
Ralph Christensen, Grants Officer.

     This study was the combined effort of many individuals at Ohio State
University.  They include Dr. Larry Wilding, Dr. Neil Smeck, Dr. Wayne Pettijohn,
Dr. Earl Whitlach, and Dr. Glenn Schwab.  Graduate students whose thesis work
contributed to the study are Fred Rhoton, Dennis McCallister, Dan Green, and
Tom Naymik.  Special study by Michael Thompson is also included in this report.

     The technical support of Rodney Smith and Ted Pohlman was critical to
the success of the project, and special thanks are due Dr. Robert Stiefel,
Director, Water Resources Center, for his continued interest and support of
the project.
                                  ii

-------
                           TABLE OP CONTENTS
  DISCLAIMER                                                              i

 ACKNOWLEDGMENTS                                                        ii

  LIST OF TABLES                                                          v

  LIST OF FIGURES                                                      viii

 ABSTKACT                                                                1

 1.     SUMMARY                                                           2

 2.     INTRODUCTION                                                      3

 3.     RESULTS                                                           g

 3.1    Phosphate  adsorption-desorption  characteristics of soils and       6
       bottom sediments  in  the Maumee River  Basin of Ohio
 3.11  Abstract                                                           g
 3.12   Introduction                                                      7
 3.13  Methods and  materials                                              7
 3.14  Results and  discussion                                             9
 3.15  Conclusions                                                      24
 3.16  Literature cited                                                 26
 3-2   Phosphate adsorption-desorption  characteristics of suspended     29
      sediments in the Maumee River Basin of Ohio
 3.21  Abstract                                                         29
 3.22  Introduction                                                    30
 3.23  Methods and  materials                                            31
 3.24  Results and  discussion                                           34
 3.25  Conclusions                                                      41
 3.26  Literature cited                                                 42
 3-3   Clay mineralogical, physical, and chemical relationships          44
      between watershed soils and runoff sediments
 3.31  Abstract                                                         44
 3.32  Introduction                                                     45
 3.33  Methods and materials                                            4g
 3.34  Results and discussion                                           47
 3.35  Conclusions                                                      50
 3.36  Literature cited                                                 54
 3-4   Coagulation and dispersion of Maumee River Basin soils           56
      and particle-size distribution of soils and sediments
 3.41  Abstract                                                         5g
 3.42  Introduction                                                     57
 3.43  Methods                                                          57
3.44  Results                                                          60
3.45  Conclusions                                                      gg
3.46  Literature cited                                                 gg
                                 ill

-------
3.5   Clay-equilibration studies in natural and simulated bottom       71
      sediment environments
3.51  Abstract                                                         71
3.52  Introduction                                                     72
3.53  Methods and materials                                            73
3.54  Results and discussion                                           74
3.55  Conclusions                                                      88
3.56  Literature cited                                                 88
3.6   Occurrence and  stability of calcite in the Maumee River          90
3.61  Abstract                                                         90
3.62  Introduction                                                     91
3.63  Materials and methods                                            92
3.64  Results and discussion                                           93
3.65  Conclusions                                                     106
3.66  Literature cited                                                106
3.7   Heavy metals in Maumee River Basin water, soil ,and sediment     108
3.71  Abstract                                                        108
3.72  Introduction                                                    109
3.73  Methods                                                         109
3.74  Results                                                         112
3.75  Conclusions                                                     125
3.8   Pesticides in watershed soils and Maumee River Basin            128
      bottom sediments
3.81  Abstrac.t                                                        128
3.82  Methods and materials                                           129
3.83  Results                                                         13°
3.84  Literature cited                                                132
                                   IV

-------
                         LIST OF TABLES

Table Number                                                             Page

2.     INTRODUCTION

1      Summary of Watershed Sites and Plots                                5

3.     RESULTS

3.1    Phosphate Adsorption-Desorption Characteristics of Soils and
       Bottom Sediments in the Maumee River Basin of Ohio

1      Some Physical and Chemical Properties of the Soils, Their           10
       Clay Fractions, and Bottom Sediments

2      Extractable Iron, Aluminum, and Silicon in the Soils,  Their         15
       Clay Fractions, and Bottom Sediments

3      Phosphate Sorption Properties of the Soils, Their Clay              17
       Fractions, and Bottom Sediments

4      Correlation Coefficients of P Sorption Parameters Versus            19
       Physical and Chemical Properties of the Soils

5      Correlation Coefficients of P Sorption Parameters Versus            20
       Physical and Chemical Properties of Clay Fractions of
       Surface Soils

6      Correlation Coefficients of P Sorption Parameters Versus            21
       Physical and Chemical Properties of Bottom Sediments (n - 5)

7      Phosphorus and Clay Enrichment of Runoff Sediment                   25

302    Phosphate Adsorption-Desorption Characteristics of
       Suspended Sediments in~the Maumee River Basin of Ohio

1      Stream Samples Composited According to Storm Flow and                33
       Calcite Content of Suspended Sediment

2      Total P, Adsorption Maximum, Adsorption Energy, Equilibrium         36
       P Concentration (EPC), and Total P Desorbed for Stream
       Composite Suspended Sediment Samples

3      Correlation Coefficients Among Phosphate Parameters and             40
       Calcite Content from the Stream-Composited Sediment Samples
       (n - 13)

-------
                        LIST OF TABLES  (continued)

Table Number                                                           Page

3.3    Clay Mineralogical, Physical and Chemical Relationships
       Between Watershed Soils and Runoff Sediments

1      Clay Mineralogy of the Total Clay Fraction from Surface            48
       Soils and Corresponding Runoff Sediments

2      Correlation Coefficients of Runoff Sediment  Mineralogy as          51
       a Function of Sediment Concentration

3      Percent Total Clay, Fine Clay, and Cation Exchange Capacity       52
       of Surface Soils and Runoff Sediments

4      Amorphous Iron and Free Iron Oxide Content (mg/g total clay)       53
       of Surface Soils and Runoff Sediments

3.4    Coagulation and Dispersion of Maumee River Basin Soils and
       Particle-Size Distribution of Soils and Sediments

1      Physical, Chemical, and Mineralogical Properties of Surface       61
       Horizons from Six Representative Soils from the Maumee Basin

2      Chemical Data for Natural Media                                    62

3      Particle Size Analysis and Bulk  Density of Maumee River            66
       Basin Soils

4      Particle Size Analysis of Runoff Sediment (Sonification)           67

3.5    Clay Equilibration Studies in Natural and Simulated Bottom
       Sediment Environments

1      Mineralogical Composition of In  Situ Equilibrated Clays            75

2      Elemented and Oxide Analysis of  In Situ Equilibrated Clays        76

3      External Surface Area, Cation Exchange Capacity and Carbonate     78
       Analysis of In Situ Equilibrated Clays

4      Measurements of (001) Illite Peak Broadening in In Situ  Clays     79
       Equilibrated for 217 and 281 Days
       Amorphous Iron Content of Laboratory Equilibrated Clays
80
                                  vi

-------
                          LIST OF TABLES (continued)

 Table Number                                                             Page

 3.6    Occurrence and Stability of Calcite in the Maumee River

 1      Temperature,  pH,  Calcium, Bicarbonate, and Electrical               94
        Conductivity  Values of the Maumee River Drainage System
        (September 1975 to September 1976)

 2      Concentrations of Calcium, Bicarbonate, Carbon Dioxide, and         96
        Several Ion Pairs (Calculated and Measured) for Selected
        Representative Sampling Sites in the Maumee River Basin

 3      Calcium Carbonate Equivalent and Calcite/Dolomite Ratios of         102
        Composite Suspended Sediment Samples

 4      Algal Populations from Stream Samples  Collected on September         105
        8,  1976

 3.7    Heavy Metals  in Maumee River Basin Water,  Soil,  and Sediment

 1      The Locations  of  Sample Sites Used in  Studying  Heavy Metal           111
        Background Concentrations

 2      Coefficients  of Variation (CV) of  Precision for  Elemental            115
       Analysis  by AA  Spectroscopy

 3       Background  Concentration  of  Heavy Metals in the Maumee River         116
        Basin and in Groundwater  (1975-77)

 4       Concentrations  of Heavy Metals in Maumee River Basin Soils,         117
        Bottom Sediments, and  Limestone Bedrock

 5      Elemental Analysis of  16 Grid Samples from Each Site (Upstream      118
       and Downstream Water and Bottom Sediments)

 6      Sample Number Required to Maintain + 10$ Precision of the Mean      124
       Concentration of Each Element      ~

 7      Heavy Metal Concentrations in Living and Dead  River Vegetation      126
       Along  the Ottawa River (ug/g)

 8      Heavy Metals Detected  in High Concentrations in Stream Hater        .127
       and Sediments  Near Cities

3.8     Pesticides in  Watershed Soils and Maunee River Basin  Bottom
       Sediments                           ~'

1      Pesticide Residues Found in Soil  and Sedisent  3aaipl.es                ".?(;

-------
                         LIST  OF FIGURES

Figure Number                                                            Page

2.     INTRODUCTION

1      Sampling Sites in the Mauraee Basin                                  4

3.     RESULTS

3.1    Phosphate Adsorption-Desorption Characteristics  of Soils and
       Bottom Sediments in the Maunee River Basin of Ohio

1      Lanpmuir Adsorption Isotherm for Phosphate on Hoytvill* Clay        13
       Loam

2      Adsorption Isotherm for Phosphate  on Hoytville Clay Loam            14

3.2    Phosphate Adsorption-Desorption Characteristics  of Suspended
       Sediments in the~Maumee River Basin of Ohio

1      Suspended Sediment Sampling Sites  in the  Haumee  River  Basin         31

2      Langmuir Adsorption Isotherm of Suspended Sediment for              35
       Phosphate

3      Adsorption/Desorption Characteristics of  Haumee  River  Basin         37
       Soils, Their Clay Fractions, and Suspended and Bottom  Sediments

303    Clay Mineralogical, Physical and Chemical Relationships
       Between Watershed Soils and Runoff Sediments

1      Ratio of Runoff Sediment to Soil-Clay Mineralogy for                49
       Individual Watersheds

3.4    Coagulation and Dispersion of Maumee River Basin Soils and
       Particle-Size Distribution of Soils and Sediments

1      Relative Dispersability of Selected Soil  Surface Horizons           60a

3.5    Clay Equilibration Studies in Natural and Simulated Bottom
       Sediment Environments

1      Concentration of Hater Soluble Silica as  a Function of Time         82
       and Atmosphere at 4° C

2      Concentration of Water Soluble Silica as  a Function of Time,        83
       Temperature, and Atmosphere

3      Concentration of Water Soluble Aluminum as a Function  of Time       84
       and Atmosphere at 4° C
                                  Vlll

-------
                         LIST OF FIGURES (continued)

Figure Number                                                            Page

3.5     Clay Equilibration Studies in Natural and Simulated Bottom
        Sediment Environments

4       Concentration of Water Soluble Aluminum as a Function of           85
        Time, Temperature, and Atmosphere

5       Concentration of Water Soluble Iron as a Function of Time,         86
        Temperature, and Atmosphere

6       Concentration of Water Soluble Manganese as a Function of          87
        Time, Temperature, and Atmosphere

3.6     Occurrence and Stability of Calcite in the Maumee River

1       X-Ray Diffractograms of Stream Suspended Sediments                 99

2       Calcium Levels in the Stream and Occurrence of Secondary           100
        Calcite

3       Bicarbonate Levels in the Stream and Occurrence of Secondary       104
        Calcite

3.7     Heavy Metals in Maumee River Basin Water, Soil, and Sediment

1       Sampling Sites Used in Determining Background Concentrations of    110
        Heavy Metals

2       Sampling Grid Pattern and Channel Cross Section of the Site of     113
        the Variability Upstream from Lima, Ohio

3       Sampling Grid Pattern and Channel Cross Section of the Site of     114
        the Variability Downstream from Lima, Ohio

4       The Variability in Strontium Concentrations in the Water at the    120
        Nearbank and the Main Flow Sampling Points Upstream from Lima,
        Ohio

5       The Variability in Chromium Concentration in the Water at the      121
        Nearbank and the Main Flow Sampling Points Downstream from
        Lima, Ohio

6       The Variability in Chromium Concentration in the Bottom            122
        Sediment at the Sampling Location Downstream from Lima, Ohio

7       Limit of Accuracy (95JJ) Curves as Percentage of the Mean for       123
        Elements in the Bottom Sediments Upstream from Point Source
                                 IX

-------

-------
                                   -1-
                                ABSTRACT


      Losses of nutrients and  sediment from agricultural  land were  monitored
 during 1975-1977 in the Maumee River  Basin of Ohio.   These  results have  been
 reported in Volume 1 of the report,   Watershed Characteristics and Pollutant
 Loadings.   A number of special studies  on the mineralogy, chemistry., arid  trans-
 port of sediment,  as well as  pesticide  and metal  sediment transport, were
 conducted as part  of the project.   They are reported  here as separate  sections.

      Maumee River  Basin suspended  sediments were  found to be higher in total-P
 and labile-P than  soils or stream-bottom sediments.   Sediments are enriched
 in P during erosion and transport  as  a  consequence of preferential transport
 of clay,which is higher in total-P than the whole soil.  Some P  enrichment of
 suspended sediment was due to concentration by algae  in  the stream.  Photo-
 synthetic consumption of C(>2  by algae was responsible for formation of secondary
 calcite in the stream.

      Changes in sediment mineralogy from its  original soil mineralogy  was a
 result of preferential clay transport.   There was no  evidence of significant
 mineral alteration during sediment transport.

      The Maumee River environment  was found to be conducive to coagulation of
 fine and coarse clay soil and sediment  into larger aggregates.  Aggregated
 clay was found  in  runoff sediment  as well  as  in stream-bottom sediments.

      Heavy metals  in Maumee River  Basin water and sediment were determined.
 Levels  were low and  indicated that groundwater and eroded soil were  the major
 source.  Although  a  number  of point sources were  identified throughout the
 Basin,  they had little  effect on loadings  to  Lake Erie,  A point-source chromium
 discharge  was absorbed  by stream sediment  and  taken up by rooted macrophytes.

      Pesticide  scan  of Maumee River Basin  soils and bottom sediments showed
 only traces  of  DDT and  other  persistent chlorinated hydrocarbons.  No other
 known pesticides in  the  scan were  detected  in  significant quantities.

     The watershed  characteristics  and pollutant  loadings are discussed in
Volume 1 of this report.

-------
                                  -2-


1.  SUMMARY
     Special studies conducted in conjunction with watershed monitoring
focused on the characteristics of stream sediment:  their susceptibility to
fluvial transport, changes in the characteristics of the original soil material
during its passage through the stream system, and the chemical reactivity of
sediment in the adsorption/desorption of nutrients and other chemicals.

     These studies showed that the Maumee River Basin sediments  are fine-
textured, chemically reactive, and high in nutrients.  This is due to the high
clay content of the native soils and their relatively youthful nature.

     Total-P content of Basin soils are high (^700 ug P/g sediment) and even
higher in sediments (> 1000 ug/g).  This is a result of the higher clay content
of sediment and the enrichment of phosphorus in the clay fraction of mineral
soils.  Basin soils and sediments have a large capacity to adsorb P, especially
stream-bottom sediments, while suspended sediments were high in  labile or P
that could be desorbed into solution.  Basin sediments, therefore,  have the
capacity to reduce high point-source concentrations of soluble P entering the
stream and yet can release large amounts of P to algae or other  aquatic
vegetation.

     Runoff and stream sediments were shown to be flocculated, i.e. clay-sized,
particles that had coagulated into silt and sand-sized floccules.   High Ca    status
of Maumee Basin streams is responsible for coagulation of negatively charged
clay particles.

     Decomposition of soil minerals or formation of secondary minerals during
sediment transport was studied.   Mineralogy of stream sediment was  found to
be somewhat different from that of the original soil.   Sediments were higher
in illite and expandable minerals and lower in vermiculite than  soils, but
this difference was attributed to preferential clay transport rather than to
mineral alteration.  Secondary calcite was formed in suspended sediment during
summer months.  High algal activity resulted in C02 consumption  and shift of
the carbonate equilibrium toward calcite formation.  In the anaerobic bottom  -
sediment environment, there appeared to be some changes in the iron oxide
mineralogy.

     Heavy metals were measured in Basin soils, sediments, and waters.  Metal
levels were at background levels in all cases, and groundwater appears to be
a major source of metal loadings for the Basin.  Point sources of metals were
scattered throughout the Basin, and, although they contribute little to the
total Basin metal load ,their effect on near-downstream water quality can be
severe.  A chromium discharge into the Ottawa River at Lima, Ohio increased
metal content of bottom sediments and rooted macrophytes.

     Pesticide scan of Basin agricultural soils and stream-bottom sediments
revealed only traces of persistent chlorinated hydrocarbons such as DDT
and dieldrin.

-------
                                   -3-
2.  INTRODUCTION
     The special studies reported here in Volume 2  of the Maumee River  Basin
Pilot Watershed Study final report were conducted separately and constituted
in most cases M.S.  or Ph.D. thesis research.   They  are accordingly treated
here as separate chapters,  each representing  a complete report  of that  work.

     For a more complete description of the Maumee  River Basin,  its  physical
characteristics, land use,and pollutant loadings,  the reader should  consult
Volume 1 of this report.  Most of the special studies were conducted on the
small agricultural  watersheds monitored during this study as well as other
sites in the Basin.  These are identified in  Figure 1 and Table 1.   More
specific site data  are provided in the individual reports.

-------
         Figure 1. Sampling sites in the Maumee Basin
   The Maumee Basin

^ Water Samples
ifc Watersheds
 1 — Hammermith Roselms
 2	Crites Roselms
 3 - — Lenewee
 4 — Blout
 5 —• Paulding
 6 — Hoytville Plots
 3jt~  Continuous mass
     transport stations
 O  Continuous rain
     gaging stations
" Defiance County
4*     2*
 Black
 Creek
 Study Area

-------
Table 1,   Summary of vatershed  sites and plots
CODE
111
201
301 &
302
501 &
502
1*01 &

611 to
682
DOMINANT
SOIL
Roselms
Roselms
Lenavee
Paulding
Blount

Hoytville
PHYSIOGRAPHIC
REGION
A
Lake
Plain
V
Till
Plain

Lake
Plain
GEOLOGIC SLOPE
MATERIALS (%}
DEFIANCE COUNTY
^ 3-15
3-5
Lake
Clays
< 1
v 1
Clay Loam 3-h
Till

WOOD COUNTY
Clay < 1
Till
HECTARES
3.2
0.6
0.8
0.1
1.0
0.9

O.OU
DRAINAGE
SYSTEM REMARKS
Surface Complex Slopes
Surface — X — -^
121 ^jc
Surface & Hi
Tile
Surface &
Tile
Surface & Dissected Uplands
Tile ,
t-n
I
Surface & OARDC Drainapc
Tile Plots

-------
                                   -6-
3.  RESULTS
3.1  Phosphate adsorption-desorption characteristics of soils and bottom
    sediments in the Maumee River Basin *>f Ohio
3.11  Abstract

     Langmuir adsorption isotherms showed that Maumee River Basin sediments
had adsorption capacities 10 to 20 times greater than Basin soils.   Although
the soil-clay fractions had adsorption capacities higher than the whole soil,
they were considerably less than those of the sediments, and the difference is
attributed to the higher content of amorphous or low-range order Fe and Al
components in the bottom sediments.  Equilibrium phosphorus concentration (EPC)
and P desorbed was similar for soils and sediments as well as total P,indi-
cating that although the bottom sediments have a high capacity to adsorb P, this
capacity has not been realized.  Correlations between adsorption-desorption
parameters and soil/sediment properties are presented.  Bray PI "available" P
was highly correlated with EPC and P desorbed in the soils but to a lesser
extent in the bottom sediments.  CDB and oxalate extractable-Fe was  highly
correlated with P adsorption capacity in the bottom sediments but not in the
soils.

-------
                                    -7-


 3.12  Introduc tion

      In recent years, increased attention has been focused on the problem of
 additions of phosphates to natural waters, particularly as evidenced in ac-
 celerated eutrophication of lakes.  Soluble phosphorus, as opposed to nitrogen,
 is often singled out as the factor limiting the growth of excess biomass in
 a body of water,due to the relatively small inputs of phosphorus compared to
 nitrogen, as well as the difficulty in controlling the many forms of N (1,11).
 Although urban and industrial wastes are major sources of phosphates to the
 Great Lakes, runoff from urban and agricultural land is assuming major impor-
 tance as other sources diminish or are removed entirely (1,2,3).   Non-point
 source losses of P in eroded material from the soil surface are diffuse and
 widespread  and as such are difficult to quantify and control.   Similarly,
 tile drainage and groundwater, which are often overlooked in typical phos-
 phorus movement studies,  can add significant quantities of P to streams and
 lakes (3,13).

      The dissolved-particulate phosphorus  balance of a stream or  lake is
 controlled largely by the adsorption-desorption reactions with  sediments carried
 by the water.  The mechanism in aquatic  systems appears to be based  on specific
 anion adsorption  such as  described by Kingston et al   (9,10),  rather than
 mechanisms proposed solely for soils  where localized P concentrations may
 be  quite high and pH's  low.   Adsorption  is most strongly associated  with
 amorphous  or,at most,short-range ordered  secondary hydroxy iron coatings  (22,26,27)
 Soluble phosphate may be  adsorbed and transported or desorbed into solution,
 depending  on ambient pH,  solution phosphorus  status,  presence of  competing
 anions,  and  redox potential  (5,14,15,25).

      The first objective  of  this  study was  to  determine the  phosphate sorption
 properties of a group of  typical  agricultural  soils  and compare them with
 stream-bottom sediments presumed  to be eroded  from similar soils.  Secondly,
 the  sorption properties were  related  to selected  chemical  and physical
 properties of the soils and  sediments  to determine  the principal  factors
 associated with sorption  by  the  two types  of adsorbents.


 3.13  Methods  and materials

      Samples  of five Ap horizons  from  soils of agricultural watersheds in
 Defiance County, Ohio, and one additional soil  from  the Northwest Branch, Ohio
 Agricultural  Research and Development Center (OARDC), were collected  for a
 study of phosphorus  sorption.  They represent some of  the major agricultural
 soils of the Maumee River Basin of northwestern Ohio and have been developed
 from glacial  lacustrine clays or clayey till.  The soils chosen for study were
Roselms silty clay  (Aerie Ochraqualf, very fine, illitic, mesic);  Broughton
 silty clay (Aerie Hapludalf, very fine, illitic, mesic); Lenawee silty clay
 loam  (Mollic Haplaquept, fine illitic, nonacid, mesic); Paulding silty clay
 (Typic Haplaquept, very fine, illitic, nonacid, mesic); Blount loam (Aerie
Ochraqualf, fine,  illitic, mesic); and Hoytville clay loam (Mollic Ochraqualf,
 fine illitic, mesic).  The soil samples were air-dried, ground,  and the less
than 2mm portion retained.

-------
     Five stream-bottom sediment samples  (0-10cm depth)  were also  collected
for comparison of their phosphate sorption behavior  to that of the surface
soils.  The sediments were stored as a slurry in tightly sealed polyethylene
containers at 4°C.  Portions of the sediments were wet-seived and  the less
than 2mm fraction retained, also as a slurry at 4°C.   The bottom sediments  are
primarily of surface soil origin,since stream-bank erosion in the  Maumee River
Basin is 00% of the total sediment load.

     The soil samples and bottom sediments, plus the total clay « 2 u)  frac-
tions of the soils, were analyzed for a series of chemical and physical  proper-
ties.  Particle'size distribution was determined by pipet with sodium hexa-
metaphosphate dispersion; pH with a 10:1 suspension:  soil ratio in 0.01M CaC^",
organic carbon by incineration^; inorganic (carbonate) carbon by gas evolution
(4); total P by perchloric acid digestion (23); inorganic P by NaOH/HCl  extrac-
tion (17); organic P by difference between the total and inorganic fractions;
and "available" P by Bray PI extraction (2).  The results of these analyses
are shown in Table 1.  The soils, sediments,and clays were also characterized
for various iron, aluminum,and silica fractions, including "amorphous" Fe and
Al  (21); "free oxides" of Fe and Al (16); and amorphous silica (8).  These
data are given in Table 2.

     The phosphate-sorption characteristics determined  for the soils, sediments,
and clays (obtained  by sedimentation after ultrasonic dispersion)  included
adsorption maximum and adsorption energy parameter, using  the Langmuir
isotherm  (18), equilibrium phosphorus concentration (EPC)  (25), and sequen-
tially desorbed phosphorus.

     Five-gram samples of  soil  or bottom sediment, or 2.5  grams of clay, the
latter two as a slurry, were equilibrated with a  graded series of phosphate
solutions  (as KI^PO^)  in polyethylene  tubes  for 24 hr on a rotary  shaker.
Five drops of toluene  were added  to retard microbial growth,  and  the tubes
were made  to volume  such  that  their  final  concentration was  0.01M with  respect
to CaCl2-  At the end  of 24 hr,  the  tubes were centrifuged until clear  and the
supernatant  filtered through a  0.2um filter membrane.  An  aliquot  of  the filtrate
was analyzed  for  soluble  inorganic  P by  the method of John (12).   P  adsorbed
or desorbed  per  unit mass  of adsorbent was  calculated and  plotted  in  the single-
phase Langmuir  isotherm  form as c/(x/m)  vs  c, where c is  final  solution con-
centration and x/m is  the  quantity of  P  adsorbed  per  unit  mass  of  adsorbent,
according to the general  equation:
                                   c/(x/m)  = c/b  + 1/kb             (17)
The constants k and  b can be  found from the slope and  intercept of the  resulting
curve where b is the P adsorption maximum  and  k is  a parameter related to
 the energy of adsorption.
 1  Post, G. J. 1958.  A study of three methods for determination of organic
    carbon in Ohio Soils of several great groups and the profile distribution
    of carbon-nitrogen ratios.  M.Sc.  thesis.   The Ohio State University.

-------
                                    -9-


      EPC, i.e.  that solution concentration in which phosphorus  is  neither
 adsorbed nor desorbed,  was determined from a plot of c  vs  x/m, where all
 units are as defined previously.   Sequentially desorbed phosphorus  was
 determined by summing the total P desorbed per unit mass of soil or sediment
 after ten successive 6-hour desorptions into 0.01M CaCl2 at a  10:1  suspension:
 soil ratio.

      Simple product-moment correlations were calculated among  all sorption
 variables (adsorption maximum,  adsorption energy parameter,  EPC,  and sequen-
 tially desorbed P)  with the chemical  and physical properties determined by
 the various  adsorbents.   Correlations were also calculated among all sorption
 variables themselves.
                 Phosphorus-Enrichment  in  Surface Runoff

      Total  sediment  phosphorus  content of runoff from  the agricultural water-
 sheds in Defiance  County  and  Hoytville plots  (Table  1, page   5 ) was determined
 by subtracting  filtered reactive-P  from total-P and  dividing  by sediment
 concentration.   This calculation  overestimates total sediment-P, because the
 total-P determination includes  dissolved  organic-P,which we did not measure.
 However,  at sediment concentrations >  100 ug/ml, this  error is negligible.
 Therefore,  only samples with  sediment  concentrations > 100 ug/ml were used.

      In addition,  on a few  selected samples, percent clay in  the soil and
 runoff were determined as well  as total-P in soil, soil-clay  fraction
 and  runoff  as described previously.  A clay-enrichment ratio was calculated
 as percent  clay in runoff divided by percent clay in surface  soil.  Phos-
 phorus-enrichment  ratio was calculated  in the same manner.


 3.14   Results and  Discussion

                       Adsorbent Characteristics

      Particle-size distribution of the  surface soils in the study (Table 1)
 strongly  shows  the influence  of parent material.   The soils range from a
 low of  22%  clay  (Blount)  to a high of 47% clay (Paulding),with the difference
 largely made up  in the sand separate.   The clays  of the surface soils are
 principally in  the coarse (0.2-2um)-size  fraction.   There is,  on the other
 hand, a much wider particle-size variation in the stream-bottom sediments
 examined.   Of special note is the Independence 12-1-75 sample,which is  very
 sandy  (88%) and atypical  of Maumee Basin sediments  as a whole.  It was  col-
 lected under high-flow conditions relatively close  to the stream bank and
may represent unmixed stream bank scour material.

     Phosphorus status of the soils, clays,and sediments  varies  widely  (Table 1)
Total P values show broad ranges for the surface  soils, from 450 to 1018 ug
P/g soil for the Blount and Roselms  whole  soils,  respectively, while the
clay fractions of  the soils show similar trends over  a  somewhat  narrower range.
Inorganic P levels  are lower and more  tightly grouped for  both soils and clays.

-------
e 1. Some physical and chemical properties
Sand Silt Fine
clay
1
	 /» 	
Soils
Roselms I 7.0 1*6.0 12.5
Broughton 6.6 1*5.1 10.8
Roselms II 10.1* 1*6.8 10.9
Lenawee 12.1* 53.7 9.5
Blount 35. ^ **2.6 1*.5
Pauldtng 5.1 1*9.5 9.7
Hoytville 18.7 1*2.1* 11.1
Soil clay fractions
Roselms I
Broughton
Roselms II
Lenawee
Blount
Paulding
Hoytville
Bottom Sediments
Ind. 12-1-75 88.9 6.0 1.1
Aug. 12-1-75 1=7 57.9 6.5
Tiffin 12-1-75 '*5.5 31.3 5.0
of the soils, their clay fractions and bottom
Coarse Total Total-P
clay clay
3l*. 5 1*7.0 1018
37.5 1*8.3 568
31.9 1*2.8 551*
21*. 1* 33.9 976
17.5 22.0 1*50
35.7 1*5.1* 780
27.8 38.9 816
889
705
738
1290
998
90l»
1120
l*.0 5.1 1*76
33,9 ItO.l* 1260
18.2 23.2 753
Inorganic-P
&/ &
701*
310
333
662
21*8
1*21
566
636
1*38
1*20
81*9
579
1*37
650
379
106^1
61*2
sediments .
Available-P
26.8
2.7
15.8
1*6.1*
13.7
8.6
21.7
nd*
nd
nd
nd
nd
nd
nd
36.7
28.6
2l|.2

Organic-C
1
1.82
1.7l*
1.76
2.13
1.1*7
2.37
2.21*
1.66
1.30
l.5l*
2.78
2,56
2.01
2.22
0.95
2.07
2. Oh

CaCO
equivalent
0
0
0
0
0
0
0
0
0
0
0
0
0
0
12.8
6.6
8.1

pH
6.1*
7.3
6.0
6.6
5.1*
6.7
7.1
nd
nd
nd
nd
nd
nd
nd
7.3
6.9
7.0

-------
Table 1.  continued
                         Sand    Silt    Fine     Coarse     Total


Ind.
Aug.

Bottom Sediments
3-2l*-76 2k. k 37.6
j-2l*-76 3.1, 1,7.5
clay clay clay

6.8 31.2 38.0 9)49
8.8 1*0.3 1*9.1 1150
Jnoreanic-P Avuiluble-P Organic-C
" g —2
692 19.0 1.91
965 13.9 1.82
CuCCU
equivalent

9.9
6.1+
PU

7.0
6.9
   *  Available  P and pH not determined for clay fractions.

-------
                                  -12-


Available P in the surface soil varies from quite low at 2.7 ug/g (Broughton)
to moderately high at 46.4 ug/g (Lenawee).

     Citrate-dithionite-bicarbonate (CDB)  and oxalate Fe and Al as well as
free carbonates have all been identified with phosphate adsorption in soil
and sediment systems and so deserve special notice.   Only the stream sediments
contained free carbonates (Table 1), with  calcium carbonate equivalents as
high as 12.8% for the Independence  12-1-75 sample.  More striking are the
differences in CDB and oxalate Fe content  of the two groups of adsorbents.   CDB
iron in the soil samples is higher than oxalate iron, which is to be expected
since the latter is presumed to remove only a fraction of the CDB iron.  On the
other hand, similar extractions of the bottom sediments yield higher ox-
alate than CDB iron values.  Possible explanations for this phenomenon will
be discussed later.  It nevertheless points to the faulty use of the terms
"amorphous" and "free oxides" of iron when referring to CDB and oxalate extracts,
especially with regard to stream sediments.


                Sorption Characteristics of Adsorbents

     The  form of  the adsorption isotherm for the whole soils, clay  fractions,
and bottom sediments followed  that  of a typical Langmuir-shaped curve, as
shown for the Hoytville silty clay  loam, Figure  1.  In every case, however,
a nearly  vertical  leg appears  in the  isotherm at  low c, giving the overall
curve a "checkmark"  shape.  This phenomenon has  been shown previously in
published data  (19,20) but was not  noted since the authors presented their
results in "simple"  adsorption curves  (x/m vs c).  An adsorption curve such
as  this  for  the Hoytville  soil gives  some  insight as to why  the  effect appears
 (Figure 2).  At  low  final  phosphorus  concentrations  « 0.1  ug/ml)  the system
buffers very strongly against  increases in P concentration,  as indicated by
the very  steep portion  of  the  curve.   The  effect of  this buffering  is  to keep
c nearly  constant while x/m rises  rapidly, such  that the ratio c/(x'/m) de-
creases  sharply  at low  added P.  As the strong buffering region  is  exceeded,
 the simple adsorption assumes  a more  characteristic  asymptotic shape,  corres-
ponding  to the  positive  slope  portion of  the Langmuir  curve.  All  calculations
 of  adsorption maxima and  energy parameters were  determined  from  the slope  and
 intercept of the Langmuir  isotherm,which  could be adequately represented by
 the single-phase form of  the adsorption equation.

      Table 2 shows values  determined  for  the four sorption properties,  i.e.
 adsorption maximum,  adsorption energy,  EPC,  and  sequentially desorbed  phosphorus.
 The adsorption maxima  fall into  three distinct  groups:   the lowest, belonging
 to  the surface soils; next,  and  not surprisingly somewhat  higher,  those of the
 clay fractions  of the  surface  soils;  last and much  higher,  the maxima  of the
 bottom sediments.  A second trend, which  is  not  apparent within groups but
 which does show up when comparing across  groups, is the roughly inverse re-
 lationship between adsorption maximum and adsorption energy.  That is, the
 bottom sediments on average have the highest adsorption maxima but the lowest
 adsorption energies, while the reverse is true for  the surface soils.   The
 mean adsorption maximum of the group of bottom sediments is significantly
 higher than that of the surface soils at the 0.5% level (t=3.88), while the
 adsorption energy of the sediments is lower at the 10% level (t=2,10).

-------
    300
    250
                          -13-
    200
         -
I   isott-
0)      o
                      y=(3.87 x 10'3)x  26.1 x 10'4
                  50          100         150
                      c(ug/ml) x 10'2
       Figurel. Langmuir adsorption isotherm for phosphate on
             Hoytville clay loam.

-------
200
                               -14-
150
                 50
100
150
200
                        c(ug/ml) x 102 EPC
     Figure 2. Adsorption isotherm for phosphate on Hoytville clay loam.

-------
Table 2.  Extractable iron,  aluminum,and  silicon in the soils,  their  clay fractions,and bottom sediments

Soils
Roselms I
Broughton
Roselms II
Lenawee
Blount
Paulding
Hoytville
Soil-clay fractions
Roselms I
Broughton
Roselms II
Lenawee
Blount
Paulding '
Hoytville
Amorphous Si

12.1
10.2
11.6
12.3
12.5
10.9
11.0

11.00
7.05
7.58
5.81
5.13
6.46
4.98
CDB-Fe

20.0
21.2
25.0
17.5
32.5
21.2
16.3

17.5
15.0
14.4
7.5
11.9
15.0
9.7
CDB-A1
	 mg/g 	
2.55
2.55
2.35
3.00
4.70
2.15
2.15

1.50
1.33
1.49
0.70
1.45
1.15
0.85
Oxalate-Fe

15.5
6.7
13.3
15.5
16.3
35.0
10.7

16.50
5.42
13.10
7.67
10.10
14.30
6.67
Oxalate-Al

2.83
2.75
3.33
5.33
3.50
5.50
4.17

1.92
1.50
2.08
2.17
1.33
2.42
2.00

-------
Table 2.  continued
Amorphous Si
CDB-Fe
Bottom Sediment
Ind.
Aug.
12-1-75
12-1-75
Tiffin 12-1-75
Ind.
Aug.
3-24-76
3-24-76
0.
6.
3.
4.
6.
98
63
17
72
22
4.4
13.8
10.0
11.2
13.8
CDB-A1
_ / _
Oxalate-Fe Oxalate-Al
	 mg/g 	
0
0
0
1
1
.40
.95
.70
.25
.40
9.
26.
17.
20.
27.
2
0
7
7
1
0
3
1
2
3
.83
.00
.83
.67
.50

-------
                               -17-
Table 3.   Phosphate sorption properties  of the  soils,  their clay  fractions,
          and bottom sediments

Adsorbent
Soils
Roselms I
Broughton
Roselms II
Paulding
Lenawee
Blount
Hoytville
Adsorption
max imum
(ug/g)

287
209
249
216
244
199
258
Adsorption
energy
(ml/ug)

1.69
4.89
2.85
4.35
0.80
2.15
1.49

EPC
(ug/ml)

0.032
0.008
0.017
0.140
0.060
0.011
0.240

P desorbed
(ug/g)

1.77
0.46
0.57
0.29
3.58
0.75
0.91
Soil-clay fractions
Roselms I clay
Broughton clay
Roselms II clay
Paulding clay
Lenawee clay
Blount clay
Hoytville clay
Bottom Sediment
Ind. 12-1-75
Aug. 12-1-75
Tiffin 12-1-75
Ind. 3-24-76
Aug. 3-24-76
393
323
411
455
422
538
623

222
4870
1930
3580
4550
0.86
4.15
1.91
1.09
0.82
7.43
1.63

1.00
0.68
1.55
1.05
1.36
0.034
0.016
0.016
0.008
0.032
0.006
0.008

0.035
0.054
0.026
0.024
0.024
2.21
0.95
0.99
1.12
3.68
1.13
1.18

3.98
3.61
1.33
1.42
1.81

-------
                                  -18-
     Simple product-moment correlations  for  the adsorption parameters of the surface
soils versus selected chemical  and  physical  properties  of those soils are^shown in
Table 4.  Only those parameters giving correlations significant at the 10%
level or lower are presented.  The positive relationship of adsorption maximum
with total and inorganic P suggests that those soils which exhibit the highest
adsorption have already retained large quantities of P from former environ-
ments and that this P is held principally in the inorganic form.   A weak posi-
tive correlation of adsorption maximum with fine clay content also appears,
most likely a function of  the large adsorptive surface of that fraction.

     Those parameters related to the stability of the complex formed between
the phosphate and the surface, i.e. adsorption energy, EPC, and sequentially
desorbed phosphorus, are  all correlated with Bray PI "available" P.  As would
be expected, the correlation with  "available" P is negative for adsorption
energy  and positive  for the other  two parameters.  This demonstrates that
these  four  parameters are indicators of the same phenomena, that is, the
extent  to which phosphorus would be released after relatively long washing or
mild acid  extraction. The correlation of EPC and sequentially desorbed P with
inorganic  P  suggests  that it is this fraction which  is  the source of phosphorus
lost  from  the adsorbent.

     Correlations involving  the clay fractions of  the watershed Ap horizons
are  shown  in Table 5.   The positive correlation between adsorption maximum
and  organic  P appears  to  be  artifactual since no similar relationship  occurs
with the remaining data or is  reported  in  the literature.  The artifact may
 arise  as a consequence  of the  sedimentation procedure  used to  fractionate  the
 clays  or their  storage  as a  slurry in methanol and water,  either  of which  may
 have altered the  inorganic P content of the clay.   The significant correlation
 of adsorption  energy with CDB  iron and  aluminum  is  the first  evidence  that
 these components, which are  operative  in P retention by more  highly weathered
 soils, may also be  contributing  to the  phenomena here.  Larger quantities  of
 total and  inorganic  P result in more P  release  in the 6-hour desorption
 sequence,  as was  noted  for surface soils.   These  last two  observations  imply
 that the desorbed phosphorus is  being  held by some fraction  other than the
 CDB Fe and Al.   This appears to  be the case since  adsorption energy and thus
 the stability of the phosphorus  complex increases  in parallel with the magni-
 tudes of these two Fe and Al fractions.

      Table 6 shows correlations  of sorption parameters determined for the
 stream-bottom sediments with certain of their chemical and physical properties.
 The most striking result  from the analyses is the strong positive correlation
 of adsorption maximum with a number of iron,  aluminum, and silica fractions
 in the sediments.  This  group includes amorphous silica,  oxalate Fe and Al,
 CDB Fe, and to a lesser  extent CDB Al.  Their presence together is in sharp
 contrast  to the surface  soils or  clays in which none of these chemical properties
 were strongly correlated with adsorption maximum.   It is likely that the Fe,
 Al, and Si  fractions are associated in some type of amorphous or no more than
 weakly crystalline  complex.  Such a complex, when highly hydrated as it
 would  be  in a stream system, has  a gel-like reactivity.  A type of gel has
 been proposed as a major agent in P sorption by lake sediments since it has
 a very large reactive  surface area with many sites  for phosphate retention
 due  to its high hydration (22,27).

-------
                                -19-
 Table 4.
Correlation coefficients of P sorption parameters versus physical
and chemical properties of the soils (n = 7)*
Adsorption
maximum
Bray PI
"Available" P n.s.
Total P 0.729
Inorganic P 0.809*
Fine clay 0.735
Adsorption EPC Desorbed P
energy
-0.868* 0.976** 0.953**
n.s. 0.768* 0.680
n.s. 0.804* 0.724
n.s. n.s. n.s.
* Values not marked are significant at the 10% level; those marked with a
   *  are significant at the 57, level; those marked with a "**" are sig-
  nificant at the 17. level; "n.s." indicates no significant relationship
  between the factors at the 107» level.

-------
                                -20-
Table 5.   Correlation coefficients of P sorption parameters  versus  physical
          and chemical properties of clay fractions  of surface soils  (n = 7)*
Adsorption
maximum
CDB-Fe n.s.
CDB-A1 n.s.
Total P n.s.
Inorganic P n.s.
Organic P 0.741
Fine clay n.s.
Adsorption EPC Desorbed P
energy
0.841* n.s. n.s.
0.823* n.s. n.s.
n.s. n.s. 0.706
n.s. n.s. 0.871*
n. s . n. s . n.s.
-0.679 n.s. n.s.
*  Values not marked are significant at the 10% level; those marked with
   a "*" are significant at the 5% level; those marked with a "**" are
   significant at the 17= level; "n.s." indicates no significant relationship
   between the factors at the 10% level.

-------
                                 -21-
 Table  6.  Correlation coefficeints of P sorption parameters versus physical
          and chemical properties of bottom sediments  (n = 5)*


Amorphous Si
CDB-Fe
CDB-A1
Oxalate-Fe
Oxalate-Al
Bray PI
"Available" P
Total P
Inorganic P
CaC03
equivalent
Total clay
Fine clay
PH
Adsorption Adsorption
maximum energy
0.997** n.s.
0.972** n.s.
0.847 n.s.
0.981** n.s.
0.974** n.s.
n.s. n.s.
0.993** n.s.
0.960** n.s.
-0.850 n.s.
0.967** n.s.
0.916* n.s.
-0.923* n.s.
EPC Desorbed P

n.s. n.s.
n.s. n.s.
n.s. n.s.
n.s. n.s.
n.s. n.s.
n.s. n. s.
n.s. n.s.
n.s. n.s.
n.s. n. s.
n.s. n.s.
n.s . n.s .
n.s. n. s.
*  Values not marked are significant at the 10% level;  those marked with
   a "*" are significant at the 5% level;  those marked  with "**"  are  sig-
   nificant at the 1% level;  "n.s." indicates  no significant relationship
   between the factors at the 10% level.

-------
                                  -22-


     Surface~area considerations also explain the correlation of adsorption
maximum with both total and fine clay content in the sediments.   As was  the
case for the surface soils and their clays,  those adsorbents which have  the
highest adsorption maxima appear to have retained the most P in the past.   This
is indicated by the highly significant correlations with both total and  in-
organic P.  Only with the bottom sediments is pH of the adsorbent also related
to adsorption - the lower pH sediments show the higher adsorption maxima,  as
would be expected from changes in surface charge.  The reason for the negative
relationship between adsorption maximum and carbonate content is not as
obvious.  It has been proposed in comparisons of the total phosphorus content
of calcareous and non-calcareous lake sediments that carbonates are a much
less energetic phosphorus retention agent than are iron and aluminum hydrous
oxides  (22).  The carbonate fraction thus acts as a dilutant to the Fe and
Al sites and so would correlate negatively with P content.  Similarly, in
reference to the adsorption maximum, carbonates are a less active adsorbent
and effectively dilute the adsorbing capacity of the sediment.

     Cross-correlations of the adsorption and desorption properties of the
soils,  clays, and sediments (data not presented) generally substantiate the
conclusions drawn above.  For the surface soils, negative correlations appear
between adsorption energy and EPC, and adsorption energy and sequentially
desorbed P  (10%), and a positive correlation between EPC and sequentially
desorbed P  (1%).  The implication is that both EPC and sequentially desorbed
P are controlled by the stability or energy of the phosphorus-colloid complex.
A less  complete set of correlations appears for the soil clays and bottom
sediments,  but the same general trends ocdur:  a positive correlation (57=)
between EPC and sequentially desorbed P for the soil clays; and a negative
correlation between EPC and adsorption energy for the stream sediments.


                   Differences in Phosphate Sorption

     Two hypotheses may be advanced to explain the significant differences
in phosphate sorption properties, especially adsorption maximum, between the
watershed surface soils and the stream-bottom sediments:  1) Preferential
erosion of  some reactive-size fraction of the whole soil followed by  further
concentration of this  fraction by selective transport in the stream itself;and
2) chemical alteration of  the eroded soil material after deposition in the
stream.

     Since  clays are the most reactive of the soil separates by nature of
their  large surface area per unit mass, the clay  fractions of the surface
soils were  used to  test the first hypothesis.  As Table 3 shows,  the  adsorp-
tion maxima of  the  total clay fractions, while higher than  those  of the whole
soils,  aare  not  even of  the same order of magnitude as those of  the sediments.
Calculations even suggest  error in  the inference  that the clay  fraction is  the
only phosphorus-adsorbing  medium  in the soil.  Expressing the adsorption
maximum of  the whole soil  on a  100% clay basis as  follows:
          maximum  (10070 clay) = maximum  (whole soil) ~ % clay
should  give a value equal  to  the  adsorption maximum of the  clay fraction.
In  every  case,  however,  the calculated value  exceeds  the  experimental,

-------
                                   -23-


 indicating the presence of other adsorbents in the whole soil besides clay.
 In addition, particle-size analysis of the stream sediments (Table 1) shows
 that they are not 100% clay at all, but generally 40 to 45% each clay and
 silt.  The eroded and transported clay fraction cannot by itself account  for
 all of the adsorption by the bottom sediments or even by the whole soils.

      The alternate hypothesis states that alterations occur in the streambed
 which markedly increase adsorption capacity of the eroded soil.   The chemical
 properties most strongly correlated with adsorption for the sediments appear
 to be the most likely indicators of such _in situ reactions,  specifically
 oxalate Fe and Al, CDB Fe and Al, and amorphous Si.   The nature  of such a
 complex has been mentioned previously>but its origin has not been described.
 The structure of iron-substituted clay minerals is largely tied  to the presence
 of iron in the +3 oxidation state.   Exposure to a redox potential only as
 low as +100 mv at pH 7 can reduce iron to Fe2+ (7),and the accompanying change
 in ionic radius and charge results  in at least partial decomposition of the
 mineral.   Redox potentials and pH's  in bottom sediments of Maumee Basin streams
 have been measured as low as -371 mv at pH 7.2 (personal communication,
 Dr.  Fred Rhoton,  Agronomy Dept.,  Ohio State University).   Upon reoxidation,
 iron will precipitate and may form  an amorphous,  highly hydrated  solid phase
 with the structural Si and Al which  were released with the reduced Fe.

      The absolute quantity of amorphous material,  especially iron,  also
 suggests  that alteration has occurred.   Normally CDB  extraction  is  assumed
 to remove all free iron and aluminum oxides,  both crystalline  and  amorphous,
 while oxalate removes the amorphous  fraction only.   Table  2  shows,  however,
 that in every stream-bottom sediment the quantity of  oxalate iron  is  about
 twice that of CDB iron.   This has been reported  elsewhere  in the  literature
 (27)  and  suggests  that the stream sediments  used  here have been subject to
 processes  causing measurable changes  in their  structure as compared  to surface
 soils.   It is likely that most of the  Fe extracted by CDB  in the bottom sed-
 iments  is  also oxalate extractable and,  in  addition,  oxalate may be  extracting
 some Fe-carbonate  forms  such as siderite.   The  latter  explanation  is  supported
 by the  lack of free  carbonate in  the soils  studied while the bottom  sediments
 did  contain carbonates.   Substantial amounts of magnetite were recovered
 from several  Maumee  River  Basin sediments  (Dr. Fred Rhoton, Agronomy Dept.,
 Ohio  State University personal communication),and Gamble and Daniels  (6)
 have  reported  that while magnetite had  appreciable solubility in oxalate
 it was  only sparingly soluble  in CDB.   Iron  is released as a weathering
 product from  soils under common ambient conditions and may find its way into
 streams,where  it coats  eroded  soil particles.  Groundwater is also a source
 of soluble iron from  pedogenic processes or when it flows through iron-
 containing rocks and  so may  add to stream Fe levels.


               Phosphorus Enrichment in Runoff Sediment

     The total particulate P concentration of runoff sediment was estimated
 by subtracting filtered reactive P (FRP) from total P for runoff  events
where suspended solids concentration was > 100 ug/ml.  The calculation over-
 estimates  total particulate P, because the total P measurement contains some
 filtered non-reactive P which was not measured routinely in this  study.  However,
at suspended solid concentrations > 100 ug/ml, this error is not  significant.

-------
                                  -24-
Th e ratio of total particulate P to total  surface soil  P is  given  in  Table  7
and is compared with the clay-enrichment ratio and the  p-enrichment ratio
determined from the P content of the clay fractions.  The P  enrichment  appears
to be somewhat correlated with clay enrichment although clay enrichment
usually underestimates P enrichment. The difference can be attributed to P
in organic matter in  finer -size fractions,  and  the adsorption  of P  from
solution by sediment during transport across  the  field.   In  the  interesting
case of the Roselms (111) soil, there is a negative P enrichment which  is
supported by the lower total P content of the clay fraction  versus the
total soil.

     It has been theorized that as erosion becomes more severe,  clay-enrichment
(and, therefore P-enrichment ratio) ratio decreases as  more  of the coarser
sized particles are eroded and transported.  This means that as  soil  loss  is
decreased, there will not be a proportional decrease in total P  loss, since
soluble P is not affected substantially by soil-loss reduction,  and the sediment
that is lost is the finer-textured fraction,which is highest in  total P.


3.15  Conclusions

     Strong contrasts appear in both adsorption and desorption behavior of
the surface soils and stream sediments.  Desorption, as measured by EPC and
sequentially desorbed phosphorus, is not significantly  different between the
two groups of adsorbents.  This is apparently due to low adsorption by  the
sediments such that their adsorbing capacities have not yet  been saturated.
The total phosphorus content of the sediments reflects  this  in that they are
not appreciably higher than the phosphorus contents of  the surface soils.

     On the other hand,  the bottom sediments  have significantly  higher  ad-
sorption maxima and significantly lower adsorption energies  than do the surface
soils.  Since the total P content of the sediments is less than  their adsorp-
tion capacities (except for the Independence  12-l-75)> it may be  concluded
that they still have a large portion of their adsorption capability remaining.
The sediments will act as phosphate "sponges" during transport until  such
time as a large enough fraction of their adsorption capacity is  occupied so
as to make desorption of weakly held phosphorus likely.  Desorption may also
become a problem if the environment of the sediment, such as pH  or solution
phosphorus status, changes in such a way as to destabilize the sediment-
phosphorus complex.  Even though only single-phase adsorption was  detected
for both soils and sediments in this study, the possibility  is still  open
that a change in phosphorus status of the sediments may alter the  energy of
the adsorbing sites, changing their role in the phosphate balance  of  the
stream or lake.

     Future study is required to identify precisely the mechanism  and nature
of the changes which occur to eroded sediment after its deposition in the
stream, as well as the behavior of the sediment as a phosphorus  adsorbent
under varying ambient conditons of redox potential, pH, and  time.

     Phosphorus in runoff sediment is higher  than in the parent  soil.  This
is due to the higher content of clay in runoff and possibly  to sorption of
soluble P by the reactive sediment during initial transport.

-------
Table   7.   Phosphorus and clay enrichment of runoff sediment
UJ.UC p
Soil

Roselms 111 1*7,0
Roselms 111 -
Roselms 201 1*2.8
Roselas 201
Blount 1*01 22.0
Blount 1*01
Hoytville 6ll 38.9
oxay
Runoff

51*. 2
61.2
1*7.8
1*0.8
1*6.7
55.1
59.8
	 f content ug/g 	 clay
Whole Clay Runoff Enrichment
soil fraction
1018 889 280
1*1*6
551* 738 ll*21
101*1
**50 998 1005
1368
8l6 1120 1591

1.15
(630)*
1.30
1.12
(731*)
0.95
2.12
2.50
1.5*
Phosphorus
Enrichment

0.28
(0.62)*
O.UU
K5
2.56 V
(1.32)
1.88
2.23
(3.26)
1.95
  Values of runoff P content and P enrichment in parentheses are means of runoff samples  from May 1975 - May 1976.

-------
                                  -26-
3.16  Literature Cited

1.  Armstrong,  D.E.,  K.W.  Lee,  P.B.  Uttomark,  D.R.  Keeney,  and  R.F.  Harris.
    1974.   Pollution of the Great Lakes by Nutrients  from Agricultural  Land.
    In:   Management programs,  research and effects  of present land  use  ac-
    tivities on water quality of the Great Lakes.   Pollution from Land  Use
    Activities  Reference Group of the International Joint Commission.

2.  Bray,  R.H.  and L.T. Kurtz.   1945.  Determination of Total,  Organic,  and
    Available Forms of Phosphorus.   Soil Science.   59:39-45.

3.  Carter, D.L., J.A. Bondurant, and C.W. Robbins.  1971.   Water-Soluble
    N03-N, P04-P, and Total Salt Balance on a Large Irrigation Tract.   Soil
    Sci. Soc. Am. Proc.  35:331-335.

4.  Dreimanis,  Aleksis.  1962.   Quantitative Gasometric Determination of
    Calcite and Dolomite by Using Chittick Apparatus.  Jour.  Sed. Petr.
    32:520-529.

5.  Edzwald, J.K., D.C. Toensing, and M.C.  Leung.   1976.  Phosphate Adsorption
    Reactions with Clay Minerals.  Environ. Sci.  Tech.  10:485-490.

6.  Gamble, E.E., and R.B. Daniels.   1972.  Iron and Silica in Water,  Acid,
    Ammonium Oxalate, and Dithionite Extracts of Some North Carolina Coastal
    Plain Soils.  Soil Sci. Soc. Amer. Proc.  36:939-944.

7.  Gotch, S.,  and W.H. Patrick.  1974.  Transformation of Iron in a Water-
    logged Soil as Influenced by Redox Potential and pH.  Soil Sci. Soc. Am.
    Proc. 38:66-71.

8.  Hashimoto,  I., and M.L. Jackson. 1958,  Rapid Dissolution of Allophane and
    Kaolinite-Halloysite Afer Dehydration. In:Clays and Clay Minerals:
    Proceedings of the Seventh National Conference  on Clays and Clay Minerals.
    Ada Swineford (ed.). Pergamon Press, Inc. New York, pp  102-113.

9.  Kingston, F.J., R. J. Atkinson,  A. M. Posner, and J. P. Quirk.  1967. Specific
    Adsorption of Anions. Nature. 215:1459-1461.

iO. 	, A. M. Posner, and J. P. Quirk.  1972. Anion Adsorption by  Goethite
    and Gibbsite. I. The Role of the Proton in Determining Adsorption Envelopes.
    Jour. Soil Sci. 23:177-192.

11. Hooper,  Frank F.  1969. Euthrophication Indices and Their Relation to Other
    Indices  of Ecosystem Change. In: Euthrophication: Causes, Consequences,
    Correctives. National Academy of Science. Washington, D.C. pp 225-235.

12, John, Matt K. 1970. Colorimetric Determination of Phosphorus in Soil and
    Plant Materials with Ascorbic Acid.  Soil  Science. 109:214-220.

-------
                                 -27-


 13. Johnson, A.H.,  D. R. Bouldin, E. A. Goyette, and A. M. Hedges: 1976.
     Phosphorus Loss by Stream Transport from a Rural Watershed: Quantities,
     Processes, Sources. Jour, Environ. Qual. 5:148-157.

 14. Kuo, S., and E. G. Lotse.  1974.  Kinetics of Phosphate Adsorption and
     Desorption by Lake Sediments. Soil Sci. Soc. Am. Proc. 38:50-54.

 15. Li, W.  C., D. E.  Armstrong,  J.D.H. Williams, R.  F. Harris, and J.K. Syers.
     1972.  Rate and  Extent of Inorganic Phosphate Exchange in Lake Sediments.
     Soil Sci.  Soc.  Am. Proc. 36:279-285.

 16. Mehra,  O.P., and  M.L. Jackson.  1958.  Iron Oxide  Removal from Soils and Clays
     by a Dithionite-Citrate System Buffered with Sodium Bicarbonate.  In: Clays
     and Clay Minerals. Proceedings  of the Seventh National Conference on Clays
     and Clay Minerals. Ada Swineford (ed).  Pergamon  Press, Inc. New York,  pp 317-327

 17. Metha,  N.C., J. 0. Legg, C.A.I.  Goring, and C.A. Black.  1954.  Determination
     of Organic Phosphorus in Soils:  I.  Extraction Method.  Soil Sci. Soc. Amer  Proc
     18:443-449.

 18. Olsen,  S.R., and  F.  S.  Watanabe.  1957.  A Method  to Determine a Phosphorus
     Adsorption Maximum of Soils  As Measured by the Langmuir  Isotherm.  Soil Sci
     Soc. Am. Proc.  21:144-149.

 19.  Ryden,  J.C.,  J. K.  Syers, and R.  F. Harris.  1972.  Potential of an  Eroding
     Urban Soil for  the Phosphorus Enrichment of  Streams:  I.  Evaluation of  Methods.
     Journ.  Environ. Qual.  1:430-434.
20.
                                                 1972. Potential of an Eroding
    Urban Soil for the Phosphorus Enrichment of Streams:  II.  Application of Adopted
    Method. Jour. Environ Qual. 1:434-438.
21. Saunders, W.M.H. 1965. Phosphate Retention by New Zealand Soils and Its Re-
    lationship to Free Sesquioxides, Organic Matter, and Other Soil Properties.
    New Zeal. Jour. Agr. Res. 8:30-57.

22. Shukla, S.S., J. K. Syers, J.D.H. Williams, D. E. Armstrong, and R. F. Harris.
    1971. Sorption of Inorganic Phosphorus by Lake Sediments. Soil Sci. Soc  Am
    Proc. 35:244-249.

23. Sommers, L.E. and D. W. Nelson. 1972. Determination of Total Phosphorus in Soils
    A Rapid Perchloric Acid Digestion Procedure.  Soil Sci. Soc.  Am. Proc.  36:902-904,

24. Stewart, K.M.,  and G.  A.  Rohlich. 1967. Eutrophication—A Review.  State of
    California.  The Resources Agency. State Water Quality Control Board. Pub.No.  34.

25. Taylor, A.W.  and H.  M.  Kunishi. 1971. Phosphate Equilibria on Stream Sediment
    and Soil in  Watershed  Draining and Agricultural Region.  Jour. Agr.  Food Chem.
    19:827-831.

-------
                                   -28-
26,  Upchurch,  Joseph B,,  James K,  Edzwald, and Charles R, O'^elia, 1974,
    Phosphates in Sediments of Pamlico Estuary. Environ, Sci, Tech, 8;56^58,

27.  Williams,  J,D.H,, J,  K. Syers, S.S. Stmkla, R. R, Harris, and D, E. Armstrong,
    1971, Levels of Inorganic and  Total Phosphorus in Lake Sediments as Related
    to Other Sediment Parameters.  Enciron. Sci, Tech. 5:113^1120.

-------
                                  -29-
3.2  Phosphate adsorption-desorption characteristics of suspended sediments
     in the Maumee River Basin of Ohio
3.21  Abstract

     P adsorption-desorption characteristics of Maumee River Basin suspended
sediments were compared with those of Basin soils and stream-bottom sediments.
Suspended sediment contained more total P than either soils or bottom sediments.
The increase in total P over soils is attributed to enrichment of P in sed-
iment by selective erosion of fine particles and adsorption of P during fluvial
transport.  The suspended sediment had higher adsorption maxima than Basin
soils but lower than bottom sediments and had lower adsorption energies than
either soils or bottom sediments.  Calcite content of the suspended sediments
was positively correlated with total P, EPC (equilibrium P concentration)  and
P desorbed and negatively correlated with adsorption energy, implying that al-
though calcite is a sink for P,  the adsorption is weak compared with other
sites for P adsorption such as hydrous oxides of Fe and Al.

-------
                                  -30-
3.22  Introduction

     Many workers (Syers et al., 1973;  Williams et al.,  1971;  Shukla et a1•>
1971) have studied the mechanisms by which P is adsorbed and retained by
sediment.  The dominant role of iron as an adsorption and binding site for
soluble P was documented, even in calcareous sediments.   Although calcite
will adsorb P (Cole et al., 1953), the iron oxides have a greater capacity
to do so.

     In calcareous streams, secondary calcite can be formed (Wall and Wilding,
1976), probably as a result of CC>2 consumption by algae,which raises pH
and shifts carbonate equilibrium toward calcite formation.  Green (1977)
found that calcite was formed in the Maumee River Basin in the summer when
algal populations in the stream were at a maximum.*-

     The objective of this study was to study the phosphate adsorption/desorp-
tion characteristics of Maumee River Basin suspended sediments containing
varying amounts of calcite and to compare these characteristics with those
of watershed soils and stream-bottom sediments (McCallister and Logan, 1978).
3.23  Methods and Materials

     A grab-sample approach was utilized to obtain suspended sediment samples
from the Maumee River and its tributaries (Figure 1).  A plastic bucket was
lowered into the middle of the stream and allowed to fill without disturbing
the stream floor.  Approximately 40 liters of streamwater were transferred
to plastic containers and transported to the laboratory for analysis.  Oc-
casionally more than 40 liters of streamwater were collected when additional
suspended sediment was needed.  Samples were collected on 20 different oc-
casions during the sampling period from September 10, 1975 to September 17,
1976.  Time limitations and little variation in solution and sediment charac-
teristics of the small Maumee River tributaries made it impractical and un-
necessary to collect a streamwater sample from all the sites on each sampling
date during the summer months (June-September, 1976).

     For the first half of the sampling period (September, 1975 - April 13,
1976), suspended sediment was concentrated by flocculation with 100 ml of
IN MgCl2 per 40 liters of samples.  The floccules were allowed to settle for
2 days and then the supernatant was siphoned off.  The concentrated sediment
was transferred to centrifuge tubes, washed with methanol and distilled water,
centrifuged for 10 minutes at 1500 rpm and decanted.  The washing procedure
was repeated twice to ensure removal of excess salts.  Samples containing
more than 2g of sediment were freeze-dried and all other samples were dried
by evaporation at room temperature.  After April 13, 1976 the suspended
sediment was allowed to settle for 4-5 days without addition of MgC^ to
eliminate sample pretreatment which might affect calcite stability.  The
   Dan B. Green.  1977.  Calcite Occurrence, Stability, and Phosphorus
    Interactions in the  FluvialMedia Exiting the Maumee River Drainage
    System.  M.S. Thesis.  Agronomy Department, The Ohio State University.

-------
                       Defiance County, Ohio
   • Tiffin R. Tributary
   + Tiffin River  Defiance- Maumee R.
   • Defiance - Maumee
   AAugiaize River
   -fc Independence - Maumee R.
   c St. Rd.  127 - Maumee R.
               <; Michigan :•:#•:•::
Maumee River
                                                          Auglaize
                                                          River
Figure 1. Suspended-sediment sampling sites in the Maumee River Basin.

-------
                                  -32-
supernatant was siphoned off and the remaining solution evaporated at room
temperature.

     During periods of medium to low stream flow, the suspended sediment
concentration was so low that only l-7g of dry sediment was collected from
80 liters of streamwater.  Since considerably more sediment was needed to
perform routine analyses, it was necessary to composite the samples exhibiting
similar stream-flow conditions and calcite content into one composite sample
per site.

     Calcite and dolomite were identified qualitatively by X-ray diffraction
and quantitatively by gasometric analysis using a Chittick apparatus following
decomposition by 6N HC1 (Dreimanis, 1962).  A summary of calcite content of
composited samples is given in Table 1.


                Phosphorus Adsorption-Desorption Study

     Total P:  Total phosphorus was determined by perchloric acid digestion
using 0.2 g of each composite sediment sample (Sommers and Nelson, 1972).
The digestion tubes were heated in a machined aluminum block to ensure uniform
heating to 205°C for one hour.  The cooled digest was diluted to 50 ml with
distilled water, inverted several times, and allowed to settle overnight.
A clear 5-ml aliquot was neutralized with 5N NaOH to a 2,  4 dinitrophenol
endpoint, and the color was developed as an ascorbic acid-reduced phosphomolyb-
date complex (Murphy and Riley, 1962, as modified by John, 1970).  The only
change in the procedure was the use of lOg of ammonium molybdate per liter
of stock solution to give better color stability with time.  A Beckman model
24 extended-range spectrophotometer was used to read the color intensity at
730 nm,which was standardized with a digested water blank and P standards.

     P adsorption:  Phosphorus additons of 0, 25, 50, 100, 200, 250, 375,
500 ug were placed in 50-ml plastic centrifuge tubes,each containing 0.5g
of composite suspended sediment.  Also 0.5 ml of 0.05 M CaCl2 and 3 drops of
toluene were added to each tube to reduce the effects of unequal ionic con-
centrations and to reduce microbial activity, respectively.  The volume was
diluted to 25 ml with distilled water and placed on a rotary shaker for 24
hours at 24i 1°C.  Following shaking, each sample was centrifuged at 2500 rpm
for 10-15 minutes, then filtered with 0.2 urn pore-size Nucleopore membranes.
A 5-20 ml aliquot of each filtrate, depending on the P additions, was used for
P determination (same as that used in total P procedure).   The P adsorption
parameters calculated from this experiment were the final concentration (c)
and the amount of P adsorbed per unit of mass (x/m) .  From these parameters,
a Langmuir adsorption isotherm was plotted (c/x/m versus c), from which the
adsorption maximum (I/slope of the line) and the adsorption bonding energy
(slope/y-intercept) was determined (Olsen and Watanabe, 1957; Hsu, 1964;
Woodruff and Kamprath, 1965; Chen et al , 1973).  The equilibrium phosphorus
concentration (EPC) was determined from the plot of c versus x/m.  The P
concentration at x/m = 0 was taken to be the EPC (Taylor and Kunishi, 1971).

-------
Table 1.  Stream samples composited according to stream flow and calcite content of suspended sediment
Composite Sediment
Sample Number
1
2
3
1*
5
6
7
8
9
10
11
12
13
11+
Site
1
1
2
3
U
5
5
6
7
8
8
8
9
11
Period Sampled
(1976)
Tiffin River tributary
3/19 - U/13
7/1 - 8/31
2/19 - U/13
2/19 - U/13
2/19 - U/13
Tiffin River
2/19 - U/13
7/1 - 9/8
Maumee River tributary
2/19 - U/13
2/19 - U/13
Maumee River
2/19 - U/13
7/1 - 7/15
7/20 - 8/18
2/19 - U/13
Auglaize River
2/19 - U/13
Stream
Flow
High
Low
High
High
High
High
Low
Med - High
High
High
Medium
Low
High
High
Percent Calcite
in Sediment
l.U
7.1
1.7
3.6
2.0
1.6
6.0
l.U
1.3
3.8
13.3
33.5
1.6
2.3
                                                                                                           u>
                                                                                                           I

-------
                                 -34-
     P Desorption:  One gram of each composite suspended-sediment sample was
diluted to 25 ml with 0.5 ml of 0.5 M CaCl2 (final concentration was 0.01
M CaCl2), 3 drops of toluene and distilled water in 50-ml plastic centrifuge
tubes.  Samples were allowed to equilibrate on a rotary shaker for 6 hours,
centrifuged at 2500 rpm for 10 minutes, and filtered through 0.2 urn pore
Nucleopore membranes.  A 20-ml aliquot was color-developed and the final P
concentration determined by the same procedure previously stated in P adsorp-
tion study.  The preceding desorption procedure was repeated 9 times and the
total P desorbed for 10 cycles was reported.
                        Statistical Analysis

     Simple product-moment correlations were calculated between calcite
content and the P parameters (total P, adsorption maximum, adsorption bonding
energy, EPC, and total P desorbed) of the composite suspended-sediment samples.
Correlations were also correlated among the P parameters themselves.
3.24  Results and Discussion

             Phosphorus Adsorption-Desorption Characteristics

     The ability to accurately predict P adsorption is dependent on data
conformity to an adsorption isotherm such as the Langmuir adsorption
isotherm (c/x/m = c/b + 1/kb) where c = equilibrium P concentration,
x/m = P adsorbed per unit weight of sediment, k = constant related to
bonding energy,and b = maximum amount of P adsorbed.  In all cases in
this study, the data did conform to the Langmuir adsorption equation by
exhibiting one linear relationship between c/x/m and c.  Figure 2 is
representative of the Langmuir plots obtained in this study.  Due to the
limited quantity of sample, only 6-8 points were obtained to define the
line.  For a few samples, which contained high amounts of P initially, P
desorption occurred at the low-equilibrium P concentrations and a "check
mark"-shaped curve resulted.  This has been reported and discussed by
McCallister and Lopan (1978).  When this situation occurred, only 3-5
points defined the normal adsorption part of the curve.  Due to the minimal
data defining the isotherms, the Langmuir adsorption isotherm plots will
only be used to indicate general trends and correlations.  More specific
statements concerning individual adsorption values will only be made with
extreme caution.

     Suspended-sediment data are given in Table 2 andare compared in Figure 3
with values obtained by McCallister and Logan (1978).  In general, total P
for composite suspended-sediment samples x^ere higher than for Maumee Mver
Basin surface soils, their clay fractions, and bottom sediments.  This indi-
cates that while P enrichment by selective erosion of the finer particles
may explain some of the increase in total P, suspended sediments entering the
Maumee River drainage system serve as a sink for P due to increased contact
between sediment and soluble P in the fluvial system and uptake of soluble P
by algae,which constitute part of the sediment, particularly at low flows.
However, once deposition occurs, P may be released by the reduction of iron
or decomposition of algae; this is indicated by the lower total P values for
bottom sediments.

-------
    30
CO
b
 x  20
    10
o
     0
                l
          y  (1.34 x 103)x+4.08x 10 3
          r  0.998
J	L
                _L
                                                                                 LO
                                                                                 Ol
200
                       800
1000
                  400       600
                   c(Mg/ml) x 10'2
Figure 2. Langmuir  adsorption isotherm of suspended sediment for phosphate.

-------
Table 2.  Total P, adsorption maximum, adsorption energy, equilibrium P concentration  (EPC), and
          total P desorbed for stream composite suspended-sediment samples.

Composite Sediment
Sample Number
1
2
3
k
5
6
7
8
9
10
11
12
13
14
Total P
ug/g
915
1080
1085
1760
970
1225
1420
1520
1070
1075
i860
1890
1360
1245
Adsorption
maximum
ug/g
817 7
- t
801.7
206?. 1
510.0
868.4
1597.1
878.5
823.9
1288.9
1205.7
483.1
778.7
744.6
Adsorption
energy
ml/ug
0.45
_
0.44
0.05
0.45
0.38
0.15
0.40
0.42
0.13
0.15
0.11
0.44
0.33
EPC
ug/ml
0.06
_
0.05
0.58
0.10
0.07
0.20
0.10
0.03
0.07
0.65
1.01
0.13
0.18
P desorbed
ug/g
10.5
20.4
13.4
93.8
9.9
18.9
4o.4
17.6
Ik. 2 ,
15.8 £
62.4
104.8
22.2
37.3
j-  not determined because of insufficient sample

-------
4200
3900
3600
3300
3000
2700
2400
2100,
1800
1500
1200
900
600
300









I
m i
T|T r
m& 1
•|:v /% / ^
* ///'' &//
«> my
<;::#:• .»:s; ^
#^£ fc:N
4.2










/
:
•


/*•
*
• •
*^
V
r »
<*'
|
3.9
3.6
3.3
3.0
2.7
o> 2.4
E 2.1
1.8
1.5
1.2
0.9
0.6
0.3
0.42
0.39

9








7
,
'
,
f 4
t *
1 /
/«
/>
r /
'/
^/
%



V



»•

0.36
0.33
0.30
0.27
- 0.24
§» 0.21
0.18
0.15
I rt ^ o
»
^ 0.09
| 0.06
V
J::i 0.03
lg






















*
r
^
•
/
Total P Langmuir Langmuir










•
*
•
«




7
'',
/
t
t
/
t
f
/
/
1
»
52
48
44
40
36
1 32
\ 28
/
J 24
\ 20
^
i 16
12
j J







1*






i
!
i
!









1
EPC P desorption
Adsorption Adsorption
Maximum Energy
                                                                         Soils
                                                                         Soil Clay Fractions (n=7)
                                                                         Suspended Sediment
                                                                         (n=13)
                                                                      23  Bottom Sediment (n=5)

                                                                      T  Standard Deviation
Figure 3. Adsorption/desorption characteristics of Maumee River Basin soils,
their clay fractions and suspended and bottom sediments.

-------
                                  -38-
     In general, Langmuir adsorption maxima for the composite suspended
sediment samples were higher than those for the Maumee River Basin surface
soils and their clay fractions, but lower than the adsorption maxima reported
for bottom sediments.  This implies that sediments entering the Maumee River
system are altered by weathering and/or adsorption, either during transport
or after deposition, in such a manner as to markedly increase adsorptive
capacity.  McCallister and Logan (1978) found that the content of amorphous
Fe, Al, and Si in bottom sediments correlated most strongly with adsorptive
capacity.  Although the same mechanisms may be responsible for increasing
adsorptive capacity of suspended sediments in this study,  no attempt was
made to verify it.  Comparison between adsorption maxima and total P shows
that all composite suspended-sediment samples contained more P than the quantity
of P retained when the adsorption maximum is reached,  except for composite
suspended-sediment samples collected at sites 5, 3 and 8 (composite samples),
respectively.  Some of the increase in total P of the suspended sediment is a
result of selective erosion and transport of clay-sized particles which contain
a higher P concentration than silt and sand sized particles.  However, the P
content of the clay fraction of the Basin soils (Figure 3) is not high enough
to account for all of the increase on the basis of clay enrichment of sediment
alone.  It is likely that during runoff and transport, there is some adsorption
of P by the sediment.  Some of the high total P for Maumee River samples
(site 8) in the summer may be due to concentration of P by algae (all samples
were examined microscopically for presence of algae), which were detected in
these samples.  The low adsorption maximum for site 8 (composite sediment) is
probably due to the high calcite content and the presence of algae.  Approx-
imately 33% of the sample is calcite.  All samples, however, exhibited an
increase in total P above those of the clay fraction of basin soils, even those
sediments that did not contain algae.  Other researchers (Shukla et al , 1971;
Williams et al , 1971; McCallister and Logan, 1978) have reported an inverse
relationship between adsorption maxima and carbonate content,which agrees
with the relationship revealed with Maumee River samples collected at Site 8
but conflicts with the relationship revealed with the Tiffin River samples
that  show an increase in adsorption maxima with an increase in calcite content.
Therefore, this study does not conclusively demonstrate the relationship
between carbonate content and P adsorption maxima.

     Adsorption energy (the energy with which P in held to the adsorbent surface)
of each composite suspended-sediment sample is given in Table 2.  All samples
had lower adsorption energies than those of the Maumee River Basin surface
soils, their clay fractions, and bottom sediments from the Maumee River drainage
system.  This indicates that suspended sediments hold P less tenaciously than
surface soils, clays ,or bottom sediments.  In this study, the lowest adsorption
bonding energies were more frequently related to high total P rather than to
high calcite content.

     The equilibrium P concentration (P concentration at which adsorption
and desorption is equal and is determined at the P concentration where x/m = 0)
for each composite suspended sediment sample is given in Table 2.  Equilibrium P
concentrations reported in this study were in the same range (0.03-0.2 ug/ml)
as the EPC range  (0.006-0.24 ug/ml) for surface soils, clays, and bottom
sediments of the Maumee River drainage system (McCallister and Logan, 1978)

-------
                                   -39-


 except for the sample collected at Site 3 and the two summer samples at site 8
 (composite samples), respectively,which show quite high (0.58,  0.65 and 1.01
 ug/ml, respectively) EPC.  The two Maumee River (Site 8)  samples  were quite
 high (13.3 and 33.3%) in calcite whereas the Tiffin River tributary (Site 3)
 sample was low in calcite (3.6%),which indicates no clear relationship be-
 tween calcite content and high EPC.   The three highest total P  values correspond
 with the three highest EPC values,which indicates that high EPC is a function
 of total P rather than calcite content.

      Results  of each 10-cycle  desorption are given in Table 2.   The three
 samples collected at Sites 3 and 8 (composite samples),which exhibited high
 EPC and low adsorption energy  yielded the highest total P desorbed (93.8,
 62.4,  104.8 ug/g, respectively).  Conversely,  those samples with  low EPC
 desorbed the  least P.  This is consistent with theory since adsorption energy,
 EPC,  and total P desorbed are  all  related to the stability of the bonds formed
 between phosphate and sediment surfaces.

 Correlations  Between Calcite Content  and P Sorption-Desorption  Parameters

      Calcite  content was  positively correlated (Table 3),with total P of
 suspended  sediment indicating  that calcite contributed to the adsorption  of
 P  by  suspended sediment.

      Calcite  content was  positively correlated with EPC and total  P desorbed,
 and negatively correlated with adsorption energy parameter.   All  three cor-
 relations  imply that whereas P content may increase with  an increase in
 calcite content,  P adsorbed  by calcite  is not  held  very strongly  and can  be
 easily desorbed.

           Correlations Among P Sorption-Desorption Parameters

     Total  P was  negatively  correlated with  the  adsorption  energy, which means
 that as  the adsorption of P  increases, the  energy holding  the P to  the
 adsorbent decreases.   Total  P was  also positively correlated  with  the  total  P
 desorbed and EPC   that  is, as  the  amount  of  P  held  by  the absorbent  increases,
 the quantity of P which can  be  desorbed increases.  Similar correlations were
 observed by McCallister and  Logan  (1978)  for Maumee River Basin soils and
 bottom  sediments.

     Adsorption maximum was  negatively correlated with adsorption energy,
which indicates that  as the  capacity  to adsorb P increases, adsorption energy
decreases.  Adsorption maximum was not positively correlated with EPC and
 total P desorbed, which is due to the method of measuring these parameters.
Total P desorbed  and EPC are determined on sediments with naturally adsorbed P,
whereas adsorption maxima are obtained from the Langmuir isotherm,which was
developed by using P concentrations far above those naturally present in streams.

     Adsorption energy was negatively correlated with total P desorbed and
EPC, which is consistent with theory which predicts that as adsorption energy
increases,  P will be held more  tenaciously and be  less  likely to  be  released.

-------
Table 3.  Correlation coefficients among phosphate parameters and calcite content from the stream-
          composited sediment samples (n = 13).
                   Total P
N. S.  not significant

**  significant at 1% level

*  significant at 5% level
Adsorption maximum     Adsorption energy    Total desorbed    EPC
CalcUe 0.67*1* N. S.
content
Total P N. S.
Adsorption
maximum
Adsorption
energy
Total
desorbed
-0.559* 0.759** 0.888 **

-0.698** 0.872** 0.867 **
-0.663** W. S. N. S.

-0.79^** -0.71^ **'

0.950 **


-------
                                  -41-
Total P desorbed was positively correlated with EPC, which means that, as the
capacity to release more P increases, a correspondingly higher EPC will result.

Applications of the P Sorption-Desorption Study to the Maumee River System

     Relationships between calcite content and total P, and between calcite
content and the bonding energy parameters (adsorption energy, EPC, and total P
desorbed), can be meaningfully applied to the Maumee River system.  First, when
precipitation of secondary calcite occurs in the Maumee River, P adsorption
capacity of the suspended sediment increases,since the quantity of total P
that can be adsorbed increases. The result of this process is a reduction of
the P available to algae and aquatic plants, retarding the rate of eutrophi-
cation.  This is particularly significant when one considers that precipita-
tion of calcite is favored by any process which results in reduction of CC>2
concentration in the fluvial media.  During periods of rapid algae growth,
CC>2 concentrations of the fluvial media are significantly reduced due to photo-
synthesis.  Thus algal growth contributed to the precipitation of calcite,
which will compete with aquatic plants for soluble P.  Consequently,even though
P adsorbed by calcite is not strongly bonded, as evidenced by the low adsorp-
tion energy, high EPC ,and P desorption values, and calcite will be the first
component to release P back into solution at low soluble P levels, calcite may
be an important factor in regulating soluble P concentrations in the stream.
In summary, calcite does contribute to P adsorption capacity of suspended
sediments, and, even though P sorption-desorption by calcite is quite dynamic,
it may significantly contribute to the role of sediment in regulating soluble
P concentrations in streams.
3.25  Conclusions

     Adsorption maxima for composite suspended samples were higher than those
for the Maumee River Basin surface soils and their clay fractions but lower
than adsorption maxima for bottom sediments from the Maumee River drainage
system as reported by McCallister and Logan (1978).  This implies that sediments
entering the Maumee River system are altered by weathering and/or adsorption,
either during transport or after deposition in such a manner as to markedly
increase the adsorptive capacity.  Composite suspended sediments generally
contained more total P than the quantity of P retained when the adsorption
maximum is reached.  This is a result of P enrichment of sediment because of
selective erosion of clay and also adsorption of soluble P by sediment
during transport.  All composite suspended-sediment samples had lower adsorp-
tion energies than the Maumee River Basin surface soils and their clay fractions
or bottom sediments from the Maumee River drainage system.  Equilibrium P
concentrations (EPC) were in the same range as those exhibited by surface soils,
clays, and bottom sediments of the Maumee River drainage system except for
samples with the highest total P content.  The three samples exhibiting
high EPC and low adsorption energy yielded the highest total P desorbed.

     Calcite content was  significantly positively correlated with total P,
EPC,  and total P desorbed and significantly negatively correlated with the
adsorption energy,which implies  that calcite contributes  to P adsorption

-------
                                 -42-
 capacity  of  suspended  sediments; however,  P  adsorbed  by calcite  is  quite
 labile.   Total  P was significantly  correlated with  the  bonding energy para-
 meters  (adsorption  energy,  EPC,  and total  P  desorbed),  which  indicates that
 as the  amount of P  held  by  the  absorbent increases, the quantity of P in
 solution  at  equilibrium  and the quantity of  P which can be  desorbed increases.
 Total P desorbed was significantly  correlated with  EPC, which indicates that
 EPC could be used to effectively estimate  the amount  of P which  can be desorbed.


 3.26 Literature Cited

 1.  Chen, Yi Shon,  James N. Butler, and Werner  Stumm.   1973.  Kinetic Studies
     of  Phosphate Reactions  with Aluminum Oxide  and  Kaolinite. Environ. Sci.
     Technol. 7:327-332.

 2.  Cole, C. V., S. R. Olsen,  and  C.  0. Scott.   1953.   The  Nature of Phosphate
     Sorption by Calcium  Carbonate.   Soil Sci. Soc.  Amer. Proc. 17:352-356.

 3.  Dreimanis,  A.   1962. Quantitative Gasometric Determination  of Calcite
     and Dolomite by Using Chittick Apparatus.   J. Sed.  Petrol.  32:520-529.

 4.  Hsu,  Pa  Ho. 1964.  Adsorption of Phosphate by  Aluminum and  Iron in
     Soils.   Soil Sci.  Soc.  Amer. Proc. 28:474-478.

 5.  John, Matt  K.   1970. Colorimetric Determination  of Phosphorus in Soil
     and Plant Materials  with Ascorbic Acid.   Soil Science.  109:214-220.

 6.  McCallister, D. L.,  and T.  J.  Logan.   1978.  Phosphate  Adsorption-Desorption
     Characteristics of Soils and Bottom Sediments in  the Maumee  River Basin
     of  Ohio. J. Environ. Qual. 7:87-92.

 7.  Murphy,  J., and J. P. Riley.  1962. A Modified Single-Solution Method
     for the  Determination of Phosphate in  Natural Waters.   Anal. Chim.
     Acta. 27:31-36.

 8.  Olsen,  S. R.,  and  F. S. Watanabe.  1957. A Method to Determine a Phosphorus
     Adsorption  Maximum of Soils as Measured by  the  Langmuir Isotherm.  Soil
     Sci.  Soc. Amer. Proc. 21:144-149.

 9.  Shukla,  S.  S.,  J.  K. Syers, J.  D. H. Williams,  D. E. Armstrong, and R.  F.
     Harris.   1971.   Sorption of Inorganic  Phosphate by Lake Sediments.  Soil
     Sci.  Soc. Amer. Proc. 35:224-249.

10.  Sommers, L. E., and  D.  W.  Nelson.  1972.  Determination of  Total Phosphorus
     in  Soils:  A Rapid Perchloric  Acid Digestion Procedure.  Soil Sci. Soc.
     Amer. Proc. 36:902-904.

11.  Syers, J. K.,  R.  F.  Harris, and D. E.  Armstrong.   1973.  Phosphate Chemistry
     in  Lake  Sediments.  J.  Environ. Quality, 2:1-4.

-------
                                 -43-


12.  Taylor, A. W., and H. M. Kunishi.  1971.  Phosphate Equilibria on Stream
     Sediment and Soil in a Watershed Draining an Agricultural Region.  Jour.
     Agr. Food Chem. 19:827-831.

13.  Wall, G. J., and L.  P. Wilding.  1976.  Mineralogy and Related Parameters
     of Fluvial Suspended Sediments in Northwestern Ohio.  J. Environ. Qual.
     5:168-173.

14.  Williams, J. D. H.,  J. K. Syers, R. F. Harris, and D. E. Armstrong.  1971.
     Fractionation of Inorganic Phosphate in Calcareous Lake Sediments.  Soil
     Sci. Soc. Amer. Proc. 35:250-255.

15.  Woodruff, J. R., and E.  J. Kamprath.  1965.   Phosphorus Adsorption Maximum
     as Measured by the Langmuir Isotherm and Its Relationship to Phosphorus
     Availability.  Soil  Sci. Soc.  Amer. Proc. 29:148-150.

-------
                                 -44-
3.3  Clay mineralogical, physical and chemical relationships between
     watershed soils and runoff sediments
3.31  Abstract

     Runoff sediments collected at sampling stations on five small watersheds
were characterized for clay mineralogy, particle-size distribution,and iron
oxide content.  Values for these parameters were used to determine the
relationship between the watershed soils and subsequent runoff sediments.
     Soil-clay mineralogy of all surface horizon samples averaged:  illite
vermiculite 23%, quartz 15% and expandables lU%.   Comparable data for runoff
sediment clay were: illite 55%, vermiculite 17%,  expandables 19% and quartz 10%.
Higher concentrations of illite and expandables,coupled with lower   concen-
trations of vermiculite and quartz in the runoff sediment than occurred in the
watershed soils,were attributed to preferential transport of fine clays from
the soil surface.

     The total clay content of the watershed soils equalled kl.Q% and contained
ill.3% fine clay  ( £ 0.2 urn) compared to 60.2% total clay and 51.W fine clay
in the runoff sediment.  Cation exchange capacity of the soil-clay fraction
increased from Hi meq/100 gms. to h6 meq/100 gms. in the runoff sediment clays
due the increased content of  fine clays.  A decrease in both amorphous iron
and free iron oxide contents  in runoff sediment clays was unexplained; however,
preferential removal of fine  clay materials may be responsible.

-------
                                  -45-
 3.32  Introduction

      Utilization of clay mineralogical data obtained from fluvial transported
 sediments to infer source areas can lead to erroneous conclusions.  Clay
 mineralogy of suspended and "bottom sediments from downstream sites may vary
 considerably from sediments obtained nearer the source area due to physical
 sorting and flocculation during transport.

      Clay mineralogical studies relating source areas and runoff sediments
 collected within the watershed are scarce.   Significant background information
 has been provided by several workers (Lund et al, 1972; Wall and Wilding,  1976;
 Jones et al, 1977).   Wall and Wilding (1976) compared clay mineralogical data
 from suspended sediments and soil profiles  in an attempt to identify source
 areas.   Their results were negative as clay mineralogy of suspended sediments
 appeared to be a function of preferential transport,  either from the soil
 surface or within the stream system.  Jones et al (1977) monitored sediment
 and solution parameters in primary drainage ditches  adjacent to small water-
 sheds.   Clay mineralogy data obtained from suspended sediments  in the ditches
 differed from that of the soils.   Both soil and suspended-sediment mineralogy
 was predominantly illitic; however, quartz  was the second most  abundant  mineral
 in the  sediments whereas expandable minerals were the second most abundant
 group   in the soils.   Wall and Wilding (1976) obtained similar results  from
 suspended sediments  collected farther downstream and listed three possible
 explanations:   l) preferential erosion of coarse clay from soil surfaces,
 which contain higher percentages  of quartz  than do fine clays;   2)  preferential
 erosion of quartz from the soil;  or  3)  a concentration of quartz relative to
 the clay minerals during fluvial  transport.

      Lund et al (1972)  found higher concentrations of montmorillonite  and
 vermiculite in reservoir bottom sediments than occurred in surrounding soils.
 Mica and quartz contents were approximately equal and less than corresponding
 amounts  in the soils.   Lower amounts of  quartz in the bottom sediments were
 attributed to  sedimentation  during transport, and vermiculite contents  in
 the sediments  were accounted for by source-area characteristics.  Higher
 concentrations  of montmorillonite  were interpreted as:   l)  fine  clay
 enrichment in  the sediments;   2) transport  of truncated B-horizon soils high
 in montmorillonite;  or  3)  easily dispersed montmorillonitic  soil  clays.
     ^    objective of the present study was to further explain some of the
previous results by investigating the possibility of preferential transport
of clay minerals from small watersheds, by analyzing runoff sediments collected
at the point of entry into the stream system.


3.33  Methods and  Materials

                        Watershed Descriptions

     Runoff sediment monitoring stations were constructed on five different
watersheds, four in Defiance County, Ohio and one in Wood County, Ohio
(Figure 1  , page  4).  Selection of a particular watershed was based on the
existence of one predominant soil series in a characteristic landscape setting,
•which occurred extensively throughout the river basin.   The soils selected for
the study included:   Roselms silty clay loam (Aerie Ochraqualf, very fine,
mixed, mesic), Roselms silty clay loam (Aerie Ochraqualf, very fine, illitic
mesic) Blount loam (Aerie Ochraqualf, fine, illitic, mesic), Paulding clay
(Typic Haplaquept,  fine, illitic, non acid, mesic)  Hoytville clay loam (Mollic

-------
                                  -46-
Ochraqualf, fine, illitic,  mesic).   Watershed sizes  ranged  from 0.32  to  3.2
hectares.  For the duration of the study,  soybeans were grown on the  watershed
following fall plowing.


                           Sample Collection

     Runoff sediment samples originating from surface horizons were obtained
by installing a Coshocton Wheel sampling device in  sediment drop boxes at  the
mouth of each watershed.   Complete details of the monitoring system are  given
in Volume 1 of this report.  Representative soil samples were obtained  from
the A-horizon in each watershed by collecting an average of 15 samples using
a radial fan grid method  and compositing to form a  single reference sample.
Sampling intervals varied between slope components  and soils; however, a minimum
of 3 samples were collected in each delineation.


                          Analytical Methods

     Sediment concentration (mg/1) was determined for runoff sediment samples
collected at each watershed by filtering a 50 ml aliquot of the runoff using
1.0 urn Nucleopore membrane.  Prior to analyses, runoff sediments were concen-
trated by flocculation with l.ON MgCl2, washed with methano1,and dried at
40°C.  Particle-size distribution was determined by the pipet   method of
Steele and Bradfield (1934) after dispersing with a Sonifier Cell Disrupter.
Samples utilized for percent fine clay were electrolytically dispersed and
fractionated according to the centrifugation procedure of Jackson (1956).
Clay-size fractions ( < 2 urn) used in all other analyses were collected  by
Bonification and standard sedimentation procedures  (Rutledge et al, 1967).

     X-ray diffraction analysis utilized a Norelco diffractometer and copper
radiation with a nickel K-beta filter, a 0.006-inch receiving slit, a 1° diver-
gence slit»and a proportional counter.  Sample preparation involved vacuum plating
0.15 grams of Mg-saturated, methano1-washed clay (  < 2 um)  on a porous ceramic
plate (Kinter and Diamond, 1956).  Two plates were prepared  for each  sample;
one was saturated with 10% ethylene glycol and scanned whereas the other was
scanned after air drying, heating to 400°C for two hours, and heating to
550°C for two hours.  Runoff sediment clays were scanned from 2° to 32° 26
at a speed of 2° 29 per minute.  Soil clays were scanned from 2° to 32° 26
at a speed of 1° 26 per minute.  Peak areas were delineated on the basis of
symmetry and average baseline heights.  Photocopies of original peaks were
cut out and weighed.  Relative percentages of clay minerals  in each sanple
were determined  by a modified Johns et al  (1954) method.

     Cation  exchange capacities of the total clay fractions were determined
by the sodium saturation method of Chapman  (1965) after removal of organic
matter with  30%  I^On*  Amorphous iron and  free  iron oxide contents of the
total clay  fraction were obtained by atomic absorption  following 1^2 treatment
and  extraction with 0.2 M  ammonium oxalate  (McKeague and Day,  1966) and
citrate-bicarbonate-dithionite  (Mehra and Jackson, 1960), respectively.

-------
                                  -47-


 3.3H  Results and Discussion

                            Clay Mineralogy

           The clay mineralogy data from composited A-horizon  soils  and
 runoff sediments  are  listed in Table  1 .   In most instances,  little
 difference is noted in relative percentages "between the watershed soils.
 Most notable  is the expandable mineral content  listed  for the Paulding  and
 Hoytville soils.   Additionally, the Blount soil contains somewhat higher
 concentrations of vermiculite and lower concentrations of illite, perhaps
 indicating an advanced weathering state.   Based on the mean values, the clay
 mineralogy of all sites follows the order:  illite > vermiculite >  quartz >
 expandables.   The order observed for runoff sediments  is:  illite > expandables >
 vermiculite > quartz.   The  differences  between  these two orders coupled with
 individual runoff and soil  data indicate the existence of a definite trend.
 As  depicted in Figure  1, expandable minerals in the runoff sediment exceed
 the amount present in the soil.   The greatest difference between the two sources
 occurs  at the Paulding site.   A slight  decrease in expandable minerals  is noted
 in  the  Hoytville  runoff sediment  relative  to the soil  clay; however, C-horizon
 materials excavated by tiling operations were left on the surface.  Reference
 soil samples  were collected in an adjacent area away from this influence }but
 runoff  sediments  may  reflect  contributions from lower horizons^hich contain
 less expandable minerals (P.  E. Rhoton, 1978.   Clay mineralogical relationships
 between watershed soils, runoff,and bottom sediments in the Maumee River Basin,
 Ohio.   Ph.D.  Thesis.   Ohio  State  Univ., Columbus).  Illite percentages  in the
 soil and  runoff sediment are  closer to  a 1:1 relationship than any of the other
 mineral species but reveal  a  trend which is analogous to the  one observed
 for expandables.   The  content  of  vermiculite in  the runoff sediment is  lower
 than that for the corresponding soils.  Quartz  varies between runoff sediments
 and soil  in a manner  similar  to vermiculite, lower concentrations in the
 runoff  sediment.

      The  differences between  soil  and runoff sediment mineralogy are not sub-
 stantial  in some  of the watersheds  considering the accuracy of quantitative
 mineralogy by X-ray diffraction procedures; however, identical trends exist
 for  all watersheds.  Therefore, it  is suggested  that the data prove conclu-
 sively  that illite and expandable minerals are preferentially transported
 from soil  surfaces, leading to an  enrichment of  these minerals in runoff
 sediments  and a decrease in the percentages of quartz and vermiculite.

      The previous  data suggest that clay mineral abundance  in the soil is
 the primary factor controlling runoff sediment mineralogy ,as  evidenced by
 the  illitic nature of both the soil and runoff sediment clay.   It is further
 suggested that observed differences in the mineral ratios between soil and
 runoff  sediment clays  are related to particle-size differences.   Based on
 the^data presented in  Figure 1  , the relative  erodibilities  are:   expandables  >
 illite  > vermiculite > quartz.  The size ranges  listed for  various  clay
minerals include:   montmorillonite  (expandables) 0.01-0.1 urn  (Zelazny and
 Calhoun, 1971), illite 0.1-0.3 urn  (Grim, 1968),  vermiculite 0.2-2.0  urn
 (Douglas,  1977),  quartz 0.2-2.0 urn (Wilding et al,  1977).   These  size  ranges
 correspond directly to the  observed relative  erodibilities.

     The manner in which the clay minerals  are removed from the soil surface
is unknown and cannot  be determined from the mineralogical  data.  The
sediment concentration parameter was selected  as a possible means of
determining if transport occurs predominantly  as discrete particles  or  as

-------
                                     -48-
Table  1  .   Clay Mineralogy of the total clay fraction from surface soils  and
            corresponding runoff sediments*
Watershed
Rose 1ms 1
Roselms 2
Blount
Paulding
Hoytville
Mean
Expandables
Soil Sediment
13
11
14
8
23
14
14
16
25
19
21
19
	 - 	 7d
Vermiculite
Soil
20
22
31
22
21
23
Sediment
17
18
22
16
11
17
r 	
Illite
Soil
57
44
41
53
45
48
Sediment
60
53
41
60
59
55
Quartz
Soil Sediment
11
23
14
18
11
15
9
13
12
5
9
10
*  Runoff sediment mineralogy values listed represent the mean of 13, 16, 9,
   15 ,and 27 samples for Roselms 1, Roselms 2, Blount, Paulding,and Hoytville,
   respectively.

j- Percent  of  the  sum  of  the  four  clay-mineral  fractions.

-------
     Ratio

OQ
c
(B
h- '
•
ft
H-
O
o
H>
i-S
£
O
l-h
CO
fl>
a
H-
3
(B
3
rt
O
«>
O
H-
1
O
I-1
Vrf
r minera
o
HI
o
H-
3
H-
<
H-
g
!-•
waters
(B
a
en




O O O O t
to -P-  cy, 00 O 10
2
"•— - ^,
to
a
a1


1

i -
1-1
H-
•'-J (B













-6V-

-------
                                 -50-
aggregates.   It was assumed that clay mineralogy and sediment concentration
fluctuated with rainfall intensity.  Clay mineralogy was expected to be a
function of rainfall intensity as it varies with particle size.   Mean-particle
size and sediment concentration should "be positively correlated with rainfall
intensity.  Furthermore, expandables and illite should be negatively correlated
with sediment concentration if transport occurs as individual particles,as
these minerals are most concentrated in the fine clay fraction.   Positive
correlations between illite and expandables and sediment concentration
could suggest transport in an aggregated form.  Vermiculite and quartz
transport as individual particles should result in positive correlations
with sediment concentration,indicating removal of the coarse clay fraction.

     Some of the clay-mineral variations were highly correlated with sediment
concentration (Table 2 ),but absolute differences between high and low
sediment concentrations were small.  This suggests the concentration of clay
minerals in the runoff sediment did not vary substantially.  However, sediment
concentrations used were biased toward high values»and no clay mineralogy
was determined at very low sediment concentrations.   A more thorough investiga-
tion is required for verification of the results.

     Particle-size analyses indicated that runoff sediments contained a
higher total clay concentration, 60.2% as opposed to ki.8% (Table 3 )
for the watershed soils.  Additionally, an average of 51-W of the total
clay fraction in the runoff sediment was fine clay ( <£. 0.2 urn) »whereas the
soil-clay fraction contained kl.3% fine clay.  These figures indicate
preferential removal of fine materials from soil surfaces and provide strong
support for the explanations given for the differences in clay mineralogy
between the watershed soils and runoff sediments.  In addition to clay
mineralogy,  the fine clay influence on runoff sediments is evident in the
recorded cation exchange capacities (Table  3 ).  The average cation exchange
capacity of the total clay fraction of the soils and runoff sediments was
hi meq/100 gms. and h6 meq/100 gms., respectively.

     Amorphous iron and total free iron oxide analyses were conducted on
the clay fractions ( < 2 urn) from the soils and runoff sediments to determine
how possible concentration differences between the two sources were related
to preferential soil transport.  Results from the analyses (Table 4 )
indicated that amorphous iron decreased from an average of 5-6 mg/g in the
surface soils to 3.U mg/g in the runoff sediment.  Free iron oxides decreased
from an average of 2k.5 mg/g in the watershed soils to 18.9 mg/g in the
runoff sediment.  Sample pretreatment with ELpC^ slightly increased the amount
of iron extracted in each instance; however  the increase did not affect the
observed trends.  The distribution of iron oxides between fine and coarse
clays  and discrete particles and clay coatings was unknown.  These materials
may occur as clay coatings involved in aggregate stabilization and less
stable aggregates containing lower iron oxide contents are dispersed and
removed; however, due to a lack of such information, the variation between
soil and runoff sediment clays is unexplained.

3-35  Conclusions

     The differences observed between watershed soils and their runoff
sediments consisting of increases in expandable minerals and illite in
addition to decreases in vermiculite and quartz in the sediment are due to
preferential transport of fine clays from the soil surface.  Higher con-
centrations of fine clay in the runoff sediment is also responsible for

-------
                                  -51-
Table   2 .  Correlation Coefficients of runoff sediment mineralogy as a
            function of sediment concentration 1"
Watershed
Roselms 1
Roselms 2
Blount
Paulding
Hoytville
N
12
15
9
14
22
Expandables
(-)ns
(-).453
ns
.456
ns
Vermiculite
(-)ns
ns
ns
(-)ns
(-).543**
Illite
ns
ns
(-).725*
(-).380
ns
Quartz
ns
ns
ns
(-)ns
ns
t  Unmarked values are significant at the 20% level;  "*" indicates signifi-
   cance at the 5% level and "**" indicates significance at the 17» level;
   ns indicates not significant at the 20% level;  (-)  indicates negative
   correlation.

-------
                                      -52-
Table  3 .   Percent total clay,  fine clay,and  cation exchange  capacity  of  surface
            soils and runoff sediments*
                     Total Clay
Fine Clay
CEC dneq/lOOgms)
Watershed
Rose 1ms 1
Roselms 2
Blount
Paulding
Hoytville
Mean
Soil
52.8
34.6
27.0
52.7
42.3
41.8
Sediment
63.8
55.8
58.2
70.0
53.3
60.2
Soil
39.1
34.3
38.6
38.1
56.6
41.3
Sediment
51.3
45.5
50.8
53.0
56.2
51.4
Soil
37
39
41
41
46
41
Sediment
40
52
43
46
47
46
*  Percent fine clay and cation exchange capacity values listed for runoff sediments
   represent the mean of 10 samples from each watershed.  Percent total clay in
   the runoff sediments represent the mean of 13, 18, 9, 15,and 38 samples for
   Roselms 1, Roselms 2, Blount, Paulding,and Hoytville, respectively.

-------
                                  -53-
 Table  4 .   Amorphous  iron and  free  iron oxide  content  (mg/g  total  clay)  of
             surface soils  and runoff sediments*
                        Amorphous  Iron
Watershed
Rose 1ms 1
Roselms 2
Blount
Paulding
Hoytville
Mean
Soil
4.4
6.2
6.8
6.3
4.2
5.6
Sediment
2.4
3.1
3.7
4.5
3.1
3.4
Soil
-' i M
23.9
25.0
31.2
21.4
21.0
24.5
i- •»- vx*.4 *~r*\ J-\JL t
S ed imen t
17.7
17.2
22.4
16.7
20.3
18.9
*  Amorphous iron and free iron oxide values listed for runoff sediments
   represent the mean of 12,  15, 9,  15,and 24 samples for Roselms 1,  Roselms  2,
   Blount, Paulding,and Hoytville,  respectively.

-------
                                 -54-
 increased  cation exchange  capacities  and may explain the lower concentrations
 of  iron oxides  in runoff sediment  clays.   Lower  concentrations of iron
 oxides  in  runoff sediment  clays  could partially  account for the increased
 cation  exchange capacity if present as clay coatings on soil clays.


 3.36  Literature Cited

 1.  Chapman,  H.  D.   1965.   Cation  Exchange Capacity.  p. 891-901.  In C.  A.
    Black  (ed.)  Methods of Soil  Analysis, Part 2.   American Society of Agronomy,
    Madison.

 2.  Douglas,  L.  A.   1977.   Vermiculites.   p. 259-292.   In J. B. Dixon and
    S.  B.  Weed  (ed.)  Minerals  in Soil Environments.   Soil Science Society
    of  America,  Madison.

 3.  Grim,  R.  E.   1968. Clay Mineralogy.   McGraw-Hill, New York.

 4.  Jackson,  M.  L.   1956.   Soil  Chemical Analysis  - Advanced Course.  Published
    by  the author.   Dept.  of Soils, Univ. of Wis., Madison.

 5.  Johns, W. D., R.  E. Grim,  and  W.  F. Bradley.  1954.  Quantitative
    Estimations of Clay Minerals by Diffraction Methods.  J. Sediment
    Petrol. 24:242-251.

 6.  Jones, L. A., N.  E.  Smeck, and L. P.  Wilding.   1977.  Quality of Water
    Discharged  from Three  Small  Agronomic Watersheds in the Maumee River
    Basin.  J.  Environ. Qual.  6:296-302.

 7.  Kinter, E.  B.,  and S.  Diamond.  1956.  A New Method for Preparation
    and Treatment of Oriented Specimens of Soil Clays for X-ray Diffraction
    Analysis.  Soil Sci.  81:111-120.

 8.  Lund,  L.  J., J. Kohnke, and  M. Paulet.  1972.   An Interpretation of
    Reservoir Sedimentation II.   Clay Mineralogy.  J. Environ. Qual. 1:303-307.

 9.  McKeague, J. A.,  and  J. H. Day.   1966.  Dithionite and Oxalate Extractable
    Iron and Aluminum as  Aids in Differentiating Various Classes of Soils.
    Can. J. Soil Sci. 46:13-22.

10.  Mehra, 0. P., and M.  L. Jackson.   1960.  Iron Oxide Removal from Soils
    and Clays by a Dithionite-Citrate System Buffered with Sodium Bicarbonate.
     Clays Clay  Miner. 7:317-327.

11.  Rutledge, E. M.,  L.  P. Wilding,  and M. Elfield.  1967.  Automated Particle-
     Size Separation by Sedimentation.  Soil Sci. Soc. Amer. Proc. 31:287-288.

12.   Steele, J.  G., and R.  Bradfield.   1934.  The Significance of Size
     Distribution in the Clay Fraction.  Amer. Soil Survey Assoc. Bull.
     15:88-93.

13.  Wall,  G.  J., and L.  P. Wilding.   1976.  Mineralogy and Related Parameters
     of Fluvial Suspended Sediments in Northwestern Ohio.  J. Environ. Qual.
     5:108-173.

-------
                                  -55-
14.  Wilding, L. P., N. E. Smeck, and L. R. Drees.  1977.  Silica in Soils:
     Quartz, Cristobalite, Tridymite, and Opal.  p. 471-552.  In J. B. Dixon
     and S. B. Weed (ed.) Minerals in Soil Environments.  Soil Science
     Society of America, Madison.

15.  Zelazny, L. W., and F. G. Calhoun.  1971.  Mineralogy and Associated
     Properties of Tropical and Temperate Soils in the Western Hemisphere.
     Soil and Crop Sci. Soc. of Fla.  31:179-189.

-------
                                  -56-
3.4  Coagulation and dispersion of Mautnee River  Basin soils  and particle-size
    distribution of spile and sediments
3.41  Abstract

     Five agricultural soils from the Maumee River  Basin were  dispersed  by
Bonification in water, septic tank effluent, secondary treated wastewater,
Maumee River water,and drainage ditch water.  In all  media  except  distilled
water, colloids coagulated rapidly while they remained dispersed  for weeks
in distilled water.  Of the natural media,septic tank effluent gave  the  least
coagula tion,while drainage ditch water gave the most.   This was attributed to
the higher concentration of sodium in septic tank effluent  and higher  calcium
and magnesium concentrations in the ditch water.

     The clay fraction  of Paulding soil was dispersed by sonification in
Maumee River water at different sediment concentrations. Coagulation  rate
increased with increasing sediment concentration.  Clay dispersed  in Maumee
River water.diluted to 0.1, 0.2, 0.25,and 0.5 strength with distilled  water,
indicated that a threshold concentration of 50 ug/ml  of (Ca +  Mg)  was  needed
for clay coagulation at a sediment concentration of 500 ug/ml.

     Dispersion ratio ofwatershed soils showed that much of the  fine clay and
total clay were aggregated into silt and sand-sized particles.

     Runoff sediment contained more clay and less silt and  sand  than the original
soil.  Sediments in tile lines were very high in clay.

-------
                                   -57-
3.42  Introduction

     Maumee River Basin soils are, with the exception of the old glacial beach
sands, very high in clay.  Clay particles once detached are more easily trans-
ported than coarser silt and sand particles.  During  initial  erosion and runoff
transport, coarser particles tend to settle in the field at points where the
slope decreases and water velocity is reduced.  As a result, runoff sediment
is higher in clay than the original surface soil from which it was eroded.
Sediments in tile lines are usually also high in clay.

     Under natural conditions, soil constituents do not behave as primary
particles; rather, they are flocculated or coagulated into larger particles.
Coagulation of primary particles, especially clays, occurs when sufficient
divalent cation (usually Ca^"*" or Mg^~*~) is present to compress the electrical
double layer surrounding primary charged particles and allow them to approach
close enough for binding forces to hold them together.  On the other hand,
monovalent cations (especially Na+) have the opposite effect and result in
clay dispersion.  In the soil, coagulated particles may be cemented together
by agents such as organic matter and iron and aluminum oxide coatings.

     The soils of the Maumee River Basin are derived from limestone till and
                                 o t       O J_
lacustrine sediments.  High in Ca + and Mg  , they are easily flocculated.
River water is also high enough in Ca   to promote coagulation of suspended
sediments.  Domestic sewage discharges, however, are high in Na+ and may
promote dispersion.

     The objectives of this study were to:
     1.  Determine the relative coagulation/dispersion of Basin soils in
         several fluvial media
     2.  Determine the effect of sediment and divalent cation concentrations
         on clay flocculation
     3.  Determine the extent to which Basin soils are flocculated
     4.  Determine clay enrichment of runoff sediment


3.43  Methods

              Soil dispersion/coagulation in natural media

     Surface soil samples were taken from the same monitored watersheds as
used in the other studies (Figure 1, page  4 ).   In addition,  Toledo clay
from Castalia, Ohio was included.

     Surface horizon samples of six different soils typical of the Maumee
Basin were collected, air-dried, and crushed to pass a 2 mm sieve.   The soils
were classified as Blount (Aerie Ocharqualf), Hoytville (Mollic Ochraqualf),
Lenawee (Mollic Haplaquept), Paulding (Typic Haplaquept), Roselms (Aerie
Ochraqualf), and Toledo (Mollic Haplaquept).  Particle-size distribution was
determined by the sedimentation-pipet method essentially as described by
Day (1965).   The samples were given no pretreatments prior to dispersion.
The soils were dispersed by ultrasonification with a Branson Model  W185
sonifier operated with an input power of approximately 125 watts for 15

-------
                                  -58-
minutes per sample.  An ice bath was used to keep the suspension temperature
at approximately 10°C.

     Organic carbon was determined by the dry-combustion method of Winters
and Smith (1929).  Soil pH was measured in a 1:1 slurry of soil and dis-
tilled water.  Exchangeable cations (Ca, Mg, K) were measured in a IN MfyOAc
(pH 7.0) extract with a Perkin-Elmer 303 Atomic Absorption Spectrometer.
Exchangeable acidity was obtained by Ba replacement procedure of Peech et.
al.  (1947).

     Mineralogy of the clay fraction (^,2 urn) was obtained with a Norelco
diffractometer equipped with a copper x-ray tube as specified by Smeck et al.
(1968).  Clays were Mg-saturated and given the following treatments:
ethylene glycolated, air dry, heated to 400°C for 2 hours, and heated to 550°C
for 2 hours.  There were no pretreatments to remove carbonates, organic matter,
or free oxides.  Amounts of the clay minerals present were semiquantitatively
estimated ($>%) by the method of Johns et al.  (1954), as modified by Rut-
ledge at al.  (1975).


                             Natural Media

     Samples of four natural media were obtained in October and November 1975
by a grab-sampling procedure.  River water was collected at Independence Dam
on the Maumee River near Defiance, Ohio (Figure 1, page  4 ); drainage ditch
water came  from an agricultural watershed dominated by Paulding soils in
Defiance County, Ohio.  Secondary-treated municipal wastewater was sampled
at the Jackson Pike Wastewater Treatment Plant, Columbus, Ohio.  Septic tank
effluent was collected from the septic tank at Don Scott Airfield, Columbus,
Ohio.  Approximately 380 liters of each medium were filtered through Whatman
No. 2 filter paper and stored at  4°C.

     Basic  cations in the media were determined by atomic absorption spectro-
scopy.  A gorning Model 12 Research pH meter was used to determine pH.
Electrical  conductivities were obtained with a conductivity bridge and cell.

     Each soil was dispersed in each of five different media:  the four natural
media plus  distilled water (originally chosen to simulate rainfall or snowmelt).
The suspensions had a concentration of approximately 25,000 mg/1 and were
dispersed by ultrasonification, according to the procedure given above.  This
allowed for a minimum of chemical alteration of the natural conditions and
was intended  to mechanically parallel dispersion by raindrop impact.  The
percent less  than 2 urn (effective diameter)  fraction was determined by the
sedimentation-pipet technique.

     A Paulding surface-horizon sample was similarly dispersed by ultra-
sonification, and the clay fraction was collected through several successive
steps of thorough stirring, sedimentation, and siphoning.  After concentrating
the clay suspension with filter candles, the approximate clay concentration
was determined by  filtering triplicate aliquots of the suspension through
micropore filters  (0.2 urn diameter pores) and weighing the residues.  The

-------
                                  -59-
day was dispersed in river water at suspended sediment concentrations of
0, 50, 100, 500, and 1000 mg/1.  Coagulation was observed after stirring with
an electric mixer, and it was recorded over time by photographing the sedimen-
tation bottles.  A similar series of photographs recorded the coagulation of
suspended sediments at a constant concentration of 500 mg/1 in river water,
which was diluted with distilled water to fractional strengths of 0.1, 0.2,
0.25, 0.3, and 0.5.  Temperature in both experiments was approximately 22°C.

     Photomicrographs of coagulates were made according to the following pro-
cedure:  Clays fractionated from the Paulding soil were suspended at several
concentrations in distilled water and in river water, and the suspensions were
mixed thoroughly with an electric mixer.  After coagulates had begun to settle,
the time elapsed since stirring was noted and a straight piece of glass tubing
(4 mm diameter) was dipped into each suspension to an arbitrary but standard
depth.  The tube containing suspension was removed from the container, and a
drop was discharged from the tip into the well of a hanging drop slide.  A
cover slip was placed on top, and the slide was examined under a petrographic
microscope at 250X or 500X.


            Soil aggregation and clay enrichment of runoff

     Surface soil samples  (0-15 cm) were taken from each mapping unit within
each of the monitored watersheds in Defiance  County (Figure 1, page 4) and
composited by area weight.  The Hoytville plots were also sampled and samples
composited.  Particle-size distribution was determined by the pipet method of
Steele  and Bradfield(1934) after dispersion by these methods:  total dispersion
in sodium hexameta-phosphate, sonification in distilled water as described
previously,and dispersion  by mild agitation in distilled water.  In all cases,
the soils were ground, screened to recover the <2 mm fraction^and dispersed
without chemical pretreatment to remove organic matter or iron oxides.  A
dispersion ratio was calculated for each soil according to the equation:

Dispersion Ratio = % soil  fraction (sand, silt or clay) by total dispersion
                   7» soil  fraction (sand, silt or clay) by water dispersion

     Sediment in runoff and tile drainage from the monitored watersheds and
plots were recovered by flocculating with IN BaCl^, washed in methanol and
dried at 40°C.  They were  then dispersed by sonification in distilled water
and particle-size distribution determined as described previously.  An
enrichment ratio was calculated using mean particle-size distribution of
samples 'analyzed for individual events during 1975-1976 and values for the
surface soils described above accoridng to the equation:

Enrichment Ratio = % runoff fraction by sonification
                   7o soil  fraction by sonification

-------
                                  -60-
3.44  Results

     Physical, chemical,and mineralogical properties  of the  surface horizons
of all six soils used in this study are given in Table 1.  These data  reveal
that the soils are moderate to high in both total clay content  (27-53% total
clay) ,and base saturation (59-88%).  The great majority of the  exchange sites
are occupied by Ca or Mg, which is primarily a reflection of the high  lime
content of the parent tills and lacustrine materials  of northwestern Ohio.
Consequently, pHs are near neutral (5.9-7.0).   The clay fraction is mostly
illite and vermiculite/chlorite,  with secondary amounts of quartz.  Such clay-
mineral suites are consistent with previous unpublished investigations.

     The natural media (Table 2 ) had alkaline pHs and, with the exception
of septic tank effluent, were roughly comparable in electrical  conductivity.
Sodium concentrations were highly variable (probably reflecting the influence
of commercial detergents), but the other cation concentrations  were always  of
the order Ca>Mg>K.  The Ca + Mg concentrations were about  100-130 mg/1.
Comparison of these data with other (U.S. Dept. Interior Geological Survey,
1975) data for the respective media indicate that these values  are approxi-
mately median and representative.

     Total soil samples ultrasonically dispersed in distilled water remained
dispersed for several weeks.  On the other hand, soil samples dispersed in
any of the natural media coagulated rapidly, usually within  3-7 minutes. A
dispersion index was calculated by dividing the weight percent  of  the  material
with an effective settling diameter less than 2 urn when the  soil was ultra-
sonically dispersed in one of the natural media by percent  total clay  (determined
by ultrasonic dispersion in distilled water) and multiplying by 100.   In
almost every case, the soils were most dispersed (least coagulated) in septic
tank effluent and least dispersed  (most coagulated) in drainage ditch  water,
with treated wastewater and river water yielding intermediate degrees  of
dispersion (Figure 1).  Comparison of data in Table 2 and Figure 1  suggests
a good correlation between sodium concentration of the suspending media and
the degree of coagulation of the soil samples.  Thus it appears that  the
sodium concentration of the fluvial media studied is a major factor controlling
the degree of dispersion or coagulation of the soil samples.

     There were also differences in coagulation among the six soils.   The
variations shown may be due in part to the degree to which  the  soil clays
were already saturated with divalent cations on the exchange complex.   The
soil with the highest amount of  exchangeable Ca + Mg (Paulding) coagulated
most readily, and the soil with  the lowest amount of exchangeable Ca + Mg
(Blount) remained dispersed the  longest.  The other four soils  were inter-
mediate in both respects.  Additionally, variability in the soils'  responses
to dispersion in  the natural media may be attributed to the percent clay in
the  surface horizon.  Figure 1 shows a decrease in the index of dispersion
as the percent clay in  the  surface horizon increased.  In other words, the
higher the clay content,  the more  susceptible  the soil was  to coagulation in
the  four media.

-------
                       FI3UKK 1.  RELATIVE DISPERSABILITY OF SELECTED SOIL SURF4CE HORIZONS
     30
                                                                                          SEPTIC TANK EFFLUENT
                                                                                          TREATED WASTEWATER
     20
  g
  rn
  as
  B5
  0.
  n
     1C
                                                                                                                                                          I
                                                                                                                                                          OA
                                                                                                                                                          o
30IL:
? CLAY:
BLOUNT
  27
HOYTVILLE
   39
LEtfAWEE
  39
                        TOLEDO
ROSELMS
  U8
PAULDIHG
   53

-------
Table 1.  Physical, chemical,and mineralogical  properties  of surface horizons  from six  representative soils  from
          the Maumee Basin.
Soil
Blount
Hoytville
Lenawee
Paulding
Rose 1ms
Toledo
Surface
depth
cm
20
25
23
20
20
18
5 Particle sizes
sand

32.5
19.4
9.9
3.4
72.0
3.4
silt
°f
40.1
42.0
50.8
43.8
44.9
56.1
clay

26.8
38.6
39.3
52.8
47.9
40.5
Exchangeable
cations
Ca

8.1
17.8
21.3
25.7
15.7
20.5
MR

1.7
4.5
4.9
6.2
7.8
4.2
K
/lOOg -
0.29
0.61
0.53
0.55
1.03
0.52
Base Organic Clay
Saturation Carbon mineralogy *
H

6.9
8.7
5.0
4.1
8.5
6.1

59
73
84
88
74
81

1.48
2.55
2.01
1.68
1.91
2.19
111

36
63
51
44
63
67
v/c

46
25
32
33
26
15
Kao Mont Qtz

T
T
T
T
T
3
"I ---
T
T
T
T
T
7

14
7
12
18
7
9
pH
H?0
5.9
6.7
7.0
7.1
6. 6 c^
h-1
1
6.8
* 111= illite, v/c= vermiculite and chlorite,  Kao=  Kaolinite,  Mont=Montmorillonite, Qtz«Quartz,  T=Trace

-------
Table 2.  Chemical data for natural media.
Cation concentrations
Medium
Na
Ca
Mg
K
Fe
Al
pH Electrical conductivit
at 25°C

	 mg/ i 	
Septic tank effluent
Treated wastewater
River water
Drainage ditch water
228
76
38
15
70
58
83
86
52
40
52
50
20
14
9
15
nd
nd
nd
nd
2
nd
nd
nd
7.
7.
8.
7.
8
5
1
9
umhos/cm
1852
840
714
746
                                                                                                                S3
                                                                                                                 I
  nd = not detectable

-------
                                 -63-


     To further examine the influence of clay concentration on coagulation,
various concentrations of clays fractionated from a Paulding surface horizon
were suspended in river water.   A series of photographs depicting coagulation
of the suspended clays over time were taken.  Visual comparison of the relative
turbidities of the different suspensions over time indicated that the higher
the clay content the faster the suspension coagulated and settled.  Both
total surface area and collision frequency increase as colloid concentration
increases.

     A similar series of photographs revealed the influence of cation con-
centration in promoting coagulation of clays.  River water which was diluted
to various  fractions of its original strength was used to suspend a constant
sediment load (500 mg/l).   Significant coagulation (within 16 hours) occurred
only at the highest concentration of river water.  We infer that the divalent
cation concentration at the dispersion/coagulation threshold for a suspended
sediment concentration of 500 mg/l lies in the vicinity of 50 mg/l (Ca + Mg).

     Photomicrographs helped establish that the morphology of coagulates
forming in clay-river water suspensions was highly variable.  Some coagulates
were very tightly packed and dense and had diameters of about 10-20 urn.   These
may represent clays truly aggregated by cation suppression of double-layer
repulsion.   Other coagulates were fairly loose, branching networks of variable
dimensions.  These may be floccules or adsorption coagulates where the clay
particles are linked by bridges of organic matter or hydrolyzed metal ions.
An alternative explanation would be that they are simply precursors to the
compact coagulates and have not yet completed aggregation.

     A number of interacting environmental factors and properties of the
system may encourage or confound coagulation in natural fluvial media.
Results given above confirm for natural media and natural colloids the well-
tested postulate that higher divalent cation concentrations are more effective
than lower concentrations in coagulating colloids.  It is also suggested
that the high base status of the soils (particularly with respect to Ca
and Mg) predisposes the soil colloids toward coagulation after their initial
dispersion.  Thus the critical coagulation concentration of Ca + Mg in fluvial
media of the Maumee Basin may be much less than for watersheds whose soils
are more acidic.

     It has recently been reported by Sherard et * 1   (1976) that erodibility
of fine-textured soils in slow-moving water is positively correlated with
water-extractable Na.  This is somewhat comparable to our finding that
natural media high in Wa supported dispersion longer than media low in Na.
It is speculated that future predictions of coagulation or dispersion in
fluvial transport may end up incorporating (Ca + Mg)/Na ratios (or simply
divalent/monvalent ratios) for the particular soils and media involved.

     The concentration of suspended sediments in transport also appears to
play a role in coagulation.  Clays at concentrations of 500 mg/l coagulated
and settled out of quiet river water after only two hours.  From October 1973
through September 197^ the Maumee River at Waterville had suspended sediment
concentrations considerably lower than those in headwater ditches in the
Basin (Jones et al, 1977).

-------
                                   -64-


      Waterville is also at the lower end  of the  Maumee  drainage
 system,  and the river's sediment load  there probably  already  reflects  the
 quieter  waters of upstream dams  and  local impoundments.  Wall et  al  (1978)
 showed that 34% of the clay in Maumee  River Basin  bottom sediments~were ag-
 gregated into silt and sand-size particles.

      The significance of adsorped species with respect  to clay coagulation
 in natural media is  difficult  to ascertain.   Soil  clays  are thought  to be
 coated with adsorbed materials like  organic  matter and  amorphous  iron and
 aluminium oxyhyroxides or amorphous  aluminosilicates.   These  may  assist in
 the flocculation of  soil colloids by forming bridges  between  the  colloids ,or
 they may simply reduce double-layer  repulsions due to their presence on colloid
 faces.   While some of the coagulates did  have the  three-dimensional  character
 to be expected of floccules, no  clear  trends  in  amorphous constituents or
 organic  carbon in the soils  parallel  the  dispersion indices (Tables  1, 2 and
 Figure 1).   The fact that coagulation  did  depend on colloid surface area in
 the suspension (i.e.,  clay concentration)  strongly suggests that  factors other
 than simply double-layer interactions  are  at  work  (Stumm and  O'Melia, 1968).
 Whitehouse et al.  (1960),  in  an extensive study of model clays,  adsorbates,
 and coagulation in artificial  seawater, indicated  that adsorbed organic com-
 pounds may have a positive,  negative,  or no  effect on clay-settling rates,
 depending  on the speicies involved and the chlorinity of the medium.   Since
 the pHs  of the media  studied here are  probably not greater than the zero points
 of charge  (ZPC)  of amorphous oxyhydroxides or aluminosilicates (Parks, 1967),
 they are likely to be  strongly attracted  to negatively charged clay surfaces,
 thus  encouraging coagulation.

     Aggregation of  clay-size  sediments may be promoted by electrically
 heterogeneous   clay  surfaces.  This occurs when the negatively charged faces
 of clay  platelets  are  attracted  to the positively charged edges of other
 platelets  (Schofield and  Samson,  1954; Bolt and Miller,  1955).  The potential
 determining  ions  are again H+  and OH", but in this case ZPC  is probably less
 than pH,  so  this  coagulation mechanism is  likely not important.

     The pH of  the suspending medium is clearly a significant  parameter
 affecting coagulation, as noted above.  Both seasonal  and daily variations
 occur in the river.  For  example, pH values as high as 9.0  (September 1974)
 and as low as  6.7  (January 1973)  were reported for the Maumee  River at Defiance,
 Ohio in  the 1973-1974 water year  (U.S.  Department of the Interior  Geological
 Survey,  1975).  Daily variations  were  usually <  0.5 pH  units; pH  variability
would have a consequent influence on the degree  of coagulation taking place.
 Seasonal  and daily temperature differences must  also be  taken  into account.
 The Maumee River at Defiance in 1973-1974 had an  overall temperature  range
 of 0-30 C, while daily temperature differences (high minus low) were  usually
around 2-4°C.  Whitehouse et al.   (1960)  have reported that  clay-settling
rates in  saline water may be decreased by about 40% when suspension temperatures
are 6UC compared to 26°C.

     The  fact that soils and fluvial  suspended sediments normally  contain more
than one  type of clay mineral complicates  greatly the  study  of natural  coagula-
tion phenomena.  Different minerals will  vary in  characteristic size,  isoelectric

-------
                                  -65-


 point,  and surface  charge  density.  Some  studies of ncdel  systems  suggest
 that when  tvo  or more  oxides  are  mixed the properties and  coagulating be-
 havior  of  the  system may "be intermediate  between the extremes of the indi-
 vidual  species (e.g.,  Healey  et al., 1973).  On the other  hand, Whitehouse
 et  al.   (i960) demonstrated that  coagulation of clay-mineral mixtures in
 artificial seawater resulted  in coagulates dominated by one clay mineral
 while material left in suspension was dominated by another.  Lund  et al.
 (19T2)  have hypothesized that some clay minerals in suspended sediments
 may be  selectively  coagulated in  reservoirs and thus occur in higher per-
 centages in bottom  sediments  than would be expected from watershed charac-
 teristics.   Wall and Wilding  (1976) essentially proposed the same  explana-
 tion to rationalize the higher percentages of quartz in suspended  river
 sediments  on progressing down the Maumee  River drainage system; they indi-
 cated that quartz must stay in suspension longer than other eroded components,
 Presumably the low  cation  exchange capacity of quartz makes it less suscep-
 tible than other clay  minerals to the chemical coagulation parameters pre-
 viously discussed.

      Finally,  turbulence of the water will influence sedimentation.  While
 it  is true that turbulence may produce more particle collisions and thus
 encourage  coagulation, velocity gradients may still be enough to keep even
 dense coagulates from  settling out of suspension.  However, we postulate that
 in  the  relatively quiet waters of artificial reservoirs, overbank flooding,
 and drainage ditches of low-gradient watersheds like the Maumee, coagulation
 of  suspended sediments is  likely, depending on the chemical and physical
 variables  given above.
                           Soil Aggregation

     Percent sand, silt, total clay,and fine clay for the watershed soils
are given in Table  3 along with their dispersion ratios.  Roselms, Brough-
ton,and Paulding soils had the highest total and fine clay contents, while
Blount was lowest in clay.  The sonification data gave results similar to
total dispersion except that it appeared that sonification was grinding some
of the sand-sized particles into smaller sizes.  Water-dispersion clay contents
were lower than by the other two methods, and the dispersion ratio indicated
that a large percentage of the fine clay and much of the total clay is ag-
gregated into larger particle sizes.  As discussed previously, the high base
status of these soils favors flocculation.

                Clay enrichment in runoff and tile drainage Table  4 gives
percent sand, silt,and clay by sonification for runoff and tile drainage
sediments.  All samples had clay contents > 55$,with the highest clay contents
in the tile sediments.  The enrichment ratio (Table   4) data show that the
sediment contained from 1.2 to 3.6 times as much clay as the surface horizon
of the soil from which it was derived.  Clay content of runoff was more uni-
form than that of the individual soils, indicating that particle-size dis-
tribution of sediment in fluvial  transport may be more a function of transport
phenomena than the size distribution of the source soil material.   The
data also indicate  that much of the coarser particles are deposited in the

-------
 Table   3.  Particle-size analysis and bulk density of Maumee River Basin soils.
Site

m
Rose 1ms
101
Rose 1ms
131
Broughton
201
Ros elms
40x
Blount
50x
Paulding
6xx
Hoytville
Particle-Size Analysis of Reference Soils (%)
Total Dispersion Bonification Water
Total Total
Sand Silt
10.1 41.0

14.6 50.1

8.5 42.9

25.3 42.3

32.8 42.0

6.4 45.7

19.4 43.9
Fine Total
Clay Clay
14.1 48.9

4.7 35.3

10.1 48.6

7.3 32.4

6.1 25.2

9.5 47.9

6.2 36.7
Total Total
Sand Silt
6.0 41.2

10.6 49.4

5.1 39.3

31.6 43.8

27.5 45.5

3.5 43 . 8

16.4 41.3
Fine Total
Clay Clay
11.4 52.8

5.7 40.0

11.3 55.6

5.5 34.6

4.1 27.0

8.4 52.7

6.6 42.3
Total Total
Sand Silt
12 . 5 53 . 7

17.6 59.5

10.4 57.7

28.5 49.4

34.5 47.7

11.9 62.9

24.4 53.4
Fine
Clay
1.9

0.8

1.3

1.1

0.8

0.8

1.1
Dispersi
Total
Clay
33.8

22.9

31.9

22.1

17.8

25.2

22.2
Total Total
Sand Silt
0.8 0.8

0.8 0.8

0.8 0.7

0.9 0.9

1.0 0.9

0.5 0.7

0.8 0.8
Bulk Density
on Ratio 	 (a/cm3)
Fine Total
Clay Clay
7.4 1.4

5.9 1.5

7.8 1.5

6.6 1.5

7.6 1.4

11.9 1.9

5.6 1.7
Field
1.209

--

1.275

1.328

1.464

1.171


Oven
1.522

--

1.595

1.564

1.638

1.540


1.   Particle-size values of reference  soils are weighted means of combined samples which represent all soil types
    within the plot.   Bulk density values are  from specific soil types within the plot.

-------
Table   4.  Particle-size analysis of runoff  sediment  (Sonification).
                                                                                                       Enrichment Ratio
Total Total Total
Sand Silt Clay
111 0.0-1.2 20.1-52.8 47.0-79.7
Rose 1ms
surface
121
Roselms
surface
131 0.4-2.0 16.6-32.7 66.3-83.0
Broughton
surface
201 0.0-20.6 17.6-69.0 29.8-82.4
Roselms
surface
401 0.2-2.4 16.9-53.5 44.7-82.9
Blount
surface
402
Blount
surface
501 0.0-3.8 13.6-47.3 49.5-86.2
Paulding
surface
502 0.0-1.8 6.4-26.5 73.5-93.6
Paulding
tile
6x1 0.0-17.6 24.8-62.4 36.0-74.4
Hoytville
surface 	 	 	
Total Total Total
Sand Silt Clay
0.3 35.9 63.8


0.0 16.2 83.8


1.1 24.8 74.1


2.0 42.3 55.8


1.3 40.5 58.2


0.0 4.2 95.8


1.3 28.8 70.0


0.4 12.5 87.1


2.4 44.0 53.3


Total Total Total
Sand Silt Clay
0.3 9.5 9.6


	 	 	


0.7 6.9 7.1


4.7 13.6 14.5


0.8 11.9 12.2


	 	 	


1.4 11.3 12.2


0.7 7.5 7.6


4.2 9.9 10.5


Total Total Total
Sand Silt Clay
0.1 0.9 1.2


0.0 0.3 2.1


0.2 0.6 1.3


0.1 1.0 1.6 ,
*s.

0.1 0.9 2.2


0.0 0.1 3.6


0.4 0.7 1.3


0.1 0.3 1.7


0.2 1.1 1.3



-------
                                  -68-


 field and do not enter the stream system.  Therefore, the 5-30% sand found
 by Wall et al (1978) in Maumee River Basin bottom sediments may be from
 streambank erosion.  The high clay content of sediment in agricultural
 runoff illustrates why these sediments are so reactive and high in total P
 (McCallister and Logan, 1978).

 3.45  Conclusions

      The coagulation-dispersion response of soil colloids in the natural
 media considered here depended on the concentration of suspended clays, the
 degree to which they were already saturated with divalent cations, and the
 cation chemistry of the suspending medium.  A survey of the various factors
 affecting natural coagulation suggests that coagulating conditions are
 most likely in the spring of the year, when suspended sediment concentrations
 are the highest, pHs are near their annual ebb,  and high discharge rates
 from contributing watersheds produce impoundments in the drainage system.
 It would seem likely that clays in fluvial transport undergo multiple coagula-
 tion-resuspension events as  they move from terrestrial to lacustrine environ-
 ments.   More detailed research is certainly needed to clarify the fate of
 natural soil colloids in natural fluvial systems.

      Runoff sediment is higher in clay than the  surface soil from which it
 is eroded and,  in most  cases,  the sediment contains > 55% clay.

      Maumee Basin soils are  aggregated with the  fine and total clay aggre-
 gates to form larger particles.

 3.46  Literature Cited

 1.  Bolt,  G.  H.  and  R.  D. Miller.   1955.   Compression Studies  of  Illite
    Suspensions.   Soil  Sci.  Soc.  Am.  Proc.  19:285-288.

 2.  Day,  P. R.   1965.   Particle  Fractionation and  Particle-Size Analysis.
    In  C. A.  Black  (ed.) Methods  of  Soil Analysis,  part  1.   Agronomy  9:545-567.

 3.  Hashimoto, I. and M. L. Jackson.   1960.  Rapid  Dissolution of Allophane
    and Kaolinite-Halloysite After Dehydration.  Clays  Clay  Miner.  7:102-113.

 4.  Healy, T. W., G. R. Wiese, D. E. Yates, and B.  V. Kavanagh.  Hetero-
    coagulation  in Mixed Oxide Colloidal Dispersions.  J. Colloid Interface
    Sci. 42:647-749.

5.  Johns, W. P., R. E. Grim, and W. F. Bradley.   1954.  Quantitative Estimations
    of Clay Minerals by Diffraction Methods.  J.  Sediment. Petrol. 24:242-251.

6.  Jones, L. A., Neil E. Smeck, and L. P. Wilding.  1977.  Quality of Water
    Discharged from Three Small Agronomic Watersheds in the Maumee River
    Basin.  J. Environ.  Qual. 7:296-302.

-------
                                   -69-
 7.   LaMer, V.  K.   1964.   Coagulation Symposium Introduction.  J. Colloid Sci.
     19:291-293.

 8.   Lund,  L.  J.,  Helmut  Kohnke, and Manuel Paulet.  1972.  An Interpretation
     of Reservoir  Sedimentation II.   Clay Mineralogy.  J. Environ. Qual.
     1:303-307.

 9.   McCallister,  D.  L. and T.  J. Logan.  1978.  Phosphate Adsorption-
     Desorption Characteristics of Soils and Bottom Sediments in the Maumee
     River  Basin of Ohio.  J. Environ. Qual. 7:82-92.

10.   Parks, G.  A.   1967.   Aqueous Surface Chemistry of Oxides and Complex
     Oxide  Minerals;  Isoelectric Point and Zero Point of Charge.  In Equili-
     brium  Concepts in Natural Water Systems.  Adv. Chem. 67:121-160.  Am.
     Chem.  Soc.,  Washington, D. C.

11.   Rutledge,  E.  M., L.  P. Wilding, G. F. Hall, and N. Holowaychuk.  1975.
     Loess  in Ohio in Relation to Several Possible Source Areas:  II.  Elemental
     and Mineralogical Composition.   Soil Sci. Soc. Am. Proc. 39:1133-1139.

12.   Saunders,  W.  M.  H.  1965.   Phosphate Retention by New Zealand Soils and
     Its Relationship to Free Sesquioxides, Organic Matter, and Other Soil
     Properties.   N.  Z. J. Agric. Res. 8:30-57.

13.   Schofield, R. K. and H. R. Samson.  1954.  The Deflocculation of Kaolinite
     Due to Attraction of Oppositely Charged Clay Crystal Faces.  Discuss.
     Faraday Soc.  18:135-145.

14.   Sherard, J.  L.,  L. P. Dunnigan, and R. S. Decker.  1976.  Identification
     and Nature of Dispersive Soils.  J. Geotech. Eng. Div., ASCE 102:287-301.

15.   Smeck, N.  E., L. P.  Wilding, and N. Holowaychuk.  1968.  Genesis of Argillic
     Horizons in Celina and Morley Soils of Western Ohio.  Soil Sci. Soc. Am.
     Proc.  32:551-556.

16.   Stumm, W.  and J. J.  Morgan.  1979.  Aquatic Chemistry.  Wiley-Interscience,
     New York.

17.   Stumm, W.  and C. R.  O'Melia.  1968.  Stoichiometry of Coagulation.
     J. Am. Water Works Assoc. 60:514-539.

18.   U. S.  Department of the Interior, Geological Survey.  1973.  Water Resources
     Data  for Ohio.  Part 2:  Water Quality Records.

19.   U. S.  Department of the Interior, Geological Survey, 1975.  Water Resources
     Data  for Ohio.  Part 2:  Water Quality Records.

20.   Verwey, E. J. W. and J. Th.  G. Overbeek.  1948.  Theory of  the  Stability
     of Lyophobic Colloids.  Elsevier, Amsterdam.

-------
                                   -70-


21.  Wall, G. J. and L. P. Wilding.  1976.  Mineralogy  and Related  Parameters
     of Fluvial Suspended Sediments in Northwestern Ohio.  J.  Environ.  Qual.
     5:168-173.

22.  Whitehouse, U. G., L. M. Jeffrey, and J. D. Debbrecht.  1960.   Differential
     settling Tendencies of Clay Minerals in Saline Waters.  Clays  Clay Miner.
     7:1-79.

23.  Winters, E. and R. S. Smith.  1929.  Determination of Total  Carbon in
     Soils.  J. Ind. Eng. Chem., Anal. Ed. 1:202.

24.  Wischmeier, W. H. and J. V. Mannering.  1969.  Relation of Soil Properties
     to Its Erodibility.  Soil  Sci. Soc. Am. Proc. 33:131-137.

25.  Steele, J. G.  and R. Bradfield.   1934.  The Significance  of  Size
     Distribution in the Clay Fraction.  Amer. Soil Survey Assoc. Bull.
     15:88-93.

26.  Wall, G. J.,  L. P. Wilding and Nt  E. Snseck.  1978.  Physical,  Chemical
     and Mineralogical Properties of Fluvial Unconsolidated Bottom  Sediments
     in Northwestern Ohio.  Jour. Envir. Qual. 7:319-325.

-------
                                 -71-
3.5  Clay-equilibration studies  in natural and simulated bottom  sediment
     environments
3.51  Abstract

     Predominantly illitic soil clays were equilibrated in  bottom-sediment
environments under natural and laboratory conditions  for 28l days  and 98
days, respectively.  Chemical, physical,and mineralogical parameters  vere
monitored to identify and characterize any alterations.

     Changes occurring after equilibrating for 28l days under natural
conditions consisted of a reduction in carbonate contents,  decrease in
particle size,and a slight loss of Al and Si.   No significant changes vere
observed in mineralogical composition.

     Clays equilibrated under laboratory conditions at 4 and 25  C in
either carbon dioxide, nitrogen,or air atmospheres revealed few changes
other than an increase in amorphous iron.  Concentrations of Al,  Si,  Fe,
Mn, K,and Ca monitored in solution above the bottom sediments varied with
the atmosphere and temperature.  The concentrations of water-soluble Al
and Si in solution appeared to be influenced by iron oxide  coatings in
the bottom sediments.

-------
                                  -72-

 3-52  Introduction

      The mode and extent of clay-mineral alteration in fluviatile  sediments
 may be limited, but lov stream gradients,  high concentrations  of flocculating
 ions,and impoundments increase the residence  time  of sediments and thus  the
 possibility of alteration.

      Grim (1968) states that clay minerals isolated from lake  and  fluviatile
 sediments should correspond with those  from the source area, differing only
 due to the removal of alkali and alkaline  earth components.  However, other
 individuals (Frink, 1969;  Lietzke and Mortland, 1973)  have  indicated that
 dechloritization of chloritized-vermiculite can occur  when  these minerals  from
 acid source areas are equilibrated in alkaline bottom  sediments.   Frink  (1969)
 further suggested that chloritized-vermiculite was  converted to  illite
 during the dechloritization process,as  clay mineralogy data indicated higher
 amounts of illite and correspondingly lower amounts  of vermiculite in the
 lake sediment than occurred in surrounding soils.   However, the  study did  not
 relate preferential transport and differential settling of  clay  minerals to
 the assemblage observed in  the sediment.   Recent studies  (Rhoton,  1978)
 have shown that clay minerals occurring in smaller  particle-sizes  are
 selectively transported from soil surfaces and can  explain  observed differences
 between bottom sediment and soil clay mineralogy.   In  other studies, Wall
 et  al (197M  investigated clay-mineral  alterations  under  laboratory conditions
 and concluded that geologic clays were  altered under static conditions as
 evidenced by  decreasing trends in mica,  expandables  and vermiculite, and
 increases in  quartz and chlorite contents.

      The present study was  designed to  extend  previous  alteration  studies
 further through utilization of supporting  data from  a  number of mineralogical,
 physical,and  chemical  analyses obtained before  and  after  equilibrating clay
 materials in  a natural bottom sediment  environment and under laboratory
 conditions.   The objectives  were to determine  the amount  and type  of clay
 mineral alterations  occurring in bottom sediment environments  and  the relative
 contributions  made by  different  types of equilibrating  environments.


 3-53  Methods  and  Materials

                         In  Situ Equilibration

      Twenty grams  of Mg-saturated  clay  ( <  2 urn) from the calcareous C-horizon
 of  a Hoytville  pedon (Mollic  Ochraqualf) were placed in Plexiglas  cylinders
 (5  cm x  20 cm).  Nucleopore membranes (pore size 3.0 urn) applied to each
 end by rubber  0-rings  contained  the clay sample.  This design permitted
 free  exchange between the cell contents and the fluvial environment without
 any  gain  or loss of material.  Cells were attached by spring-type clips  to
 a plexiglas plate which was  surrounded by  Fiberglas screen.   Scuba divers
 staked the plate at the sediment-water interface above Edison Power Dam on
the Auglaize River, northwestern Ohio.  Individual cells were retrieved by
 divers at intervals of 38,  ikQ, 217,and 28l days.  Cell contents were compared
to reference samples using:   clay mineralogy,  broadening of the (001)  illite
peak,  cation exchange capacity, percent  calcite and dolomite,  amorphous  iron,
aluminum and silica, free iron oxides, total iron,  aluminum, silica and
potassium, and external surface area.

-------
                                 -73-
                       Laboratory Equilibration

     Approximately 3 ,330 grams of untreated Auglaize River bottom sediment
and 13.0 liters of river water were placed in each of 10 aquariums (18.9 liter)
and pre-equilibrated for one week at a controlled temperature of k° C in
one of three atmospheres:  compressed air, nitrogen gas (99-9%},and 20% carbon
dioxide.  Four aquariums each were equilibrated with compressed air and
nitrogen gas whereas the remaining two were equilibrated with carbon dioxide.
Respective gases were bubbled through tygon tubing at the sediment-water
interface.

     Three cells containing fifteen grams of either Hoytville A or C horizon
clay were partially submerged in the sediment of each aquarium following the
pre-equilibration.  Only calcareous C-horizon clays were equilibrated in the
two carbon dioxide aquariums.  Once the clay cells were in place,aquariums
were sealed with Plexiglas lids.

     The scheme for addition of gas into the systems was arbitrarily
determined by the gas.  Nitrogen gas was bubbled through the aquariums daily
for 12 hours ,whereas carbon dioxide was bubbled for 6 hours daily,then
shut off for 18 consecutive hours.  Compressed air was bubbled continuously.
All aquariums were equilibrated as such for U5 days at h° C,at which time
five aquariums (2 air, 2 nitrogen, 1 C02 ) were discontinued.  The remaining
five were equilibrated an additional 53 days at 2|?  C.

     Solution samples and clay cells were removed periodically at each
temperature.  Solution samples were withdrawn in 200 ml aliquots at a constant
distance parallel to the sediment-water interface immediately following an
equilibration period.  One set of five aquariums was  sampled frequently at
k° C (days 1, 3, 5, 7, I1*, 21, 28, bj) then discarded due to depletion of
solution.  The other set was sampled only occasionally at h  C( days 1, 21, U5)
but more frequently at 25° C (days h9, 5k, 69, 85, 98).  Reference solution
samples were collected from each aquarium after pre-equilibrating and
immediately prior to adding clay cells.  Individual clay cells were removed
for analysis at approximately two-week intervals at both h° C and 25° C.
Thus, comparisons could be made of the relative temperature effects on
both clay and solution parameters.  All solution and clay samples were stored
at k° C until analyses were completed.

     Solution parameters analyzed included:  dissolved silica, iron, aluminum,
manganese, calcium?and potassium.  Clay samples were analyzed for amorphous
iron, free iron oxides, total silica, aluminum, iron, potassium,and calcium,
percent calcite and dolomite.  No analyses were conducted for clay mineralogy,
peak broadening, surface area, amorphous aluminum,and silica, or cation
exchange capacity.  Redox potential and pH measurements were made on the
sediment and solution phases before clay cells were added to the aquariums;
thereafter these parameters were determined only for the solution phase at
the time of sampling.

                          Analytical Techniques

     Clay samples were Mg-saturated and methanol-washed prior to  and after
equilibration.  X-ray diffraction utilized a Norelco diffractometer, copper
radiation, nickel K-beta filter, 0.006-inch receiving slit, 1° divergence
slit, proportional counter, and a chart speed of 30 inches/hour.  Sample
preparation involved vacuum plating 0.15 grams of clay on ceramic plates

-------
                                  -74-
 (Kinter and Diamond,  1956).   Sample  treatments  included  saturation with
 ethylene glycol,  drying at  25°  C, heating  to  UOOc C  and  550°  C.  Relative
 percentages of illite,  chloritized vermiculute, kaolinite,and quartz were
 determined following  a  modified Johns  et al (195^) method.  Peak areas
 were  delineated according to  symmetry  and  average  baseline heights.
 Photocopies of the  peaks were weighed  and  clay mineral percentages reported
 as  a  mean of the  subsamples extracted  from each cell.  Measurements of peak-
 broadening utilized the procedure of Jackson  (1977).

      Cation exchange  capacities  were determined by the sodium saturation
 method of Chapman (1965) after  diaiyzing the  clays against a  10$ solution
 of  acetic acid to remove carbonates.   Calcite and  dolomite percentages were
 obtained by the Chittick apparatus (Dreimanis, 1962).  Amorphous iron and
 free  iron oxide were  extracted with  0.2 M  ammonium oxalate (McKeague and
 Day,  1966)  and citrate-bicarbonate-dithionite (Mehra and Jackson, I960),
 respectively,and  quantified by atomic  absorption.  Amorphous  aluminum and
 silica contents were  determined  by treating clays with boiling 0.5 N; NaOH
 (Hashimoto and Jackson, 1960).   Total  silica  and aluminum samples were
 decomposed in  a Teflon  bomb (Bernas, 1968) and analyzed by atomic absorption.
 Total iron and potassium contents were obtained  by  X-ray spectrographic
 methods.   External  surface area  was  determined  by   a BET method by
 Micrometrics  Instrument Corporation, Norcross, Georgia.

      Dissolved silica,  aluminum, iron ,and manganese were determined by the
 colorimetric method of  Rainwater and Thatcher (i960).  Dissolved potassium
 and calcium analyses  followed atomic absorption and  emission procedures.
 Redox potential and pH  measurments were made by inserting platinum wire
 or  glass  electrodes through holes in the aquarium lids, into solution.  Redox
 potentials  were read  from the millivolt scale of the pH meter.


 3.5^   Results  and Discussion

                          In  Situ Equilibration

      The  in  situ  equilibration study was designed for a duration of two
 years  but was  destroyed by flooding after 28l days.  Only four cells were
 collected,  at  intervals of 38, ihQ, 217 ,and 28l days.  The clay mineralogy
 of  these  samples  (Table 1 ) reveals only slight variation with time and are
 well within  the margin of error  for such semi-quantitative analyses.  In
 addition to  time, the lack of alterations of sufficient magnitude to
 result in significant mineralogical changes may be due to the formation of
 iron  oxide  coatings  on clay mineral surface which prevent equilibrium reactions
with  the river  environment.  Eswaran and Heng  (1976) indicated that goethite
 coatings were  effective in reducing the alteration of vermiculite and
 interstratified minerals.

     Elemental analyses (Table  2)   indicated some statistically significant
 (5$ level) changes with time;  however,  absolute concentration differences
are minimal in most  cases.   Largest changes occurred in total aluminum
and silica concentrations as both decreased through 217 days.   The increases after
 281   days may be due to sample contamination.   Absolute concentration of
total iron and potassium shows little  change with time.

     The small increases in  amorphous iron  and free iron  oxide contents were
nonsignificant (5% level).   The  greatest  increase  in  these two parameters

-------
                                  -75-
Table 1.  Mineralogical composition of in situ  equilibrated clays

Length of               Chloritized Vermiculite
Equilibration(Days)      	Vermiculite	  Illite    Kaolinite   Quartz

0
38
148
217
281

17
15
16
17
15
cy
76
75
76
73
77

2
2
2
4
3

5
8
6
6
5

-------
                                         -76-
Table 2.  Elemented and oxide analysis of In situ equilibrated clays
ibrationfDavs}


0
38
148
217
281
Fe


49.0
49.0
48.5
49.8
49.8
Al


112.0
110.0
105.0
98.6
131.0
Si


198.
197.
183.
173.
178.



0
0
0
0
0
K


29.0
29.0
29.8
30.4
30.5


m§/g ~
0
0
0
0
1
Fe


.80
.78
.84
.86
.16
Al


2.5
3.3
3.5
2.5
2.5
Si


4.5
5.5
8.0
5.8
8.5
J. A.
Iron


22
23
25
23
26
w^.
Oxides


.9
.3
.5
.8
.4

-------
                                 -77-
occurred at 28l days when FeS crystals  were observed in the cell.   Amorphous
aluminum shows little change with time,whereas  the amorphous silica concen-
tration appeared to increase.

     External surface area, cation exchange capacitysand carbonate content
changes with equilibration time are listed in Table 3  .  Although  the
increase in surface area suggests a decrease in particle size,  the decrease
may be influenced more by the progressive dissolution of carbonates than
by   clay   mineral decomposition with  time.  The carbonate analyses indicated
calcite decreased at a slightly greater rate than dolomite.  Carbonate
dissolution is attributed to submergence of the clay cells in the  bottom
sediment where the pH approximates 7.2.  Calcite precipitation has been
observed (Green et al, 1978) in the overlying river water where the pH
ranges from 8.2-8.5.  Cation exchange capacity values fluctuated considerably,
indicating no time trend or correlation with the increased surface area.  The
erratic values may be due to incomplete carbonate dissolution by the acetic
acid pretreatment.

     Measurements conducted to determine the degree of peak-broadening
(Table  4)   were obtained from the air dry, (001) illite peak.  Results
indicated that broadening had occurred, which normally indicates a particle-
size decrease.  This suggests that the observed increase in external surface
area may be related to clay-mineral degradation,which would presumably occur
as a surface hydrolysis reaction disrupting the (001) reflection plane.  Iron
oxide contributions to the observed peak-broadening cannot be discounted
since these materials were not removed prior to scanning the clays.

                        Laboratory Equilibration

     Amorphous iron contents from the laboratory equilibrated study varied
according to clay-material source  (Table 5  ).  C-horizon clays were relatively
unaffected other than the  increase under the CC2 atmosphere at 25  C,whereas
the amorphous  iron  content of A-horizon clays increased substantially, especially
in the nitrogen atmosphere.  The source of  the increase was observed in later
samples.  The  aquarium systems met the criteria listed as  necessary for FeS
formation  (Connell  and Patrick,  1968; Doner and Lynn,  1977).  The  greater
increase of amorphous iron under the nitrogen atmosphere in A-horizon clays
is believed to be  due to  lower redox potentials in  the overlying water,
which permitted greater  concentrations of  sulfide ions and ferrous iron to
diffuse  across the  sediment-water  interface.  Critical redox potential measurements
from the aquarium sediment were  less than  -300 mv.  The redox potential of
the  solution phase  in the nitrogen atmosphere eventually decreased to less
than -kOO mv,whereas  similar measurements  exceeded  +300 mv for other
atmospheres.   The  lack of any  substantial  increases of amorphous  iron in
C-horizon  clay cells may be  related to the carbonate material maintaining
the  pH  near  the upper limit  of the FeS precipitation range.

     No  significant differences  or trends  were  identified  in the  free iron
oxide  contents or in the total  elemental  analyses  (Al, Si,  Fe, Ca,  K).regard-
less  of temperatures  or  atmosphere.  Additionally,  calcite and  dolomite
percentages  were  not  affected  by the equilibration. All  observed variations
were within  the analytical margin  of error.

-------
                                      -78-
Table 3.  External surface area,  cation exchange capacity and  carbonate  analysis
          of in situ equilibrated clays

Length of                Surface  Area    Cation Exchange Capacity  %  Carbonates
Equilibration(Davs')       (meters2/gram)      (meq./lOO  grams)     Calcite   Dolomite

         0                   46.1

        38                   N/A

       148                   48.2

       217                   51.9

       281                   51.6
25
29
24
20
39
7.0
8.2
6.6
4.9
4.9
4.0
2.3
3.0
2.8
2.4

-------
                                      -79-
Table  4 .   Measurements of (001)  illite peak broadening in in situ clays
            equilibrated for 217 and 281 days
                 Peak Height (mm)
Written at % Peak Height (mm)
Number
1
2
3
4
5
6
7
8
9
10
Mean
Width/Height
Reference
186
186
190
181
191
197
176
190
178
184
185.9
.026
217
149
132
142
128
139
141
147
141
158
138
141.5
.036
281
171
154
182
172
161
162
150
163
163
161
163.9
.032
Reference
5.5
4.8
5.3
5.0
4.8
4.5
4.8
5.0
5.5
4.8
5.0

217
5.0
4.5
4.8
5.5
5.0
5.0
5.0
5.3
5.0
5.5
5.1

281
5.0
5.3
5.5
5.5
5.3
5.3
5.5
5.3
5.0
5.3
5.3


-------
                                      -80-
Table   5.  Amorphous iron content of laboratory equilibrated clays


Length of                                 C-Horizon Clay           A-Horizon Clayf
Equilibration(Days)


0
14
28
45
69
85
98
Temp. (°C)*


__
4
4
4
25
25
25
N2



1.2
1.2
1.2
1.7
1.2
1.4
CO,


1.1+
1.4
1.1
1.4
2.0
1.9
2.4
Air



1.4
1.5
1.4
1.0
1.3
1.2
NO Air

.idy 	
3.8+
4.6 3.9
4.3 4.2
4.6 3.6
6.6 4.1
6.4 5.0
8.0 5.4
*  At the end of 45 days the temperature was increased from 4°C  to 25°C.
j-  A-horizon clays were not equilibrated in CO  atmospheres.
+  Reference value.

-------
                                 -81-
     Concentrations of water-soluble silica, aluminum, iron, manganese,
calcium,and potassium were monitored in all aquariums at "both h° C and
25  C.  Redox potentials and pH were monitored simultaneously on identical
solutions.  Silica concentrations in solution were substantially higher
under the CC>2 atmosphere than under the N2 or air atmosphere ( Figure  1 )
and increased considerably after the temperature was increased to 25° C
(Figure   2).  The air atmosphere system contained initially higher concen-
trations than the nitrogen atmosphere (Figure l); however, more silica was
present in the nitrogen system after ^5 days.  The nitrogen system also
contained a higher  concentration at 25° C (Figure  2) »as both absolute
concentration and percentage increase exceeded that of the other atmospheres.

     Aluminum concentrations in solution at ^° C (Figure 3) varied in a manner
similar to silica.  Essentially no differences existed between the nitrogen
and air atmosphere,but concentrations in the CC>2 atmosphere were exceedingly
higher, increasing with time.  The nitrogen gas system again exhibited the
higher concentrations in solution at 25°C (Figure  4).

     Silica and aluminum concentrations in solution appear to be related.
Absolute concentrations and percentage increases were at a maximum in the
C02 system at U° C.  The average solution pH for the CC^ system was 6.2 at
h  C ,compared to 9-0 for nitrogen and 8.2 for air.  Initial sediment pH
readings for these systems were 6.6, 7-9>and 7-6, respectively, after the
one week pre-equilibration period.  The lower pH reading for the CCu system
was due to H+ production from lUCCL dissociation.  Since the solubility
of amorphous silica is nearly constant below pH 9.0,it is assumed the
increased solubility of silica in the CCu system is due to removal of Al from
amorphous aluminosilicates ,leaving silica relics which are less stable
(Wildman et al, 1968).

  o+ Wildinget al (1977) indicate that metallic ion coatings, particularly
Al   and Fe   . can significantly reduce the solubility of amorphous silica.
Removal of Fe   coatings as reducing conditions become more intense may
account for the higher concentrations of silica and aluminum in the
nitrogen atmosphere at 25° C.  However, increased reaction rates at 2p  C
and the higher pH of the nitrogen system cannot be disregarded.  The higher
pH is related to the production of ammonia, and hydroxyl ions released as
a result of reduction of ferric and manganic compounds as the redox potential
decreased (Redman and Patrick, 1965).  Although the sediment pH was not
monitored after the initial reading of 7-9 >it is assumed that the value may
have eventually approximated that of the overlying water.

     Iron concentrations in solution at h° C were low and differed little
as a function of atmosphere.  At 25° C, the nitrogen atmosphere affected
a substantial, erratic increase ,whereas concentration changes in the CO?
and air atmospheres were slight and approximately equal (Figure  5).  Manganese
concentrations were too erratic to report at ^° C; however, the higher
concentrations were detected in the C02 system.  Manganese was detected in
only one sample from the nitrogen systemfwhereas none of the air-equilibrated
samples contained detectable amounts.  The C02 system also had the greatest
effect on manganese concentrations at 25° C  (Figure  6 ), but substantial
amounts were also recorded in the  nitrogen system.  Again, no detectable
manganese was found in the air equilibrated system at 25° C.

     The differences in iron concentrations between equilibrating systems
appear to be strictly related to redox potential.  Sufficiently low redox

-------
O
£
O
co
                                   14            21             28
                                       Equilibration Period  (days)
                                                                                C02
                                                                                Air
                                                                                N2
                                                                                                            I
                                                                                                            00
45
      Figure 1.  Concentration of water-soluble silica as a function of time and atmosphere  at 4°C.

-------
20 r
                      21
       45  49  54

Equilibration Period (days)
69
                                                                                                    oo
                                                                                                    w
                                                                                                    i
                                                                                               98
   Figure 2.  Concentration of water-soluble silica as a function of time,

              temperature,and  atmosphere.

-------
      357
14
          21              28

Equilibration Period (days)
                                                                                                  00
                                                                                                  -P-
                                                                                                   i
45
Figure 3.  Concentration ofwater-soluble aluminum as a function of time and atmosphere
           at 4°C.

-------
o>
    801-
    60
o
o  40
O
    20
                                           4°C  25°C
                         21
          45  49   54

Equilibration Period  (days)
69
85
                                                                                                      00
                                                                                                      Ln
                                                                                                       I
98
       Figure 4.  Concentration of water-soluble aluminum as a function of  time, temperature,

                 and atmosphere.

-------
                   21
                                                                                                  I
                                                                                                 00
          45  49  54

Equilibration Period (days)
69
85
98
Figure 5.   Concentration of water-soluble iron as a function of time,  temperature,
           and  atmosphere.

-------
                   21
         45  49  54
Equilibration  Period  (days)
                                                                69
                                                                                                      I
                                                                                                     00
85
                                                            98
Figure 6.  Concentration of water-soluble manganese as a function of time, temperature ,
           and atmosphere.

-------
potentials  in the nitrogen system led  to  the reduction of  ferric  compounds
and perhaps decreased the thickness of  the oxidized zone at  the interface,
permitting  a greater amount of iron to  diffuse through into  solution.   The
processes were greatly affected by a change in temperature,  since  comparable
redox potentials were recorded at both  temperatures but much higher  concen-
trations existed at 25° C.

     No significant differences existed for water-soluble  calcium  between
nitrogen and air systems at either temperature; however, the CC^ system
yielded substantially higher concentrations that appeared  to increase with
time at both 4 and 25  C.  Apparently,  calcium concentrations were controlled
by the dissolution rate of calcite and/or dolomite in the  sediments, since
the higher  concentrations were associated with the lower pH  values.  Potassium
concentrations exhibited little change  with time, regardless of temperature
or equilibrating atmosphere.

3.55  Conclusions

     Clay-mineral alterations in freshwater bottom sediment  environments
are minimal when equilibrated for approximately one year.  Any alterations
occurring may involve surface reactions leading to particle-size reduction
but not a change in basal spacings.

     The existence of iron oxide coatings on amorphous aluminosilicates
in bottom sediments appears to control  dissolution rates and may influence
clay minerals assemblages in bottom sediments.

3.56  Literature Cited

1.  Bernas, B.  1968.  A New Method for Decomposition and  Comprehensive
    Analysis of Silicates by Atomic Absorption Spectrometry.  Anal.  Chem.
    40:1682-1686.

2.  Chapman, H. D.  1965.  Cation Exchange Capacity.  In Methods of  Soil
    Analysis (Edited by C. A.  Black).   American Society of Agronomy, Madison.

3.  Connell, W. E. and W. H. Patrick, Jr.   1968.  Sulfate  Reduction  in
    Soil.  Effects of Redox Potential and pH.  Science 159:86-87.

4.  Doner, H.  E.  and W.  C. Lynn.   1977.  Carbonate, Halide,  Sulfate,
    and Sulfide Minerals.  In Minerals  in Soil Environments  (Edited by
    J.  B. Dixon and S.  B. Weed).   Soil  Science Society of America, Madison.

5.  Dreimanis, A.  1962.   Quantitative Gasometric Determination of Calcite
    and Dolomite by Using Chittick Apparatus.  J.  Sediment.  Petrol. 32:520-529.

6.  Eswaran, H. and Y.  Y. Heng.   1976.   The Weathering of Biotite in a
    Profile on Gneiss in Malaysia.   Geoderma 16:9-20.

7.  Frink, C.  R.   1969.   Chemical and Mineralogical Characteristics of
    Eutrophic  Lake Sediments.   Soil Sci. Soc.  Am.  Prof.  33:369-372.

8.  Green, D.  B., T.J.  Logan and  N.  E.  Smeck.   1978.   Phosphate Adsorption
    Desorption Characteristics  of Suspended Sediments  in the Maumee River
    Basin of Ohio.   J.  Environ.  Qual.  7:208-212.

-------
                               -89-
 9.  Grim, R. E.  1968.  Clay Mineralogy.  McGraw-Hill, New York.

10.  Hashimoto, I. and M. L. Jackson.  1960.  Rapid Dissolution of
     Allophane and Kaolinite-Halloysite after Dehydration.  Clays and
     Clay Minerals 7:102-113.

11.  Jackson, T. A.  1977.  A Relationship Between Crystallographic Properties
     of Illite and Chemical Properties of Extractable Organic Matter in
     Pre-Pleanerozoic and Phanerozoic Sediments.  Clays and Clay Minerals
     25:187-195.

12.  Johns, W. D., R. E. Grim and W. F. Bradley.  1954.  Quantitative
     Estimations of Clay Minerals by Diffraction Methods.  J. Sediment.
     Petrol. 24:242-251.

13.  Kinter, E. B. and S. Diamond.  1956.  A New Method for Preparation
     and Treatment of Oriented Specimens of Soil Clays for X-Ray Diffraction
     Analysis.  Soil Sci. 81:111-120.

14.  Kietzke, D. A. and M. M. Mortland.  1973.  The Dynamic Character of
     a Chloritized Vermiculite Soil Clay.  Soil Sci. Soc. Am. Proc. 37:651-656.

15.  McKeague, J. A. and J. H. Day.  1966.  Dithionite and Oxalate Extractable
     Iron and Aluminum as Aids in Differentiating Various Classes of Soils.
     Can. J. Soil Sci. 46:13-22,

16.  Mehra, 0. P. and M. L. Jackson.  1960.  Iron Oxide Removal from Soils
     and Clays by a Dithionite-Citrate System Buffered with Sodium Bicarbonate.
     Clays and Clay Minerals.  7:317-327.

17.  Rainwater, F. 0. H. and L. L. Thatcher.  1960.  Methods for Collection
     and Analysis of Water Samples.  Geol. Surv. Water Supply paper 1454.

18.  Redman, I. H. and W. H. Patrick.  1965.  Effect of Submergence on
     Several Biological and Chemical Soil Properties.  La. State Univ. Res.
     Bull. 592.

19.  Rhoton, F. E.  1978.  Clay Mineralogical Relationships Between Watershed
     Soils, Runoff and Bottom Sediments in the Mautnee River Basin, Ohio.
     Unpublished Ph.D. Thesis, The Ohio State Univ., Columbus.

20.  Wall, G. J., L. P. Wilding, and R. H. Miller.  1974.  Biological
     Transformations of Clay-Sized Sediments in Simulated Aquatic Environments
     Proc. 17th Conf. Great Lakes Res., 207-211.

21.  Wilding, L. P., N. E. Smeck, and L. R. Drees.  1977.  Silica in Soils:
     Quartz, Cristobalite, Tridymite and Opal.  In Minerals in Soil Environments
     (Edited by J. B. Dixon and S. B. Weed).  Soil Science Society of America,
     Madison.

22.  Wildman, W. E., M. L. Jackson, and L. D. Whittig.  1968.  Serpentinite
     Rock Dissolution as a Function of Carbon Dioxide Pressure in Aqueous Solution.
     Am. Miner. 54:1252-1263.

-------
                                 -90-
3.6  Occurrence and Stability of Calcite in the Maumee River


3.6l  Abstract

    The occurrence of secondary (precipitated)  calcite and the concentra-
tions of parameters controlling calcite dissolution-precipitation were mon-
itored at six sites in the Maumee River drainage system during 1975-76.
Secondary calcite was identified on the basis of calcite/dolomite ratios
as determined by X-ray diffraction and a volumetric procedure employing the
Chittick apparatus.  Additional evidence for the occurrence of secondary
calcite as well as crystal morphology was obtained by optical microscopy.
A positive correlation between high calcium concentrations and the occurrence
of secondary calcite could not be established;  however, a relationship between
algal blooms and calcite formation was noted.  It is concluded that a reduc-
tion in the CC^ content of the stream due to algal biomass production is
the factor responsible for initiating calcite precipitation in the Maumee
River system.  Calcite equilibria in the stream is quite dynamic with secondary
calcite appearing and disappearing in relatively short periods of time.

-------
                                  -91-
3.62  Introduction

     The contribution of calcite, which is a common component of lake and
stream sediments, to the sorption of inorganic phosphate by sediments has
been examined (Shukla et al., 1971;  Williams et al., 1971;  McCallister and
Logan, 1978, Green et al., 1978); however, little attention has been focused
on the origin of calcite in sediments.  In this paper,  secondary calcite
will refer to calcite precipitated after entry into the drainage system
whereas primary calcite will refer to calcite which enters  the drainage
system in particulate form.

     In the Maumee River drainage system of northwestern Ohio, Wall and Wilding
(1976) found that the silt fraction of suspended sediments  contained < 57»
calcite during periods of high and medium discharge but was dominated by
calcite during low flow regimes.  Semi-quantitative estimates of the calcite
content of the suspended silt fraction by X-ray diffraction techniques
revealed that the suspended silt fraction contained up  to 64°L calcite during
periods of low flow.  Calculations indicated that the stream water was
saturated with calcium carbonate; thus calcite precipitation could be expected.
Determination of calcite-dolomite ratios and optical observations of the
sediments with a light microscope confirmed the secondary origin of much of
the calcite.

     The equilibrium of calcite in an aqueous system open to the,.,atmosphere
is a function of the following:  calcium (Ca  ), carbonate (COo  ), bicarbonate
(HCO~3), hydroxide (OH"), and hydrogen (H+) activities  and the partial
pressure of carbon dioxide (PC02).  The effect of these parameters is evident
in the following reaction equation for the aqueous-carbonate system:
                         CaC03 + H2C03^Ca+2 + 2HCO"3
                                 U
                                 C02 + H20
As temperature increases, the Ksp for calcite decreases; thus calcite
solubility decreases.  The relative proportions of ^003, HCO~3, and CO^
in the carbonate system are a function of pH; 003   is  the dominant species
above pH 10, H2C03 below pH 6.5, and HCO~3 between those pH values (King, 1970).

     Studies of ionic concentrations in the Maumee River by Jones et al (1977)
revealed that concentrations of Ca   and HCO 3 are also highest during low
flow.  Thus there appeared to be a positive correlation between high ion
concentrations and calcite precipitation during periods of low flow.  Our
initial hypothesis in this investigation was that calcite precipitation is
initiated by increasing calcium concentrations during periods of low flow
due to decreasing dilution by surface runoff.  Alternate hypotheses for consid-
eration were that calcite precipitation is induced by (1) declining C02
contents in the  fluvial media,  (2) rising temperatures, or (3) increasing
pH of the fluvial media.  The objectives of this study were  (1) to investigate
the occurrence and stability of secondary calcite in stream discharge in the
Maumee River Basin and  (2) to determine the factors controlling the occurrence
of secondary calcite in suspended sediments.

-------
                                   -92-


3.63  Materials and  Methods.

                           Sample Collection

     During this study, solution parameters were periodically monitored and
suspended sediments collected at fourteen sites in the Maumee River drainage
system.  Data from only six of the sites will be discussed in this report,
since secondary calcite was never detected at the sites further upstream and
would thus contribute little to the testing of the hypotheses.  The locations
of the six sites are shown in Figure 1, page 31.

     A grab-sample approach was utilized to obtain samples from the streams.
A plastic bucket was lowered into the middle of the stream and allowed to
fill without disturbing the stream floor.  Approximately 40 liters of stream
water were collected when additional suspended sediment was needed for analysis.
Samples of the streams were collected on 20 different occasions during the
sampling period from September 10, 1975 to September 17, 1976.


                           Sediment Analysis

     Mineralogy;  An aliquot of stream water was filtered through a ceramic
plate to collect suspended sediments for X-ray analysis.  After drying, the
samples on the ceramic plates were irradiated with a Norelco diffractometer
equipped with a copper X-ray tube as described by Smeck et al   (1968).
Specimens were scanned from 26° to 32° 20 at a rate of 2° 20/min and plotted
linearly.  The time constant was one second and the counting rate was 500
counts/sec.  The peaks at 29.5° 26 and 30.8° 29 yield a qualitative estimation
of the content of calcite and dolomite, respectively.

     Carbonate Content;  Calcite and dolomite content and calcium carbonate
equivalent of the suspended sediment were determined volumetrically with a
Chittick apparatus (Dreimanis,  1962).  To obtain sufficient suspended sediment
for the determination of calcite and dolomite, the stream samples remaining
after aliquots  were removed for all other analyses was flocculated with MgCl2
or simply allowed to settle for 4-5 days.  The supernatant was then siphoned
off.  The concentrated suspended sediment was washed and dried by freeze-
drying or evaporation at room temperature.  Samples showing similar charac-
teristics as determined by X-ray and optical techniques  were composited for
analysis.

     Optical Microscopy;  Immediately upon return from sampling the streams,
50 ml of stream water were centrifuged to concentrate the sediments.  The
concentrated sediment was permanently mounted on a microscopic slide and the
abundance, sizesand shape of calcite crystals were observed with a Leitz
Ortholux petrographic microscope (250X magnification).

     Algae Count;  One-mi of each fluvial media sample was diluted with one
ml of silica gel ; then a 0.01 ml sample was spread over a one-cm  area of a
microscope slide and gently heat-fixed.  Fluvial media smears were stained  for

-------
                                  -93-


3 minutes with a solution consisting of 0.20 ml of 0.5 M sodium carbonate
buffer (pH 9.6), 1.1 ml of 0.01 M potassium phosphate buffer (pH 7.2),  1.1 ml
of 0.857o saline, and 1.0 mg of flurescein isothiocyonate (FITC).  The solution
was mixed at room temperature and used immediately.  The smear was then washed
in 0.5 M sodium carbonate buffer (pH 9.6) for 10 minutes and in  5% sodium
pyrophosphate for 2 minutes (Babiuk and Paul, 1970).  The smears were im-
mediately mounted in glycerol and observed with a Leitz-Ortholux microscope
equipped with a super-pressure mercury lamp-HBO 200W LI Green.  At a magni-
fication of 675X, the total number of algal cells per field was  recorded.
The average number of algal cells for 20 random fields was used  to calculate
the algal cell concentration (algal cells per 1 ml of fluvial media).


                           Solution Analysis

     Calcium;  Calcium was determined by atomic adsorption using a Varian
Techtron AA6 Atomic Adsorption Spectrophotometer.  A 25-ml filtered aliquot
of drainage water was diluted to 50 ml with 2.5 ml of 1,000 mg/1 lanthanum
solution and distilled water.  Standard solutions were prepared  to calibrate
the Spectrophotometer.  The concentrations (mg/1) of calcium were read directly
from the Spectrophotometer.

     Electrical Conductivity;  A conductivity bridge and cell was used to
determine the electrical conductivity of the stream sample (Richards, 1954).

     pH;  The pH was determined on a Beckman Expandomatic SS-2 model pH
meter.

     Bicarbonate;  HCOo concentrations in the water sample were  determined by
titration with dilute (0.01-0.03 N) H2S04.using bromphenol  blue as the end-
point indicator (Jackson, 1958).


3.64  Results and  Discussion

                   Characteristics of Fluvial Media

     Tributaries of the Maumee River tended to yield higher concentrations
of calcium and bicarbonate  and higher electrical conductivities than the main
stem.  Median concentrations of calcium and bicarbonate and electrical conduc-
tivities ranged from 68 to 74 mg/1, 135 to 216 mg/1, 520 to 721  umhos/cm,
respectively,for tributaries of the Maumee Riverswhereas the three sampling
sites in the Maumee River yielded values ranging from 48 to 57 mg/1, 152 to
181 mg/1, and 509 to 566 umhos/cm for these parameters, respectively (Table 1 ).
These values are similar in magnitude to those reported by Jones et al (1977)
for the Maumee River at Waterville which is farther downstream.

     Little  importance was placed on measurement of temperature and pH due
to the narrow range encountered in previous studies (Jones et al , 1977).
Thus initial   measurements of these parameters were not obtained; however, to
provide some indication of their magnitude and variability, temperature and pH
were monitored during August and September, 1976.  Both temperature and pH were

-------
      TABLE 1.   TEMPERATURE,  pH,  CALCIUM, BICARBONATE,AND ELECTRICAL  CONDUCTIVITY VALUES OF THE
                MAUMEE RIVER  DRAINAGE  SYSTEM (SEPTEMBER  1975  TO SEPTEMBER  1976)
Site

Calcium
(mg/1)

Bicarbonate
(mg/1)

Electrical
Conductivity
(y mhos /cm)
pH*


Temperature*
(°c)


Minimum
Maximum
Median
Minimum
Maximum
Median
Minimum
Maximum
Median
Minimum
Maximum
Median
Minimum
Maximum
Median
Tiffin R.
Tributary
30
86
72
112
285
206
2U3
632
520
7.9
8.1
8.0
17
22
20
Tiffin
River
32
95
7U
12k
286
216
259
Ilk
577
8.0
8.0
8.0
19
23
21
Maumee
River
St. Rd. 127
32
81
53
129
280
156
302
569
509
8.5
9-1
8.9
23
28
2U
Auglaize
River
53
71
68
117
166
135
551
826
721
8.3
8.7
8.5
19
23
22
Maumee
River
Independence
31
66
57
105
218
181
1*71
775
566
8.1
9.0
8.7
22
28
2U
Mauraee
River
Defiance
39
68
1*8
10 U
200
152
1*72
653
510
8.U
8.9
8.6
22
27
23
Temperature and pH vere only monitored during  August  and September of 1976.

-------
                                  -95-
slightly higher in the Maumee River than its tributaries.  Median temperature
and pH ranged from 20 to 22°C and 8.0 8.5, respectively,for tributaries of the
Maumee River,whereas these parameters ranged from 23 to 24 C and 8.6 to 8.9,
respectively,for the Maumee River during the two-month period (Table 1 ).
At any one site, pH and temperature showed less than 1 pH unit and 6°C var-
iation, respectively, during the two-month period.  Both temperature and pH
tend to favor precipitation of secondary calcite in the Maumee River rather
than its tributaries ,whereas the concentrations of calcium, bicarbonate, and
carbonate favor precipitation in the tributaries rather than the Maumee
River proper.


                     Calculated Calcite Equilibria

     Calcite equilibria in selected stream discharges representing high and
low calcium and bicarbonate concentrations found during this study were
calculated using the procedure outlined by Garrels and Christ (1965).  Ionic
strengths (I) were computed from electrical conductivity using the method of
Ponnamperuma _et al.  (1966).  Concentrations of Ca+2 and HCOo were converted
to activities by using activity coefficients obtained from Figures 2.15 and
4.5 in Garrels and Christ (1965^ which relate activity coefficients  to ionic
strength.  The equilibrium concentration of Ca+2, at standard temperature and
pressure, was calculated using measured bicarbonate concentrations and pH.
First  the activity of CCs  was ascertained by substituting the activities
of ff1" and HCOg into  the following equation and solving  for
                 (airf) (ac03'2)
                                = v     =10-10.3
                 	   KHC03   iu
                      HC03              _i-o
     The equilibrium concentration of Ca+^ was then calculated by substituting
aco    into  the following equation and solving for aCa+2 which is converted
to concentration:
                   (aCa+2)(aco-2)
                  	 = KCaCo3 = 10 8'3                        (2)
                     CaC03(s)

     The calculated  equilibrium values of Ca 2 concentrations are reported
in Table    2.  In  all  cases,  the calculated equilibrium Ca+2 concentrations
are considerably less  than measured  calcium concentrations}implying  that
measured Ca concentrations are not in equilibrium with  the pH and bicarbonate
concentrations  found in the Maumee River  system.

                                                                          +2
     Some of the difference  between  measured and  calculated  equilibrium Ca
concentrations may be  due  to  the  formation  of  ion-pairs involving Ca  .
Analytical  determinations  of  calcium concentrations will not only measure Ca
 ions but also  all  Ca ion-pair species;  thus these ion-pair  species must be
subtracted  from measured calcium concentrations  to  yield actual  Ca    ionic
concentrations  (Adams,  1971).  The concentrations of  CaHC03, CaCO°3, CaHPO°4,
 and CaOH+  ion-pairs  were calculated  using equlibrium  constants  reported by
Adams  (1971) and measured  Ca  concentrations.   The latter results in  the cal-
 culation  of maximum concentrations  for  each ion-pair  since some of  the measured
 Ca  is  associated with other  ion-pairs,which reduces its activity accordingly.
 To  calculate actual concentrations  of all the  ion-pairs, all the related

-------
                    TABLE  2.  CONCENTRATIONS OF Ca+2 , HCO^, CCj2, AND SIWJRAL ION PAIRfi  (CALCULATED AND MEASURED)
                             FOR SELECTED, REPRESENTATIVE SAMPLING SITES IN THE MAUMEE  RIVER BASIN


Site and
Sampling Date
Auglaize R.
8/11/76
Maumee R. at
S.R.-127
9/8/76
Tiffin R.
8/11/76
Tiffin R.
8/31/76
Maumee R. at
S.R.-127
8/11/76
Maumee P. at
S.R.-127
8/31/76
Maumee R . at
Defiance
8/11/76
Measured Stream Water Paramete
Hro
Ca "UU3
pH ttg/1 pCa mg/1 pHC03
8.7 70 2,76 Ilt9 2.6l

9.1 32 3.10 lltl 2.6U


8.0 72 2.7lt 165 2.57

8.0 75 2.73 286 2.33

8.9 63 2.80 l61t 2.57


9.1 to 3.00 151 2.61


8.9 68 2.77 13^ 2.66


>rs Calculated Equilibrium
Concentrations
Contribution
of ion-pairs
ion-pairs §


0

0


0

0

0


0


0



I*
.013

.008


.010

.011

.009


.009


.009



pCa**
3.83

It. 27


3.20

3.U3

It. 12


It. 28


It. 03



PC02 +
0.00032

0 .00012


0.00179

0 .00303

0.00023


0 .00013


0 .00019




0

0


0

0

0


0


0



Pco tt pCaHCOl)
.00378 U. 37

.00185 14.67


.00522 U. 28

.015142 It.0l4

.00lt76 It. 32


.00250 It. 56


.00337 >4.38



pCaCOjSS
It. 23

ii.oS


It. 82

14.59

3.95


3.99


It. 01



pCaHPo'n
5.6it

5.93


5.60

5. 60

5.65


5,85


5-62


i
pCaCH+
6.89

6.78


7-55

7.55

6.70


6.70


6.67


to
Pa — Ca^ni ml

6.5

ilt.it


6.c

8.0

10.7


13. B


9-2


 * Ionic strength (I) calculated from measurements  of electrical  conductivity  (Ponnamperuma et_ jil. ,  1966).
** Calculated using measured values  for pH and HCOs  concentrations.
 t Calculated using measured values  for pH and HCOa  concentrations .
tt Calculated using measured values  for Ca, and HC03  concentrations.
 § Calculated using equilibrium constants reported, by Adams  (1971).
§§ Carbonate concentrations  were estimated by using  measured values  for pH  and HC03  and the equilibrium constant,
 M HPO^2 was taken to be O.It mg/1 which is the median reported  by Jones et  al.  (1977)  for the  Maumee River.
1111 The concentrations of all four ion pairs vere  summed  and  expressed as  a  percentage  of difference  between  the measured and  calculated
   Ca concentrations.

-------
                                 -97-


equations must be solved simultaneously; however,since our only interest was
in maximum concentrations of ion-pairs, such a rigorous mathematical solution
was not necessary.  The calculated maximum concentrations of the selected
ion-pairs are reported in Table 2.   The concentrations of all four ion-pairs
are low (lO~4 to 10~7 moles/I),which agrees with the contention of Garrels
and Christ (1965) that the effect of ion-pairs in natural waters less con-
centrated than sea water is relatively slight.  The contribution of ion-
pairs to the difference between measured and calculated equilibrium Ca+^
concentrations is also reported in Table 2.  Note that, in all cases, ion-
pairs account for less than 15$ of the difference.  Thus the formation of
ion-pairs can not account for the apparent oversaturation of the fluvial
media with calcium.

     The partial pressure of C02 in the streams which, is required for equili-
brium with the measured pH and bicarbonate concentrations,was calculated using
the following two equations:
                                               = 10-6-^                    (3)
                          PC02


The concentrations and activity coefficients of HCO~ and H+ were substituted
into Equation 3 which,was then solved for the activity of H COo.  That value
was then placed into Equation h ,which was solved for PCO .   The values ob-
tained for the partial pressure of C02 in the streams (T§ble 2) approximate
atmospheric C02 partial pressure except for the two Tiffin River samples ,
which exhibit values which are an order of magnitude higher than atmospheric.
Due to respiration of aquatic organisms, decomposition of organic components
in the water, and a relatively slow C02 exchange process between the river and
the atmosphere, the C02 content of the river water should be greater than in
the atmosphere.  Atmospheric C02 exchange with the river is particularly slow
in sluggish streams characteristic of the Maumee River Basin.   If the H+ ion
concentration is a function of dissolved C02 content and the C0.2 content is
higher than atmospheric, then the pH of the river should be considerably
lower than that reported in Table 2.  This implies that the pH of the stream
is more closely related to ion exchange and additions of strong bases than
to the HgiCO-j content of the stream.  Increasing pH's on moving downstream
exhibited by data in Table 2 and by Jones et al   (1977) tend to substantiate
this hypothesis,since the quantity of industrial and municipal wastes and
effluent increases on moving downstream.

     Consequently, the partial pressure of C02 in the river at the selected
sites was also calculated ,assuming that calcite is in equilibrium with H CCU
(dissolved C02) in the river rather than with pH using the  following equation:

-------
                                   -98-


      Equation  5 was solved  for  the activity of lUCC^ >which was then sub-
 stituted  into  Equation 4,which was solved for PC09-  The C02 partial pressures
 in equilibrium with Ca Z and HCO^ are much higher than atmospheric (Table  2 ),
 ranging up  to  1.57», but are believed to be quite reasonable for the Maumee
 River system.  Thus the Ca-HC03-H2C03-C02 system may be more nearly at equili-
 brium than  the H-HCO-j-I^CC^-CC^ system.  This would also explain why calcite
 is not continually being precipitated in the Maumee River system.  In either
 case,  it  is apparent  that the calcium concentrations are not in equilibrium
 with  the  pH exhibited by the streams.


                    Occurrence of Secondary Calcite

      X-ray diffraction analyses of all suspended sediments collected in this
 study were conducted  to give a qualitative indication of the presence of
 secondary calcite.  Ratios  of calcite to dolomite peaks were used to distinguish
 between primary and secondary calcite.  Calcareous glacial deposits in the
 lake  basin exhibit calcite/dolomite ratios of approximately 1.8 (Smith and
 Wilding,  1972); thus primary carbonates produce calcite and dolomite peaks
 of approximately equal intensity as illustrated by the sample from the Auglaize
 River  in  Figure  2«  Since  dolomite is less soluble than calcite and the pre-
 cipitation of  dolomite in fresh water has never been reported, a ratio
 of calcite-to-dolomite of greater than 1:8 is indicative of secondary precipi-
 tation of calcite in  the stream,as illustrated by the two Maumee River sites
 in Figure  1.  In the study by Wall and Wilding (1976), the presence of
 secondary calcite produced  calcite peaks 10-15 times the intensity of dolomite
 peaks.  Only one suspended-sediment sample was void of crystalline calcite
 and dolomite (a sample from St. Rd. 127-Maumee River collected on March 10,
 1976)  throughout the sampling period.  This indicates small amounts of crys-
 talline calcite and dolomite are a basic ingredient of the suspended sediments.
 Erosion of stream banks (which contain calcareous glacial  till and lacus-
 trine  deposits) contributes primary calcite and dolomite.  A sample from St.
 Rd. 127-Maumee River collected on April 13,  1976 contained the first evidence
 of secondary calcite found  in this study.   The calcite peak was more distinct
 and 10 times greater in intensity than the dolomite and calcite peaks recorded
 from  earlier samples.   During the study period only suspended-sediment
 samples from the Maumee River at St.  Rd. 127,  Independence, and Defiance
 produced high-intensity calcite peaks;  however,  intermediate calcite peaks
 (2-3  times the dolomite peak)  were observed  for samples collected from the
 Tiffin River and Tiffin River tributary (Figure  1 ).

     In the Maumee River,  the occurrence and quantity of secondary calcite
were variable, depending on sampling location and date.   Figure  2 gives  Ca+2
 concentration and corresponding secondary calcite content (estimated  by X-ray
 analysis)  for each sample during the summer  months (June-September,  1976).
 Several observations  are evident:   (1)  The St.  Rd.  127-Maumee River site was
 the most consistent in exhibiting secondary  calcite.   Except for  the  sample
collected  on July 1,  1976,which was collected  immediately following a heavy
precipitation event, medium to high amounts  of secondary calcite  were ob-
 served throughout the  summer.   (2)  The Defiance-Maumee River site,  downstream
 from St. Rd.  127 approximately ten miles,  was  next most consistent in ex-
hibiting secondary calcite.   (3)  The  Independence-Maumee River site,  five

-------
             -99-
                                     3.34A Quartz
                                     3.04A Calc'rte
                                       2.89 Dolomite
                                     Maumee River
                                     (St. Rd. 127)
                                     Maumee River
                                     (Independence)
                                     Tiffin River
                                     Auglaize River
                2.89 3.04     3.34
           d-spacing (A)
Figure 1. X-ray diffractograms of stream-suspended sediments

-------
   80r
   70
o> 60
 03
O
    50
    40
    30
    25
O  No secondary calcite

®  Med. secondary calcite

0 High secondary calcite
                                                             Maumee R.
                                                             Auglaize R.
                                                             Maumee R.
                                                             Maumee R.
                                            St. Rd. 127
                                       	Independence
                                       	Defiance
                                                                                         O
                                                      \           £
                                                \--n-\-	O  f
                    July
15
                   Aug.
                    1976
15
Sept.
15
             Figure 2. Calcium levels in the stream and occurrence of secondary calcite.

-------
                                 -101-


miles downstream from Defiance, was most variable in secondary calcite content.
The addition of industrial and municipal wastewater from Defiance and dilution
by the Auglaize River may change the solution parameters enough to influence
the precipitation of secondary calcite at this site.  (4) The Auglaize River
showed no evidence of secondary calcite throughout the sampling period.

     After April 13, 1976)when high intensity calcite peaks were first observed
by X-ray analysis, microscopic observations of the suspended sediment were
made to examine primary and secondary calcite crystals and to describe their
morphology.  Differentiating criteria for distinguishing between primary and
secondary calcite crystals were made by observing samples with and without
abundant secondary calcite as revealed by X-ray analysis, and noticing morpholo-
gical differences between the two types of crystals.   Secondary calcite
crystals were distinctly different than primary calcite crystals in two aspects.
First, secondary calcite crystals ranged from 2-20 urn in size (with the majority
of crystals ranging from 2-12 urn), whereas primary calcite crystals ranged from
10-35 urn (with the majority of crystals between 15-25 urn).  The latter is
consistent with a study of carbonate particle-size distribution in tills of
Western Ohio by Smeck et al  (1968),   which revealed that calcite attains a
maximum in the fine silt (2-20 um) fraction.  Second, secondary calcite crystals
appeared to be irregular clusters of many individual smaller crystals, whereas
primary calcite crystals occurred as angular, dense, discrete units.   Secon-
dary calcite crystals in suspended sediment collected from St.  Rd. 127-Maumee
River tended to be smaller than secondary calcite crystals farther downstream
at Independence and Defiance,which indicates the possibility that smaller
crystals dissolve at the expense of large crystal growth.  Large populations
of biological cells were apparent on the slides of suspended sediment.   A
variety of different species of algae and diatoms were noticed especially
from samples collected at St.  Rd. 127-Maumee River.   This observation prompted
the hypothesis that algae control the PcOo in stream water,which in turn
controls the carbonate equilibria.

     Quantitative determinations of calcite and dolomite contents in the
suspended sediments  were determined on composited suspended sediment samples
(Table  3).  The calcium carbonate equivalents of most of the composited
suspended sediment samples were less than 15% with calcite/dolomite ratios of
less than 1.5,which is in agreement with data reported by Wall and Wilding
(1976) for both medium and high-flow conditions in the Maumee River system.
Since these values are very similar to the calcium carbonate equivalent and
calcite/dolomite ratios of calcareous glacial deposits in the basin,  these
samples may contain primary carbonates contributed to the streams by bank
erosion.  However>a composite sample of suspended sediments collected at
State Route 127 on the Maumee River from July 20 to August 18 exhibited a
higher calcium carbonate equivalent (38.9%)  and calcite/dolomite ratio (6.7)
than calcareous glacial deposits*which is indicative of calcite enrichment
of the sediments by calcite precipitation in the river.   Suspended sediment
samples collected at this location from July 20 to August 18 also exhibited
high intensity calcite peaks by X-ray diffraction.

-------
TABLE 3-   CALCIUM CARBONATE EQUIVALENT AND CALCITE/DOLOMITE RATIOS
          OF COMPOSITE SUSPENDED-SEDIMENT SAMPLES
Site
Tiffin R. Tributary
1!
Tiffin River
it
St. Rd. 127-Maumee R.
tt
it
Auglaize River
Independence-Maumee R.
Date
2/19
7/1
2/19
7/1
2/19
7/1
7/20
2/19
2/19
Sampled
- U/13
- 8/31
- U/13
- 9/8
- U/13
- 7/15
- 8/18
- 9/8
- U/13
Calcite
i
I
7
1
6
3
13
33
2
1
%
.U
.1
.6
.0
.8
.3
.5
.3
.6
Dolomite
1
5
2
U
3
7
5
2
2
.8
.2
.1
.3
.2
.U
.0
.U
.5
Ratio
Cal/Dol
0
1
0
1
1
1
6
1
0
.8
.U
.8
.U
.2
.8
• 7
.0
.6
CaC03 Equiv.
(
3
12
3
10
7
21
38
U
U
%
• U
o
.7 ?
.8
.7
.3
.3
• 9
.9
.3

-------
                                -103-
               Parameters Controlling the Precipitation
                         of Secondary Calcite


     In this study, the precipitation of secondary calcite was  only observed
in the Maumee River and only during the summer months.   In an earlier study
in the Maumee River drainage system (Wall and Wilding 1976),  the precipita-
tion of secondary calcite was only observed during low flow.   High Ca+2
concentrations during low-flow periods were considered the controlling factor
in the precipitation of calcite.   From the data presented in  Figure 2,
which relates Ca+2 concentrations to semi-quantitative estimates of the
abundance of secondary calcite, it is evident that there is considerable
variation in Ga+2 concentrations  (20-50 mg/l) during the three-month-period
with no discernable relationship  between Ca+2 concentrations  and the occur-
rence of secondary calcite.  In fact, the Auglaize River generally contained
higher Ca+2 concentrations than the Maumee River,but no precipitation of
secondary calcite was observed in the Auglaize River (Figure   2 }.  In the
Maumee River, at the time of the  highest recorded Ca+2 concentration during
the summer months (78 mg/l on July 1, 1976), secondary calcite  was not found.
Furthermore, many of the observations of abundant secondary calcite occurred
during periods of low calcium concentrations.  Thus it is concluded that Ca
is not the primary parameter controlling calcite precipitation, since at even
the lowest recorded Ca+2 concentrations (30 mg/l), calcite precipitation
was observed.

     During the summer months, temperature and pH remained relatively
constant-, while secondary calcite  appeared ( June 23) and disappeared  (July l)
within short time periods.  Also, temperature and pH values recorded in the
Auglaize and Maumee Rivers were almost equal throughout the summer, but
secondary calcite was never observed in the Auglaize River (Table  3).  Thus
neither temperature nor pH appears to be an important parameter controlling
secondary calcite precipitation in these fluvial systems.

     Bicarbonate concentrations not only exhibited considerable variation
during the year (Table  1 ) but also showed considerable variation between
sampling periods (Figure  3).  However, there is not an evident relationship
between bicarbonate concentrations and the occurrence of secondary
calcite (Figure  3).

     As the study progressed, microscopic observations of slides of suspended
sediments revealed a relationship between the abundance of algal cells on
the slide and the presence of secondary calcite.  Consequently, algal popu-
lations were measured on a series of sediment samples collected on September 8.
Algal populations ranged from 0.62 x 105 (Tiffin River) to 11 x 1C)5 (St.
Rt. 127-Maumee River) cells/ml (Table   k).  Samples collected at State
Route 127 on the Maumee River contained 2-3 times more algal cells than any
other site and was the only site  on that date to show significant quantities
of secondary calcite.  As a result of these observations, it is speculated
that during periods of algal bloom, the partial pressure of C02 in the river
is reduced by photosynthesis to levels which will result in the precipitation

-------
   225
O)
O
c
CD

O
.Q

CD
O

CO
    150
   125
   100
o No secondary calcite

© Med. secondary calcite

• High secondary calcite
Maumee R.__	  St. Rd. 127

Auglaize R.	

Maumee R	Independence

Maumee R	Defiance
                                                                \
                                                                 \
                                                                   \
                                                                     \
                                                                      \
                                                                       Sept.
                                                           15
      Figure 3. Bicarbonate levels in the stream and occurrence of secondary calcite.

-------
                                  -105-
TABLE k.  ALGAL POPULATIONS FROM STREAM SAMPLES COLLECTED ON SEPTEMBER 8,  19?6
                                               Algae Population
                      Stream Site              (cells/ml x io5)
               St.  Rd. 127-Maumee River              11.00

               Tiffin R.  Tributary                    2.17

               Tiffin River                           0.62

               Defi an ce-Maumee River                  U.19

               Auglaize River                         U.3^

               Independence-Maumee River              2.U8

-------
                                -106-
of calcite.  As revealed by data in Table 2, the C02 partial pressure would
not even have to be reduced to atmospheric CC>2 partial pressure to induce
calcite precipitation.  Additional evidence for this mechanism was observed
during late June and early July at St. Rd. 127-Maumee River.  Samples collected
in late June contained high concentrations of algal cells and secondary
calcite; however, samples collected one week later following a heavy rain
contained neither algal cells nor secondary calcite.  Evidently the increased
discharge not only flushed out the system but the increased turbidity of
the fluvial media retarded algal photosynthesis.  It seems reasonable to
expect the CC>2 partial pressure to increase with depth in the stream during
algal blooms, due to decreasing utilization by algae.  The C02 content of
the lower part of the stream will be maintained at relatively high levels
due to decomposition of organic components and a slow CC>2 equilibration
rate.  Thus calcite which is precipitated near the surface during algal blooms
may redissolve as it settles toward the stream bottom.  The hypothesis is
supported by examination of the bottom sediments at these sites by Wall et al
(1978).  By studying the calcium carbonate content and calcite/dolomite ratios,
they concluded that calcite had been dissolved from the bottom sediments
during the transport process or on the stream bottom.  Thus calcite which is
precipitated in the stream is not being contributed to the bottom sediments.


3.65  Conclusions

     Though secondary calcite is not ubitiquous in the Maumee River drainage
system, its occurrence has been established; however, calcite equilibria in
the stream system are quite dynamic.  Secondary calcite may precipitate and,
subsequently, dissolve in relatively short periods of time in the stream and
seems to be related to algal blooms.  To maintain the relatively high Ca^+
concentrations present in the drainage system, C02 partial pressure must
be considerably higher than atmospheric.  Algal blooms reduce the CC^ content
resulting in the precipitation of calcite.  Thus it is concluded that C02
fluctuation rather than changing Ca   concentrations is the factor initiating
calcite precipitation and dissolution in the Maumee River stream system.


3.66  Literature Cited

1.  Adams, Fred.  1971.  Ionic Concentrations and Activities in Soil Solutions.
    Soil Sci. Soc. Amer. Proc.  35:420-426.

2.  Babiuk, L.A. and E.A. Paul.  1970.  The Use of Flurescein Isothiocyonate
    in the Determination of the Bacterial Biomass of Grassland Soil.  Can.
    J.  Microbio.  16:57-62.

3.  Dreimanis, A.  1962.  Quantitative Gasometric Determination of Calcite
    and Dolomite by Using Chittick Apparatus.  J. Sed. Pet.  32:520-529.

4.  Garrels, R.M. and C.L. Christ.   1965.  Solutions, Minerals, and
    Equilibria.  Harper and Row, New York.

-------
                              -107-
 5,   Green,  Dan B.,  T.J.  Logan,  and N.E.  Smeck.   1978.   Phosphate Adsorption-
     Desorption Characteristics  of Suspended Sedimants  in the Maumee River
     Basin of Ohio.   J.  Environ.  Qual.   7:208-213.

 6.   Jackson, M.L.   1958.  Soil  Chemical Analysis.   Prentice-Hall, Inc.,
     Englewood Cliffs, New Jersey.

 7.   Jones,  L.A., N.E. Smeck,  and L.P.  Wilding.   1977.   Quality of Water
     Discharged from Three Small Agronomic Watersheds in the Maumee River
     Basin.   J. Environ.  Qual.  6:296-302.

 8.   King, D.L.  1970.  The Role of Carbon in Eutrophication.  J. Water
     Pollut. Cont.  Fed.   42:2035-2051.

 9.   McCallister, D.L. and T.J.  Logan.   1978.  Phosphate Adsorption-
     Desorption Characteristics  of Soils and Bottom Sediments in the Maumee
     River Basin of Ohio.  J.  Environ.  Qual.  7:87-92,

10.   Ponnamperuma,  F.N., E.M.  Tianco, and T.A. Loy.  1966.  Ionic Strength
     of the Solutions of Flooded Soils from Specific Conductance.  Soil
     Science.  102:408-413.

11.   Richards, L.A. ed.   1954.  Diagnosis and Improvement of Saline and
     Alkali Soils.   Salinity Laboratory, USDA Handbook No. 60, p. 160.

12.   Shukla, S.S., J.K. Syers, J.D.H. Williams, D.E. Armstrong, and
     R.F. Harris.  1971.  Sorption of Inorganic Phosphate by Lake Sediments.
     Soil Sci. Soc. Amer. Proc.   35:244-249.

13.   Smeck, N.E., L.P. Wilding,  and N. Holowaychuk.  1968.  Genesis of
     Argillic Horizons in Celina and Morley Soils in Western Ohio.  Soil
     Sci. Soc. Amer. Proc.  32:550-556.

14.   Smith, Horace and L.P. Wilding.  1972.  Genesis of Argillic Horizons
     in Ochraqualfs Derived from Fine-Textured Till Deposits of Northwestern
     Ohio and  Southeastern Michigan.  Soil  Sci. Soc. Amer. Proc.  36:808-815.

15.   Wall,  G.J.  and L.P. Wilding.  1976.  Mineralogy and Related Parameters
     of Fluvial  Suspended Sediments in Northwestern Ohio.  J. Environ.
     Qual.   5:168-173.

16.   Wall,  G.J., L.P. Wilding, and N.E. Smeck.  1978.  Physical,  Chemical
     and Mineralogical Properties of Fluvial Unconsolidated  Bottom  Sediments
     in Northwestern Ohio.  J. Envorn. Qual.  7:319-325.

17.   Williams, J.D.H., J.K. Syers, R.R. Harris, and D.E. Armstrong.   1971.
     Fractionation of Inorganic Phosphate in Calcareous Lake  Sediments.
     Soil Sci. Soc. Amer. Proc.   35:250-255.

-------
                                 -108-
3.7  Heavy metals in Maumee River Basin water.  soil> and  sediment
3.71  Abstract

     Heavy metal content of streamwater,  groundwater,  stream-bottom sediments,
and soils were measured in the Maumee River Basin.   Metal levels  were low
in all cases and groundwater appears to contribute  a significant  percentage
of the total heavy metal tributary load.   Point sources  were identified
as industrial discharges throughout the Basin  but  had no detectable effect
on downstream water or sediment metal concentrations.

     A point-source chromium discharge was monitored above and  below Lima,
Ohio.  Chromium and other metals associated with the point source and other
sources in the Lima area were adsorbed by bottom sediment and taken up by
rooted macrophytes.  This metal assimilation by sediment and aquatic plants
could pose a long-term hazard to the local stream environment.

-------
                                -109-
3-72  Introduction

     The Maumee River Basin is primarily of agricultural land use with a
number of urban centers on tributaries and the Maumee River itself.  Heavy
metals draining the Basin may be from industrial and municipal point
discharges, groundwater, and attached to sediment.  Little if any previous
information on heavy metals in the Basin is available.

     Point sources entering the stream as soluble metal may be diluted
by upstream drainage, be adsorbed and retained by stream sediment  or be
assimilated by stream organisms.

     The objectives.,of this study were:

     1.  To determine background levels of heavy metals in streamwater,
         stream sediments, groundwater, and agricultural soils.

     2.  To determine the fate of a point source discharge of heavy metal
         to a stream.
3.T3  Methods

            Metals in water, sediment, soils and bedrock

     Stream water and bottom sediments were sampled at 20 locations on four
different occasions over a 2*§ year period (Figure   1).  Sites were chosen
along rivers where industrial and municipal effluent was absent or greatly
diluted; these sites are listed in Table  1.  Samples were taken 10-21-75,
1-20-76, 7-10-76,and 1-29-77.  Samples were taken by hand and uniformly
across the stream cross-section wherever possible.  Water samples and
sediment samples (after digestion) were analyzed by flame atomic absorption
spectroscopy.

     Metals in bottom sediments were digested with 1:1 cone. KNCg and
cone. HC1.  Samples were digested at *J 90° C for two hours, the digest
removed, the samples redigested and the digests combined for analysis.
The same procedure was used to determine metal content of surface (0-15 cm)
soils from the Defiance County monitored agricultural watersheds and plots
at Hoytville (Figure  1, page  4  )•

     In addition, total metals in bedrock samples within the Basin (sampled
from a number of quarries throughout the area) were determined, aswere dissolved
metals in groundwater from 27 wells in Defiance County.  These values are
intended for comparison only.

                 Metals from point-source discharge

     Lima, Ohio was  chosen as the study site to determine the heavy metal
variability  in the stream environment  because the Ottawa River is easily
sampled at this point and a well-defined heavy-metal point  source  (primarily
chromium)  exists downstream from Lima  at the Vistron Corporation.
The river at Lima is representative of the  stream  conditions  in most of
the Basin, in that it drains both ground moraine and end moraine deposits.

-------
                             -110-
Figure 1. Sampling sites used in determining background concentrations of
         heavy metals.

-------
                                 -Ill-
Table  1.  The locations of sample sites used in studying heavy metal
           background concentrations.
S amp1e
Number	Sample Site	

10-1                  St. Marys River at Rt. 33 (U.S.)
10-2                  St. Marys River at SR 118
10-3                  St. Marys River at Decatur gauging station
10-4                  St. Marys River in Ft. Wayne
10-5                  St. Joseph River in Ft. Wayne
10-6                  Tributary to St. Marys River near Ft.  Wayne
10-7                  St. Joseph River near Hursh
10-8                  St. Joseph River, Newville gauging station SR 249
10-9                  St. Joseph River, at U.S. 6 near  Edgerton
10-10                 St. Joseph River at Rt. 20 (U.S.)
10-11                 Maumee River at New Haven gaiging station
10-12                 Maumee River at Antwerp SR 49
10-13                 Maumee River and Auglaize River confluence
10-14                 Tiffin River near Evansport
10-15                 Maumee River at Independence Dam
10-16                 Maumee River at Napolean
10-17                 Maumee River at Waterville
10-18                 Blanchard River near Dupont
10-19                 Auglaize River at Ottoville
10-20                 Blanchard River at Ottawa

-------
                                -112-

     Sample sites were established on the Ottawa River upstream and downstream
from Lima.  The upstream site is characterized by low-relief farmland with
no industrial discharge.  At this point heavy metals  could enter only by
surface runoff or ground water discharge, i.e. diffuse-source conditions.
The downstream site is in an industrial park where waste discharge into
the river is common.   The river was sampled about 150 feet downstream from
one such point source, the Vistron Corporation.  At each location 16 water
and bottom sediment samples were taken.

     An 18 x 18 feet grid pattern was sampled at each site ( Figs.  2 and    3).
Water and bottom-sediment samples were collected at the nodes numbered 1-16.
Nodal numbers increase away from the bank and in the downstream direction
with a 6-feet spacing between them.

     Water samples were collected about:5 cm below the water surface and
stored in polyethylene bottles.  Unfiltered water samples were analyzed
for Cr, Cd, Pb, Sr, and Zn using a Perkin-Elmer 303 AA spectrophotometer.
The precision for this instrument under these conditions is reported in
Table  2  in terms of coefficient of variation for duplicate aliquots of
water and sediment extracts.  The bottom sediment was taken with a piston
core sampler.  Cores were stored in polyethylene wrapping with only the
upper 2 inches being analyzed for metals.  The metals were extracted from
the sediment with 0.05IJ HCL + 0.025J! H2SO^  , since the study was concerned
mainly with easily extractable metals.


3-7^  Results

              Dissolved metals in stream and groundwater

     Stream water at 20 sampling sites throughout the Maumee Basin was sampled
10-21-75, 1-20-76, 7-10-76 and 1-29-77.  Nickel and zinc were detected most
frequently and Ni gave the highest concentrations.  Strontium was included
for comparison.  There appeared to be no seasonal effect on heavy metal
concentrations,but this is a tentative conclusion considering the low
frequency of sampling.  No individual site  appeared to be higher than others
for any of the metals, not surprising since these sites represent diffuse
sources only.  Mean dissolved metal  concentrations are given in Table  3
together with mean values  for 27 test wells.   Groundwater sources were
generally higher than stream water.  Based  on  the analysis of groundwater
contribution to total flow, it would appear that groundwater is the major
source  of dissolved metals in the Maumee.   Waterville groundwater accounted
for 38% of the total  flow  in 1976  and,given the  concentrations given  in
Table   3 , the contribution of groundwater  to  the amounts of each dissolved
metal  discharged from the  Maumee can be  estimated  (Table    3 ).  The  data
show that groundwater contributes most of the  dissolved metals except  cadmium.

-------
                     -113-
                  Downstream
              CHANNEL X-SECTION
Figure 2. Sampling grid pattern and channel cross-section
of the site of the variability upstream from Lima, Ohio.

-------
                       -114-
                     UPSTREAM

              CHANNEL X-SECTION
Figure 3. Sampling grid pattern and channel cross-section
of the site of the variability downstream from Lima, Ohio.

-------
                 -115-
Table  2.  Coefficients of variation (CV) of
          precision for elemental analysis by
          AA spectroscopy.
Element                  C.V. of Precision (%)
                          water      sediment

  Cr                       0.02        0.80
  Cd                       0.01        0.09
  Pb                       0.04        0.91
  Sr                       0.06        0.76
  Zn                       0.10        0.33

-------
 Table
3 .
                                 -116-
Background concentration of heavy metals in the Maumee River Basin
and in groundwater (1975-77)
Streamvater Groundwater
Background
11 f /Tffl
Percent of total
discharge as
groundwater*

Cd
Co
Cr
Cu
Ni
Pb
Zn
Sr
0
0
0
0
0
0
0
0
.011
.010
.003
.003
.082
.020
.021
• 570
(20.
(21.
(20.
(16.
(77-
(28.
(85.
(100
o)t
3)
0)
3)
5)
8)
0)
-0)
0.
0.
0.
0.
0.
0.
0.
1.
009
080
098
250
950
09k
95 k
650
33.
83.
95.
98.
87.
Ik.
96.
6k.
k
1
2
1
7
2
5
0
              *  Assumes 38% of total discharge in groundwater

              t  Percent of samples where metal was detected
     Heavy metals in watershed soils and Maumee River bottom sediments

      Table 4  gives the mean heavy metal concentrations of the surface soil
 horizons of the Defiance county and Hoytville sites  and bottom sediments
 from the 20 metal sampling sites in the Maumee.   Metal content of limestone
 bedrock of the area is included for comparison.   Values given in Table 4
 are for aqua regia extraction.   This procedure does  not extract all the
 structural metal, i.e. metal held within the crystal lattice of minerals,  but
 it does extract those compounds that would be environmentally active.   Of
 the metals, cadmium has the lowest concentration and  zinc the highest in both
 soil and sediment.  Metal concentrations in both soil and sediment appear  to
 reflect bedrock composition somewhat.   Only cobalt appears to be enriched
 in the sediment compared to soil while all other metals are considerably lower
 in the sediment.  Variability was remarkably low and there appeared to be
 little regional differences.  In addition, metal concentrations were not
 correlated with each other.  It should be reemphasized that the sampling sites
 were chosen to reflect background metal levels and were not close to known
 point sources.  While our estimates of sediment-bound metals is underestimated
 because our extraction procedure does not extract total metal, the data still
 show that dissolved metal accourts for a high percentage of the total load.
 Taking into account our findings that the groundwater accounts for a high
 percentage of the dissolved load, it would appear that metals in groundwater
are the major source of metals leaving the Maumee.

                      Point-source metal discharge

      Concentration means and ranges and coefficients of variation were
 determined for each metal at both locations for water and bottom sediment
 (Table    5).  The high concentrations of all elements except strontium in
 the downstream water and sediment samples reflect the point-source loading
 at Lima.  The major component of this discharge is chromium., and this is
 reflected in the values in Table  5.   Chromium in excess of O.k mg/1 persisted

-------
Table  4 .Concentrations  of  heavy metals in Maumee River Basin soils, bottom sediments ,


Cd 0
Co 1
Cr 12
Cu 9
Ni 25
Pb 21
Zn 41
Sr

Range
.10-0.70
.80-2.30
.00-13.80
.60-27.80
.80-42.00
.60-29.40
.30-69.60

Soils
Mean
0.35
1.98
15.30
20.20
33.75
25.20
49.15

Sediment
S.D.
0.26
0.22
4.17
8.62
6.63
3.23
13.65

Range
0
4
0
4
6
3
6
50
»"t>< t>
.04- 0.39
.25-14
.72- 2
.38-10
.42-16
.84-10
.95-24
.10-93
.31
.54
.11
.89
.70
.68
.60
Mean
0.
9.
1.
6.
11.
7.
15.
71.
15
11
55
49
21
33
77
77
S.D.
0.09
2.26
0.46
1.27
2.39
1.55
3.32
7.89
Bedrock

1.
1.
2.
8.
34.
33.
250.
57.

94
27
63
52
12
50
50
80

-------
                              -118-

Tablc  5.  Elemental analysis of 16 grid samples from each site
          (upstream and downstream water and bottom sediments).
Element           Concentration                                  C.V.
                  Mean                        Range               %

                Upstream Water (mg/1)

  Cr              0.12                     0.10 - 0.18           4.1
  Cd              0.007                    0.005 - 0.200         7.0
  Pb              0.02                     0.01 - 0.02           5.4
  Sr              0.782                    0.750 - 0.800         6.3
  Zn              0.018                    0.010 - 0.030         9.1
                                                        Mean     6.4

                Downstream Water (mg/1)

  Cr              3.93                     0.10 - 6.98           8.1
  Cd              0.060                    0.050 - 0.070         1.0
  Pb              1.10                     1.08 - 1.13          .1.5
  Sr              0.754                    0.640 - 0.820         3.3
  Zn              0.912          ,          0.880 - 0.980         1.4
                                                        Mean     3.1

                Upstream Bottom Sediment (ug/g)

  Cr              1.69                     1.48 - 2.10           9.3
  Cd              0.94                     0.52 - 4.38          11.2
  Pb              1.1                      0.5 - 1.5             6.2
  Sr             26.9                     17.7 - 29.3           10.6
  Zn              1.30                     1.09 - 9.38          13.5
               ;                                         Mean    10.1

                Downstream Bottom Sediment (ug/g)

  Cr              195                      176 - 230            51.4
  Cd              1.20                     0.76 - 12.10         19.6
  Pb             19.3                     15.6 - 22.6            9.5
  Sr             29.7                     19.8 - 34.7           31.6
  Zn             12.10                    17.60 - 26.30         13.5
                                                        Mean    25.1

-------
                                 -119-
four miles downstream from the discharge tunnel,  hugging  the  river  bank
of the source.  Under diffuse source conditions,  concentration  variability
in water was much lower in the higher velocity (main flow)  portion  of  the
stream than near the bank (Fig. 4).   The difference is  more than  likely
a direct result of mixing in the more turbulent stream  water.   This suggests
that sampling under diffuse source conditions should be carried out in the
most rapidly flowing part of the stream.  Strontium concentration is high
throughout the basin since the area is underlain by Silurian  and  Devonian
carbonates containing
     The variability in the metal content of the downstream water was  due
to effluent mixing and dilution.  Chromium was the best element to demonstrate
mixing (Fig. 5).  In grid observation numbers 1-6, the variability was
low because the effluent was still highly concentrated.  At grid numbers
7-10, mixing occurs and variability was high.  Outside of the mixing zone
nodes 10-16, both concentrations and variability were low.

     In contrast to the easily recognized trend of the water, the bottom
sediment at the same nodes had a highly variable chromium concentration
(Fig. 6).  The coefficient of variation for these data is 51.4 percent.
The variation is attributed to shifting of sediment during high flow,  and
differences in sediment distribution and grain size.  The coefficient of
variation suggests taking a large number of samples to approach the mean
concentration.

     In order to compare the data in terms of sample numbers needed to
maintain a precision of 10 percent on either side of the mean, limit -of-
precision curves (95 percent confidence interval) were constructed for
each element and graphed for each location, water sample, and bottom sediment.
In Figure 7 such a graph is presented for the upstream bottom sediments.
Zinc is  the most variable, with  lead being the least variable; precision
increases as sample number increases.  Similar graphs were obtained at
the  downstream  location; however, downstream, chromium was  the most
variable  element as demonstrated by the CV's  in  Table  5  •

     Sample numbers needed to maintain a precision  - 10  percent on  either
side of  the mean were  calculated and are presented  in  Table  6.  The number
of  bottom -sediment samples required were higher  than for water  samples.
Since  upstream  and downstream values were averaged, chromium required  the
largest  number  of  samples  for I 10 percent  precision.  The  number of samples
which  should  be taken to meet the  specified precision  was  found  to  be  at
 least  4  water and  7 bottom sediment at any  one  location.  With  a  change
 in  variables,  such as  river  stage,  groundwater  contribution,  industrial
discharge,  etc.,  the  sample  number would  be subject to change.   The varia-
 bility encountered in stream-sampling  programs, especially  sediments,
necessitates  preliminary study  to determine the degree of  sampling  required.
Monitoring the  distribution of  point-source pollutants in  stream water and
 sediments will  require a carefully designed monitoring program that considers
 the dynamics  of the  pollutant in the water-sediment system as it mixes with
 the diffuse-source load.

-------
                           UPSTREAM WATER
    0.81
   0.80-

   0.79-
   0.78-
   0.77-

"5; 0.76 -
 E
 ,- 0.75 H
CO
   0.74-
   0.73-

   0.72-

   0.71 -
   0.70 H
                 Near Bank
                  Main Flow
             i'Mean  0.777 mg/l
          1    2   3
78    9   10  11   12   13   14   15   16
                          Grid Observation Number
        Figure 4. The variability in strontium concentrations in the water at the near-
        bank and the main-flow sampling points upsteam from Lima, Ohio.
                                               ro
                                               o

-------
                       DOWMSTREAM WATER
5.0 -

4.0 -

3.0 -

2.0 -

1.0 -
0.0  J
         Near Bank //vy/vyY/
          Effluent
Effluent-Stream
Water Mixing
                                High
                              Variability
Stream
Water
                        Low
                      Variability
      1   234    5   6   7   8   9  10  11   12  13   14  15  16

                   GRID OBSERVATION NUMBER

    Figure 5. The variability in chromium concentration in the water at the near
    bank and the main-flow sampling points downstream from Lima, Ohio.

-------
     230
     220 -




     210 -


Cr ug/g


     200 -




     190 -




     180 -
                          DOWNSTREAM SEDIMENT
MEAN=195 ug/g
                                                     I
                                                     l->
                                                     ro
           1   2   3  4   5   6   7  8   9  10  11   12 13  14  15  16


                            Grid observation number

          Figure 6. The variability in chromium  concentration in the bottom sediment

          at the sampling location downstream from Lima, Ohio.

-------
             401
             35-
             30-
LIMIT OF
ACCURACY
(95%)        25-
PERCENTAGE
OF MEAN
             20-
             15-
             10-
              5-
              0-
UPSTREAM BOTTOM SEDIMENT
                    MEAN (ug/g)

                 Zn  1.3
                 Cd  0.94
                 Sr  26.8
                                      i NUMBER OF
                                      T OBSERVATIONS
                  T—i—i—|—l—l	1—l—i	1—I	1	1—I—I
               01  23456   7  8  9  10 11 12 13 14 15 16
       Figure 7. Limit of accuracy (95%) curves as percentage of the mean for elements in
              the bottom sediments  upstream from point source.

-------
                    -124-
Table  6.   Sample number required to maintain
           ±10% precision of  the mean  concentration
           of each element.
Element
Water
                                    Sediment
Cr
Cd
Pb
Sr
Zn
Mean
7
5
2
4
5
4
14
6
4
6
5
7

-------
                                 -125-
     Th e data indicate:  1)  under diffuse-source conditions,  heavy metal
variability increases laterally away from the main channel  flow,"  2)  in
close proximity, approximately 150 feet downstream from point sources,  heavy
metal variability in bottom sediment is higher than that of the stream water.

                 Heavy Metals in River Bank Vegetation

     Samples of plants, mainly river grasses and weeds,  were collected at
sites upstream and downstream from Lima.  Sixteen samples were collected  at
each location, once during the growing season and once during the winter.
Samples were digested by both the dry-ash method and by perchloric acid
digestion.  There was no significant difference between the two digestion
methods.  The arithmetic means of the concentrations are reported in Table 7.

     From the vegetation analysis the following conclusions can be made:
1) stream environment vegetation concentrates heavy metals; 2) metal
uptake by vegetation is much greater downstream of cities;  3) river grasses
release a high proportion of their metal content during the autumn.   Stream
vegetation, therefore, is one mechanism for concentrating heavy metal on
stream bottoms.

                       Heavy Metal Point Sources

     Heavy metals appearing in anomalously high concentrations were found
in the vicinity of eight cities (Table 8).  It is probable that the
high concentrations of various heavy metals in the water and sediment near
these eight cities are a result of industrial pollution.
3.75  Conclusions

     Levels of heavy metals in Maumee River Basin water, soils,and sediments
reflect normal geochemistry of the area  and indicate that point source
discharges are small compared to loadings from natural sources.

     Groundwater appears to contribute a substantial portion of the heavy-
metal load from the Basin.

     Point-source discharges are adsorbed by sediments in the stream and
taken up by rooted macrophytes.  These discharges may pose a threat to
local downstream biota.

-------
                               -126-
Table   7.  Heavy-metal concentrations in living and dead river
           vegetation along the Ottawa River  (ug/g).
Element
Cd
Co
Cr
Cu
Ni
Pb
Sr
Zn
Growing
Upstream
Lima
0.001
1.27
1.35
48.72
9.78
0.95
35.7
2.55
Vegetation
Downstream
Lima
0.89
34.2
440.0
537.0
98.5
6.3
105.7
15.9
; Dead Ve2<
Upstream
Lima
0.001
0.02
0.02
1.78
0.90
0.02
10.20
0.85
station
Downstream
Lima
0.001
1.2
64.8
62.6
5.6
1.01
45.7
9.8

-------
                               -127-
Table  8.  Heavy metals detected in high concentrations in
           stream water and sediments near cities.
      City.
                                   Sediment                    Water
Butler, Ind.                  Cd,Co,Cu,Ni,Pb                     Cd

Decatur, Ohio                 Cr,Cu                              Cd

Defiance, Ohio                Cr,Ni,Pb                           pb

Findlay, Ohio                 Ni                              Co,Sr,Pb

Fort Wayne, Ind.              Cd,Cr                              Co

Hudson, Mich.                 Cd,Co,Cr,Cu,Ni,Pb,                 Cu
                              Sr,Zn

Lima,  Ohio                    Cr,Cu,Pb                           Cr

Maumee, Ohio                  Cd,Cr,Cu,Ni,Pb,Zn                  Cd

-------
                                 -128-


 3-8  Pesticides  in watershed  soils and Maumee River Basin bottom


 3-8l  Abstract

     A pesticide scan was run on surface soils from the monitored agricul-
 tural sites in Defiance County and at Hoytville and bottom sediments from
 the Maumee River and its tributaries.  Of the compounds screened, p,p'-DDD
 was found in one soil and p,p'-DDD, p, 0-DDD and dieldrin vere detected in
 bottom sediments from Maumee River tributaries.   No other compounds vere
 detected in any samples.  Insecticide usage is low on the corn, soybeans, and
wheat crops,which account for most of the agricultural production in the
 .DEIS in •

-------
                                -129-
3.82  Methods and Materials

     Surface (0-15 cm) soil samples were taken from each  of the  monitored
watersheds in Defiance County and a composite sample from the  plots  at
Hoytville (Figure 1, page  4).  The samples were air-dried, ground and
screened and the < 2 mm fraction retained.   Bottom-sediment samples  were
taken by a scuba diver from the 0-10 cm water-sediment interface and sealed
in plastic jars.  Sediment was oven-dried prior to pesticide analysis.
Samples were taken from the Auglaize and Tiffin Rivers and the Maumee River
at Independence Dam (Figure 1, page  4).  Pesticide analysis was performed
at The Ohio State University Pesticide Analytical Laboratory by  Dr.  Acie
Waldron.

     Fifty grams of sample was blended with 300 ml of glass distilled acetone
in a Waring blender for 5 minutes at slow speed (30 volts on the Variac).  The
solution was filtered under light vacuum through Whatman #1 filter paper and
the volume of filtrate measured.

     The volume of  filtrate was reduced by evaporation to 150 ml and then com-
bined with 500 ml of  2% sodium sulfate solution in a 1-liter separatory
funnel.

     The aqueous  solution was extracted by shaking vigorously for 1-2 minutes
in  sequence with  200  ml and 100 ml  of glass distilled petroleum ether.
The combined petroleum ether  extracts were backwashed by  shaking with 500 ml
of  2%  NaHC03.   The  volume  of  petroleum  ether  extract was  recorded,,then
reduced by vacuum rotary evaporation to approximately 20  ml volume.

     The  concentrated extract was  placed on  a column of  cleanup material
in  a 250 ml  reservoired, 19 mm  diameter chromatography column containing
from bottom to  top  a  glass wool plug, 2% inches of activated  florisil,
2%  inches of silica gel, and  topped with 1 inch of anhydrous  sodium sulfate.

     The  pesticide  residues were  eluted from the  column  in sequence with
200 ml of 5$ benzene  in  petroleum ether, 250 ml of 100$  benzene,  and 200  ml
of  10% ethyl acetate  in  benzene.   Organochlorine  pesticides  elute in the
 first  two eluates and organophosphate pesticides  in the  ethyl acetate-
benzene eluant.

      Eluates were then reduced to an  appropriate  volume  with a  rotating
 flask vacuum evaporator  and aliquots  injected on  the columns  of a gas  chroma-
 tograph (Tracor 222 equiped with multiple  detectors).

      Organochlorine pesticide residues  were determined in response  to  an
 Electron Capture detector with additional detection and confirmation by  a
 Hall Electroconductivity detector.  Organophosphate pesticide residues were
 determined by response to a Flame Photometric detector specific for phosphorus
 detection.

-------
                                -130-


 3.83  Results

     The results of the pesticide scan for watershed soils and Maumee River
 Basin bottom sediments are given in Table  1.  Pesticide standards used in
 the scan are given below:

     Organochlorine
          Standard A -     Aldrin; o,p-DDE; o,p-DDD; p,p'-DDD
          Standard B -     Heptachlor; P5p-E>DE; o,p-DDT; p,p'-DDT
          Standard C -     Lindane; Heptachlor epoxide; Dieldrin; Methoxychlor

          Chlordane
          Toxaphene

     Or ganophosphate
          Thimet (Phorate)
          Diazinon
          Malathion
          Methyl Parathion
          Ethyl Parathion
          Guthion (will not respond without forming a derivative)

     Each extract solution was analyzed with all three dectectors,although the
 identity of peaks on the chromatogram correspond only to the type of eluate
 and the detector system which has been determined in past research to relate
 to the specific pesticide.

     Several peaks were observed on the chromatogram that were not identi-
 fied.  Extraneous peaks are common with the Electron Capture detector.   Some
 very prominent peaks were detected with the Electron Capture detector or the
 Hall Electroconductivity detector or with both detectors but were not identi-
 fied.  The Electron Capture detector responds to any compounds that will
 capture electrons (chlorinated hydrocarbons more pronounced and sensitive),
 and the Electroconductivity detector is specific for chlorinated compounds
but not restricted to pesticides.


 Table  1 .   Pesticide Residues Found in Soil and Sediment Samples.
Sample
No.

1.
2.
3.
h.
5.
6.
7-
Sample
Description
Watershed Surface
Hoytville
Hammersmith Roselms
Hammersmith Broughton
Speiser Paulding
Rohrs Lenawee
Heisler Blount
Crites Roselms
Pesticide
Organochlorine
Soils I/
None
None
None
None
Residues (ppb)
Organophost)hate
I/
None
None
None
None
0.89 p,p'-DDD None
None
None
None
None

-------
                                 -131-
 Table  1.   (Cont)
 Sample
 No.
   Sample
Description
       Pesticide Residues (ppb)
Organochlorine	Organophosphate
 9-
10.
     Bottom Sediments
Maumee River (independence
     Dam)
Auglaize River
Tiffin River
     None
     2.77 p,p'-DDD
     0.9h Dieldrin
None
None
None
 _!/  None means no residues detected at the sensitivity of the method which
     could "be identified in relation to the pesticide standards used.

      A very prominent peak was chromatographed in the 10$ ethyl acetate-
 benzene eluate of the ten samples,but it did not correspond to any of the
 standards used.  The retention time did not basically correspond to that of
 other organophosphate standards analyzed in previous research in the labora-
 tory including DDVP, Ronnel, Ciodrin, and Dyfonate.   Dimethoate also re-
 quired the formation of a derivative for gas chromatographic detection.   In
 addition, one or two prominent peaks were observed in the chromatograms  of
 the 5$ benzene in petroleum ether eluate and the 100$ benzene eluate.  These
 peaks did not correspond to any of the standards; in addition, under the
 conditions of the research procedures, the organophosphate pesticides
 related to the standards used should have eluted only in the ethyl acetate-
 benzene solution.  Sample No. 10 had a very prominent peak with the retention
 time for diazanon, but it was in 100$ benzene eluate and no indication of
 detection at all in the ethyl acetate-benzene eluate.  The Flame Photometric
 detector is specific for phosphorus compounds but is not limited to only the
 organophosphate pesticides.  Thus the peaks observed are likely due to a
 phosphate or phoshorylated compound, but the identity remains unresolved
 at present.

      Based on the results of this scan, no further analyses were made.
 Waldron  (197^) in a previous study on the Maumee and several other Ohio
 tributaries draining into Lake Erie found similar low values for water and
 bottom sediments.  When detected at all, pesticide residues were generally
 less than 10 ppb, while triazine herbicides were usually less than 50 ppb.
 He found that DDT, diazanon and dieldrin were the common insecticides de-
 tected, while atrazine was the herbicide found most frequently.  The generally
 low levels of insecticides found in the Maumee reflect  the land use of the
 area.  About 70$ of the Maumee Basin is in cropland and,of that, grain
 crops are dominant.  Insecticide usage by grain  farmers in Ohio is quite
 low, although it is expected that there will be  some increase in insecticide
 application as acreages of minimum and no-till increase.  Herbicide usage
 is more  common with atrazine the most common material.  It is recommended
 at rates of 1-U kg/ha  for  corn  (Ohio Agronomy Guide, 1978), while materials
 such as  lasso  (l-3 kg/ha) plus lorox or sencor  (0.5 to 2 kg/ha) are recom-

-------
                                -132-
mended for soybeans.  Herbicide usage on wheat is minimal.   Herbicide usage
by Ohio grain farmers continues to increase as more and better compounds
are introduced  and will be an integral part of minimum or no-till farming
in the future.  Most pesticides are applied at or near planting and so
discharge  to streams should be greatest in late April through May in the
Maumee.  Therefore, pesticide runoff should only be significant in the early
spring thaw events as residues from the previous year's application.  This
will not be a problem with the more degradable compounds.


3.Qh  Literature  Cited

1.  Agronomy Guide.  1978-1978.  Bulletin k^2.  Cooperative Extension
    Service, The Ohio State University.

2.  Waldron, A.C.   197^.  Pesticide Movement  from  Cropland to Lake Erie.
    EPA Tech. Series.  EPA 660/2-7^-032.

-------
                                          •ITU	
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
 . REPORT NO.
                              2.
                                                           3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
4aumee River Basin,  Ohio,  Pilot Watershed Study.
Volume 2.  Sediment,  Phosphate, and Heavy Metal
Transport.
              5. REPORT DATE March  1979 _
              of preparation
             6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
ferry J. Logan, Agronomy Department, Ohio State
Jniversity
             8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
)hio State University
tesearch Foundation
L314 Kinnear Road
tolumbus, Ohio   43212
              10. PROGRAM ELEMENT NO.

              	2BA645	
              11. CONTRACT/GRANT NO.

               Grant R005145-01
12. SPONSORING AGENCY NAME AND ADDRESS

Jreat Lakes National Program  Office
J.  S. Environmental Protection  Agency,  Region V
536 South Clark Street, Room  932,  Chgo.,  IL 60605
              13. TYPE OF REPORT AND PERIOD COVERED
             Final Report, May 1975-Mav  1977
              14. SPONSORING AGENCY CODE
is. SUPPLEMENTARY NOTES This study,  funded by Great Lakes  Program grants from the U.S. EPA,
was conducted  as part  of the Task C-Pilot Watershed  Program for the International
Joint Commission's  Reference Group on Pollution from Land Use Activities.
16. ABSTRACT Losses  of nutrients and sediment from agricultural land were monitored during
1975-1977 in the Maumee River Basin, Ohio.  These results  have been reported in Volume
L, Watershed Characteristics and Pollutant Loadings,   Special studies were made on
sediment mineralogy  and chemistry and on sediment, pesticide, and metal sediment
transport.
     Suspended  sediments were higher in total-P  (phosphorus)  and labile-P than soils or
stream-bottom sediments.   Sediments are enriched in P  during  erosion and transport
Because of preferential transport of clay which is higher  total P than the whole
soil.  Some P enrichment of suspended sediment was due to  concentration by algae in
the stream.  Photo-synthetic consumption of carbon dioxide by algae caused formation
of secondary calcite.
     Preferential  clay transport changed sediment mineralogy  from its original soil
mineralogy.  No significant mineral alteration during  sediment transport was found.
Aggregated clay was  found in runoff sediment as well as in stream-bottom sediment.
     Heavy metal concentrations were low and indicated that groundwater and eroded soil
were the major  source.   A point-source chrominum discharge was absorbed by stream
sediment and taken up by rooted macrophytes.  Pesticide scan  of soils and bottom
sediments showed only traces of DDT and other persistent chlorinated hydrocarbons.  No
other pesticides were detacted in significant quantities.
     The watershed characteristics and pollutant loadings  are discussed in Volume 1.
                               KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.IDENTIFIERS/OPEN ENDED TERMS
                           c. COS AT I Field/Group
Agricultural pilot  watershed,  agricultural
runoff, agricultural  pollutant loadings,
 ediment transport, sediment adsorption/
desorption, heavy metals,  nutrients, soil
types/sediment and  nutrient loss.
 Ohio State University,
 Pollution from Land Use
 Activities, Great Lakes
 Basin,  International
 Jo int Commi s s ion.
 8. DISTRIBUTION STATEMENT
Document is  available  to the public through
the National Technical Information Service
             VA  22161
19. SECURITY CLASS (ThisReport)
	  None
21. NO. OF PAGES
  143
20. SECURITY CLASS (This page)
          None
                           22. PRICE
EPA Form 2220-1 (Rev. 4-77)   PREVIOUS EDITION is OBSOLETE
                                                                               : 1979 — 652-885

-------