U.S. DEPARTMENT OF COMMERCE
                         National Technical Information Service
                         PB-263 970
Lake Drawdown as a Method
of Improving Water  Quality

Florida Univ, Gainesville
Prepared fa*

Corvollis Environmental Research lab, Greg


Jan 77

-------
EPA-600/3-77-005

January 1977
                                                              PB  263   970
Ecological Research Series
                      W wo.iucio BY

                     NATIONAL TECHNICAL

                    INFORMATION SKVICE  t
                         0[P*»IKfNt Of CODKttCI
                         SPRiriGHUo. V*. :;iu

-------
                RESEARCH REPORTING SERIES

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

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

This report has been assigned to ihe ECOLOGICAL RESEARCH series This series
describes research  on  the effects of  pollution on r-umans. plant  and ammai
cpecies. ard materials  Problems are assessed lor loeir  long-  and short-term
influences Investigations mcluda formation, transport, and pathway studies to
determine the fate Jf pollutants an' their effects Th-s work provides the technical
basis for setting standards to mmir.Mo undosiraOie changes in living organisms
in the aq^.tic. terrestrial, and atmospheric environments.
This document is available to the public through the National Technical Informa-
tion Service. Springfield. Virginia 22161.

-------
                                                  EPA-600/3-77-005
                                                  January 1977
    LAKE DRAWDOWN AS A METHOD OF IMPROVING  WATER QUALITY
                            By

Jackson L. Fox. Patrick L. Brezonik, and Michael A. Kelrn
                  University of Florida
               Gainesville. Florida   32611
                   Grant Number R80030S

                     Project Officer
                    Donald W.  Schults
           Marine and Freshwater Ecology Branch
       Corvallis Environmental Research Laboratory
                Corvallis,  Oregon   97330
      CORVALLIS EN\IRONMENTAL RESEARCH LABORATORY
           OFFICE OF RESEARCH AND DEVELOPMENT
          U.S. ENVIRONMENTAL PROTECTION AGENCY
               CORVALLIS, OREGON   97330

-------
                             DISCLAT'CR
     This report has been reviewed by the Corvallis Environmental
Research Laboratory, U.S. Environmental Protection Agency,  and
approved for publication.  Approval does not signify that  the
contents necessarily reflect the views and policies of the  U.S.
Environmental Protection Agency, nor docs mention of trade  names
or comrofcial products constitute endorsement or recommendation
for use.
                                   11

-------
                              FOREWORD
Effective regulatory and enforcement actions by the Environmental
Protection Agency would be virtually impossible without sound scientific
data on pollutants and their impact on environmental stability and human
health.  Responsibility for building this data bsse has been assigned to
EPA's Office of Research and Development and its IS major field instal-
lations, one of which is the Corvallis Environmental Research Laboratory
(CERL).

The primary mission of the Corvallis Laboratory is research on the
effects of environmental pollutants on terrestrial, freshwater, and
marine ecosystems; the behavior, effects and contra! of pollutants in
lake systems; and the development of predictive models on the movement
of pollutants in the biosphere.

This report describes the results «•' pilot scale investigations to
determine If the drawdown of Lake Apopka, Florida, would be an effective
restoration technique in reducing algal blooms and increasing the popu-
lation of desirable macroscopic aquatic plants.
                                              A.F.  Bartsch
                                              Director,  CERL
                                 ill

-------
                               ABSTRACT

Investigations were made to determine the feasibility of radical  draw-
down as a restoration technique for Lake Apopkn, Florida,  a 12,545
hectare lake in central Florida.  Field studies showed the lake  to  be
hypereutrophic with continual algal blooms, mats of floating water
hyacinths, and a flocculent organic muck bottom rich in interstitial
water nutrients.  Sediments were dredged from the lake bottom and
placed in aquaria, columns, tanks and pools.  Following dcwaterlng
and varying drying periods, the containers of sediment were refilled.
A large number of physical, chemical and biological parameters were
monitored before, during and following sediment drying.  Results indi-
cate that drawdown improves subsequent refill water quality.  In muck
sediments, drying causes significant water loss and shrinkage.  Loss
of organic material is minimal.  During and following refill, sediment
is colonized by two macroscopic aquatic plants, Typha (cattail)  and
the alga, Chara.  Drying results in only minor chemical changes  In
muck sediment.  Refill water In the pool test simulations has the
same or lower nutrient content, lower turbidity, higher dissolved
oxygen, lower temperature, fewer algae and a more diverse benthic
invertebrate population.  Based on these laboratory scale investiga-
tions, drawdown appears to be an effective restoration technique for
Lake Apopka.

This report was submitted in fulfillment of Grant Number R800305
by the University of Florida under the sponsorship of the Environmental
Protection Agency.  Work was completed as of June, 1974.
                                   Iv

-------
                           CONTENTS
Abstract                                                   iv

List of Figures                                            vl

List of Tables                                             vlll

Acknowledgments                                            *

Sections

1       Conclusions                                        1

II      Recommendations                                    2

III     Introduction                                       *

IV      Study Approach                                    13

V       Trophic State and Sediment Characteristics        17
        of Lake Apopka

VI      Effec-s of Drawdown OR Sediment                   28

VII     Effects of Drawdown on Overlying Water Quality    57

Vtll    References                                        88

IX      Appendix                                          91

-------
                                  FIGURES

  No.                                                                 _
  	                                                                Page

   1    Lake Apopka and  Major Land  Use Types                            5

   2    Oklawaha  Chain of  Lakes                                         6

   3    Plexiglaa Column Design                                        14

   A    Steel Tank Design                                              jj


   5    Chlorophyll ^ and  Carbon-H Productivity at Station 3,         21
       Lake Apopka


   6    Diurnal Changes  of Dissolved Oxygen ;md Terpcrature,           22
       Lake Apopka, August,  1972


   7    Location of Sediment  Sampling Stations                         25


  8    Effect of Sun Drying on Moisture Content of Muck Sedl-         29
      mcnt (Aqutrluo Experiment)


  9   Consolidation of Muck Sediment Upon Sun DrylnR (Aquarium       30
      Experiment)

 10   Consolidation of  a  Drained and Undrlcd Column  o' Muck           31
      Sediment


 11   Consolidation of  Muck Sediment (Column-Tank Study)              y,

 12   Percent  Water In  Muck Sediment (Column-Tank Study)              35

 13   Volatile Solids  In  Muck Sediment (Column-Tank  Study)           36

 14   Percent  Water  In  Muck Sediment  (Pool Study)                    33

 15   Vein' ile Solids In  Muck Sediment (Pool" Study)                  39

 16   Total Phosphorus  and  Nitrogen Changes  In Column and            49
      Tank Sediments


 17    Changes  In Ammonia Nitrogen  In Column and Tnnk Sediments       50

 18    Changes  In Orthophcsphate  In Column and Tank Simulation        58
     Water


 19   Changes  in Nitrate in Colu-im and Tank Simulation Water         59
              «
20   Changes In Total Nltropcn  in Column nnd Tank Simulation        60
     Water
                                 vi

-------
                                 FIGURES

 No-                                                                Page

 21   Changes  In Ammonia In  Column and  Tank  Simulation Water          61

 22   Changes  in Dissolved Oxygen In Column  and Tank Simula-          62
     tlon Water

 23   Changes  in pH in Column  and Tank  Simulation Water               63

 24   Weekly Changes  in  Turbidity In Pool Simulations                 65

 25   Weekly Changes  in  Temperature in  Pool  Simulations               66

 26   Weekly Changes  in  Alkalinity In Pool Simulations                67

 27   Weekly Changes  in  pH in  Pool Simulations                        68

 28   Weekly Changes  in  Orthophosphatc  in Pool Simulations            69

 29   Total Phosphorus in Pool Simulation Refill Water                70

 30   Ammonia  in Pool Simulation  Refill Water                         72

 31   Weekly Changes  In  Nitrate in Pool Simulations                   73

 32   Weekly Changes  In  Dissolved  Oxygen in Pool Simulations          75

 33   Effect of  Orylnc Period on  Sediment and Refill Water            78
     Quality  in Laboratory  Experiment

 34   Primary Productivity and Chlorophyll a in Control and           80
     Test Tanks

35   Selenpstnin Growth in Wet Sediment Suspensions                  82

36   felcnaatrum Growth In Dry Sediment Suspensions                  83

37   Diversity  Indices  for Test and Control Pool Simulation          85
     Phytoplankton Populations

38   Chlorophyll Concentrations In Test and Control Pool            86
     Simulations Following Refill
                                  vll

-------
                                TABLES

N'o.                                                                Pace

 1   Chemical and Biological Conditions in Oklavaha Lakes,          8
     October, 1968

 2   Major Lake Apopka Restoration Features                         12

 3   Summary of Water Chemistry Data fron Lake Apopka Moni-         18
     torlni;, 1972-1974

 l»   Summary of Water Quality Data fron Lake Apopka for             19
     1969-1970

 5   Community Metabolism in Lake Apopka                            23

 6   Comparison of Muck fron Different Locations in Lake            24
     Apopka

 7   Nutrients in Interstitial Water of Like Apopka                 26

 8   Summary of Physical Characteristics of Muck in Column          32
     Experiment

 9   Aquatic Vascular Plants of Lake Apopka                         41

10   Seeds Found In 17 Samples of Sediment from Lake Apopkn         42

11   Plants Colonizing Dried Sediment                               43

12   Plant Blomnss Harvested from Pool Simulations                  44

13   Bcnthlc Invertebrates Populating Pool Simulations Fol-         47
     lowing Drawdown and Rof.II

14   Nltrop.cn and Phosphorus Distribution In Sediment Cores         52
     from Tank Experiment

IS   Metal Analyses on Sediments fron Lake Apopka Pool Slmu-        51
     lotions

16   Interstitial Water Nutrient Content of Apopka Pool Slrau-       54
     lotions

17   Summary of Lake Apopka Synoptic Survey                         56

18   Nutrient Leachablllty of Apopka Pool Simulation Scdi-          56
     rants
                                  vltl

-------
                                TABLES

No.
19   Analysis of Variance Results for Water Quality Para-            76
     meters In Pool Simulations

20   Post Drawdown Pool Simulation P'uytoplankton Results             87

-------
                            ACKNOWLEDGMENTS
The efforts of all of our personnel are gratefully acknowledged.
Michael Mover and Richard Bloom made most of the field collections.
Chemical analyses wore performed by Karen Tuttle, Boyd Welsch,  and
Lloyd Chesney.  The alfial assay work was done by William Moyer.
Donna Hunt performed the phytoplnnkton counts and the Lake Apopka
seed identifications were made by Mrs. L. A. Hetrick.  Joseph Hand
performed the statistical analyses.

We also thank Donald Schults of the Corvallis EPA Laboratory, who
served as Project Officer, for his advice and patience.

-------
                                SECTION I

                               CONCLUSIONS

 In laboratory simulations, we have found that drying muck sediments
 from Lake Apop'-a, Florida, results in significant water loss and
 shrinkage.  Loss of organic matevial is minimal.   Peat sediments,
 while they lose water as a result of drying, do not consolidate
 appreciably.

 The only two aquatic macrophytes which sprout on  drying sediments are
 TyPha. (cattail) and Eichhornia (water hyacinth).   Following flooding
 of the dried sediments, Typha is the only vascular plant which survives.
 In several simulations, the macroscopic green alga Chara (stonewort)
 colonized the dried sediments after flooding.  Chara is easily trans-
 planted by sprigging.  Dried sediments remain consolidated  for long
 periods of time (at least a year)  following refill.

 The benthic invertebrates on dried sediments following four months
 of refill show greater diversity than populations  on control sediraents.

 Few differences were found for nitrogen and phosphorus forns between
 dried  and undried sediment.   Drying results In  only  minor chemical
 changes  in auck sediment.   Lake  Apopka  sediment has  a  low capacity  to
 absorb added  phosphate,  and  mixing of unconsolidated sediment with
 overlying watei  acts as a  source rather than a  sink  for  aqueous
 phosphorus.

 Nutrient  levels  in water above dried  sediments  (pool simulations) are
 the  same  or lower than levels  In control simulations.  Because of the
 higher turbidity levels  in the controls, levels of particulate nitro-
 gen and phosphorus are lower in  the treated  pools.

 Refill water in  the  pool test simulations has lower  turbidities.
 higher dissolved  oxygen  values, and lower temperatures (because of
 lower turbidities) than  water In the  control pools.  Phytoplankton
 in refill water above  dried sediments is less numerous and contains
 fewer cyanophytes than water above undried sediments.  Chlorophyll
 levels are also lower  in the test simulations.  A laboratory algal
assay showed that ground up dried sediment supports lower algal
populations than does wet sediment.  Adenosinc triphosphate levels
and total plate counts at the sediment water Interface of dried
versus undried sediments were not significantly different, lending
support to the conclusion that the changes occurring during sediment
drying are predominantly physical.

-------
                              SECTION II

                            RECOMMENDATIONS

Drawdovn/sedlnent consolidation appears to be a  feasible  lake restora-
tion technique.  Its use is recommended for shallow lakes in which
extensive muck deposits cause problems because of th««ir ready suspen-
sion by wind-driven waves and outboard motors and their consequent
nutrient release.  Drawdown is most attractive for lakes  in which the
littoral zone slopes gently so that a small vertical  drawdown exposes
large areas of bottom.  Any drawdown prelect needs to consider the
effects of increased flow and possible sediment  transport on downstream
water bodies.  If pumping is required to achieve substantial drawdown,
care must be taken to mlniaize the amount of loose sediment that is
pumped from the lake.

A serious effort should also be made to evaluate the potential for
physical removal of muck and peat from lake bottoms for purposes of
soil amendment.  It 1- not naive to consider muck and peat a misplaced
resource; the highly fertile and profitable muck farms north of Lake
Apopka attest to this fact.  An integrated program of 
-------
should be given to removal of dried muck from shore areas, perhaps
transporting it by truck to the nearby nuck farms.  The possibility of
pumping the wet muck sediments onto the nuck farms should also be
explored.  Fish spawning areas need to be added to the lake,  either by
removing the muck down to the sand bottom or by covering the  diied muck
with sand.  The former method would seen preferable since it  will
increase the lake depth.  Consideration should be given to planting
or seeding the littoral areas with desirable forms of aquatic vegetation
The cessation of preventable nutrient loading is, of course,  Impera^-o.

-------
                              SECTION III

                              INTRODUCTION


Florida's mild climate, extensive coastline, and more than 8,000 lakes
effectively attract tourists, new residents, and industry.  The rapid
growth of Florida's population has exerted heavy demands on its water
resources, especially for fishing, boating, and water contact sports.
In an effort to meet the desire for water front property, Florida s
coastlines are being developed using dredge, fill, and channelization
techniques.  The same procedures are being used on natuial and manmade
lake shorelines.  Flood control structures often are installed and pre-
vent natural fluctuations in water level.  Development on floodplains
sometimes necessitates river channelization for rapid removal of excess
water.  Vihen these physical manipulations of -..atural systems are coupled
with wastes generated by the resulting population increase, serious
water quality degradation occurs.

Water quality degradation in lakes is characterized by a shortening of
food chains and by respiratory dominance.  Unique vascular plant prob-
lems occur in Florida as a result of the proliferation of introduced
exotics,  such as Florida elodea  (Hydrilla verticlllata) and water hya-
cinth (Etchhornta crassipes).  In extreme cases, eutrophic lakes are
characterized by odoriferous algal scums, fish  kills, oxygen depletion
and resultant hydrogen sulfide production, buildup of benthic organic
deposits, and floating islands of hyacinths or  impenetrable growths of
submerged vegetation.  Fish  camps and resorts on such deteriorated lakes
are unable to attract  tourists and the resulting loss of  income and de-
creased  property values  can  be significant.  Public  discontent  is  pro-
portional to  economic  loss.


LAKE APOPKA BACKGROUND AND HISTORY

Lake Apopka,  a  12,545  ha lake  (Figure 1)  in central  Florida,  exhibits
most of  the systems  of degradation mentioned above.   According to Brezonik
et al.  (1969),  Apopka  is the most eutrophic of  the  lakes in  the Oklawaha
chain  (Figure 2).  Gourd Neck  Springs,  in the  southwestern portion of
Lake Apopka  is  considered the  headwaters  of the chain.   Water flows  north
through  Lake  Apopka  into Lake  Beauclair via the Apopka-Beauclair Canal.
Lake Beauclair  empties directly  into Laire Dora, which drains into Lake
Eustis  through  the Dora  Canal.   Lake Harris water also enters Lake Eustis
via Dead River.  Lake  Eustis is  connected to Lake Griffin by Haines  Creek,
and Lake Griffin discharges  directly into the  Oklawaha River.  A manmade
canal  connects  Lakes Griffin and Yale (State of Florida, 1970).  Table 1

-------
   APOPKA-
  BEAUCLAIR
  CANAL
                                                               MAGNOLIA
                                                               COUNTV
                                                               PARK
       GOURD
       NECK
         SPRINGS
                                                       WINTER
                                                       GARDEN
                                               R8KS5H
                                               n    K-ijEaa
                                                MILES
Figure 1.  Lake Apopka and major land use types (modified from Perez
          1972)

-------
                                                           WINTER
                                                           GARDEN
Figure 2.  Oklavaha chain of lakes

-------
 provides a comparison of water quality parameters for the 5 lakes.  The
 Okiawaha River has been and remains  the subject of state and national
 concern, for It forms pare of the Cross-Florida Barge Canal.  A section
 of the Okiawaha already has been damned as a part of canal construction.
 Ttie resulting 5,260 ha impoundment is known as Rodman Reservoir.  Further
 north and east, just above Lake George, the Okiawaha joins the St. Johns
 River, which discharges into the Atlantic Ocean east of Jacksonville.

 A number of nutrient sources have been cited as causative factors in the
 eutrophlcation of Lake Apopka.  These include agricultural runoff from
 vegetable farms (muck farms) and citrus groves, domestic waste effluent
 from the City of Winter Garden, and wastes from citrus processing plants.
 The relative importance of each of these nutrient sources has been the
 subject of considerable controversy.  Perez (1972),  Shet.'ield (1970),
 Heaney et. al. (1971) and many others have all studied aspects of the prob-
 lem.   The lack of data has hindered the investigations,  and  an accurate
 hydrologlc model and/or nutrient budget for Lake Apor
-------
   TABLE 1.  CHEMICAL AND BIOLOGICAL CONTUTIOSS IN OKLAWAHA LAKES.
                            OCTOBER, 1968a
                      (FROM BREZONIK ET AL. 1969)
Constituent1*
Apopka
Dora
Harris
Eustls
                                            Crlffln
Dlss. 02
Conductivity
pit
Alkalinity
COD
Color
Sccchi Disc
Turbidity
Suspended Solids
TON
NH3-N
Ortho P03, nitrogen species in  ag N/l,  phosphorus species In ug P/l, color
 in ng/1 as Pt., specific conductance  in umho cm'1, Secchi disc visibil-
 ity  in meters, chlorophyll  a_ in rog/m3,

cPrimary production in tug C  flxed/1-hr.  Composite  lake water samples
 were run  in a  laboratory incubator.

-------
The persistence of phytoplantcton blooms and adverse public opinion prompted
Ihe 1962-1964 Florida State Board of Health study (Huffstutler e_t al. 1965).
In their report they concluded, among other things, that "the lake is In an
undeiirable biological condition and does not provide a satisfactory environ-
ment for fish food.  This is caused by excessive nutrients contributed by
many years, from the following sources:  (a) domestic waste, presently
treated, but untreated in years past, (b) untreated citrus packing, canning,
and concentrating waste, (c) leaching from citrus growing, fertilizing and
Irrigating, (d) irrigation, drainage, and spraying in the truck farm areas,
(ex allowing dead fish to remain and decay in the lake, and (f) allowing
destroyed hyacinths to remain and decay in the lake."  These findings, as
well as newspaper articles such as the one entitled "Lake Apopka's Lingering
Death" (Lane 1966, cited by Forbes 1968) further increa°*d public discontent.

In 1967, the Governor of Florida requested the aid of the Federal Water Pol-
lution Control Administration in evaluating the restoration of Lake Apopka.
In March of 1968, the F.W.P.C.A. surveyed the lake to characterize the bot-
tom materials and nutrient contributions by spring water, citrus grove run-
off, and rainwater.  They found 90 percent of the bottom to be covered by
unconsolidated muck averaging 1.5 m (5 ft) in thickness.  Only 5 percent of
the bottom was characterized as being biologically productive.  The distri-
bution of nutrients in the bottom was relatively unlfoTo over the lake.
The muck contained from 1 to 4 percent (dry weight) total nitrogen, 0.02 to
0.21 percent total phosphorus, and 90 percent organic naterial.  The F.W.P.C.A.
considered citrus grove runoff and rainfall as significant, sources of ni'ro-
gen and phosphorus.  Gourd Neck spring water contributed low concentration
of these nutrients.

Additional data on the Apopka sediments were gathered by the Orange County
Pollution Control Department (Sheffield 196?).  Thia report concluded that
the top portion of the sediment contained higher amounts of phosphorus and
organic matter than underlying sediment.  Calcium, magne-ium, and sodium
increased with sediment depth.  The report also stated that "results of
this study would tend to indicate that if the bottom material were allowed
to dry and oxidize by the sun, a reduction in organic material would be
realised."

The Corps of Engineers (Department of the Army 1967) also studied Lake
Apopka sediments.  In 1961, dry rod probings were made to determine sedi-
ment thickness along the northern shoreline.  In 1965 and 1967, tventy-nine
core borings were made in the same area,and lengths and physical descrip-
tions of each core were presented.  The surface of the study area was foand
to be covered by either peat or soft organic silt and clay deposits varying
in depth trom less than 0.3 m (1 ft) to over 15 m (50 ft).  Below these
organic sediments are interbedded deposits of clay, sand, and silt.

While there is a long history of disagreement concerning capability for
nutrient inputs to Lake Apopka, it la generally agreed that the lake is
hypereutrophlc and that something should be done about the situation.  However,
what should be done Is also a subject of controversy, with suggestions ranging
from commercial catfish farming tc radical drawdown and sediment removal.

-------
Some have suggested the entire lake bottom be converted to additional
muck farm acreage.
LAKE RESTORATION BY DRAWDOWN

The Florida Department of Pollution Control advocates drawdown and in
1972 submitted a proposal to the Environmental Protection Agency for
funds to dewater the lake sediments (Florida Department of Pollution
Control 1972).  Drawdown as a method of lake restoration was suggested
by reports of the Florida Game and Fresh Water Fish Commission on Lakes
Trafford and Hancock.  Both lakes showed Improvement in water quality
as a result of natural drawdown and subsequent exposure of bottom sedi-
ments.  Lake Trafford, prior to its 1962-1963 drawdown, had a game to
rough fish ratio of approximately 1 to 6, extensive algal blooms, and
an unconsolldated muck bottom with few benthic organisms and little
rooted aquatic vegetation.  Drawdown caused by a drought exposed 15 per-
cent of the bottom sediment for a period of 6 months.  The original 30
cm (1 ft) of muck consolidated to 5-10 cm (2-4 in).  After the lake re-
filled naturally, the consolidated sediments supported bcnthlc organisms
and rooted aquatic plants.  With a restored habitat and food supply,
the game fish returned.  Less than a year after refilling, (.lie game fish
population had increased from its pre-drawdown 20 percent to 35 percent
of the total.  In 1968, 5 years after drawdown, game fish comprised 76
percent of the total population.

Information on Lake Hancock is less detailed, but the story is similar.
In 1965, the lake was reported to be highly eutrophic, characterized
by almost continual algal blooms, unconsolidated muck, and a paucity
of both benthic organisms and rooted aquatic vegetation.  In 1968, nat-
ural drawdown occurred and sediments were exposed for an unreported
period.  The effects of exposure Included increases in water transpar-
ency, bottom firmness, benthic organisms in the littoral zone, and rooted
aquatic vegetation.

More recent experiences with drawdown of Florida lakes suggest the method
la effective in Improving certain aspects of water quality.  In 1971,
Lake Tohopekaliga, a 9,200 ha lake in Osceola County, was drawn down in
an effort to improve the fish habitat.  According to a report by Florida
Cane and Fresh Water Fish Commission (1972-1973 Annual Report), the
standing crop of fish doubled as a result of drawdown.  The invertebrate
population also Increased.  Hydrilla verticillata, however, became es-
tablished following refloodlng.  Hydrilla is a particularly troublesome
submersed vascular plant in Florida.  Although the consolidated sediments
remained firm following refill, the report concluded that at the present
rate of organic deposition (of decomposing hyacinths and algae) the con-
dition of the substrate in the littoral zone would be comparable to that
which existed prior to drawdown in a relatively short period.  In addition,
algal diversity decreased and bloom frequency increased after refill, and
chemical water quality did not Improve.  Total organic nitrogen and total
phosphate were higher following drawdown.  The Tohopekaliga experiment.
                                     10

-------
 while successful in terns of Increasing fish blomass, did not cau«e a
 long-tern change In trophic status, undoubtedly In major part because
 significant amounts of sewage effluent (a total of 2.0 mgd) from 5 plants
 have continued to enter the lake.  The authors concluded that "any long-
 term Improvement iji the lake requires complete removal of_ the large and
 ever-increasing volume of_ sewage being discharged t± It."

 Under the direction of the Orange County Department of Pollution Control,
 Lake Eola, In Orlando, was drawn down In 1972.  According to C. W. Sheffield,
 (1974), who directed the project, "at least partial success was attained."
 Although fishing has Improved, there are still occasional algal blooms.
 The blooms are probably caused by the continued Input of storm water drain-
 age.  The Eola experience again points to the need for curtailment of nu-
 trient sources as an essential component of any lake restoration scheme.


 LAKE AP07KA RESTORATION PLAN

 From a practical standpoint, drawdown appears to be the oopt feasible
 technique for restoring a lake the size of Apopka.  Other restoration
 techniques, such as hypollmnetlc aeration, dredging, bottom sealing, and
 chemical treatment  are Impractical for or not applicable to Lake Apopka
 for a number of obvious reasons.  Because of past experience with draw-
 down and Its apparent  economic feasibility,  the Florida Department of
 Pollution Control has  chosen that method  of  restoration for Lake Apopka.
 The 1972 D.P.C. pronosal to the Environmental Protection Agency basically
 calls for pumping Apopka water through the Beauclalr Canal,  through the
 Oklawaha lake chain and into the Oklawaha River.   Extensive  pumping would
 create a relatively narrow river of  water from Gourd Neck Springs,  In
 the southern end of the lake,  to the pumps at the northern end.   Following
 the drying period,  refill would occur naturally,  principally from Gourd
 Neck Springs Inflow, subsurface seepage,  and surface runoff.  Table 2
 shows a summary of  the Florida  Department  of Pollution  Control's restor-
 ation plan.

 Anticipated  results  include  lake deepening and  an  Improvement In water
 quality due  to  retardation of nutrient  flux  to  overlying waters.  This
 reduction In nutrient  transport  will  theoretically result  from both oxi-
 dation  of the sediment  surface  (Mortimer  1941)  and the  physical  consoli-
 dation  of the sediment.  Consolidated sediments are  less likely  to  be
 physically suspended by agitation than are unconsolldated sediments.  This
 latter  factor is particularly important in Lake Apopka, for  turbulence
 produced  by m*ld storms and even  small outboard motor boats readily brings
 bottom  sediments to the water surface.  The  hardened substratum should
 provide a growth platform for rooted aquatics,  thereby  increasing the
 available niches to support a diversified food web.

 In view of the large size of the proposed project  and the lack of in-
 formation concerning the effect of drying on Apopka sediment, the Environ-
mental Protection Agency understandnbly required further data before
making a  decision on the State's proposal.  The present project was funded
 In order  to obtain some basic data concerning the effects of dewatcrlng
Apopka sediments on  subsequent refill water quality.

                                    11

-------
            TABLE 2.  MAJOR LAKE APOPKA RESTORATION FEATURES
          (FROM FLORIDA DEPARTMENT OF POLLUHON CONTROL,  1972)
Abatement of man-made wastes into lake:

     From farms by July - September 1973
     From Winter Garden municipal waste treatment by mid-1971

Consolidation of lake bottom sediments:

     Total bottom sediments exposed:  29,500 acres
     Bottom sediments dried:  16,500 acres
     2 bottom sediments dried/total lake sediments:  502

Water quantities:

     Water drawdown from approx. 66.5 ft msl to 58 ft msl
     Water In lake at 66.5 ft msl approx. 191,000 to 197,000 nc.-ft.
     Water In lake at 58 ft msl approx. 1500 to 7500 uc.-ft.
     Water Into lake from springs:  approx. 60 cfs (38.3 mgd)

Lake dredging:

     To provide settling basin for silt removal prior to water being
        pumped downstream
     To connect some deep water pockets (10 ft deep or more)

Treatment of drawdown water:

     Silt settling using special barriers
     Time phasing and flow regulation to reduce high nutrient loads
     Contingency operations Involving diversion and pump stoppage
     In-canal aeration

     Total project length:  700 to 850 days (3 concurrent fiscal yrs)
                                   12

-------
                               SECTION IV

                             STUDY APPROACH


 In predictive studies of large lake phenomena, several approaches are
 possible.  These include Kathematical modeling, in situ lake studies,
 and laboratory simulations.  Economic and scientific constraints usually
 dictate which technique Is feasible.  In this study, laboratory simula-
 tions were used for several reasons.  Modeling was Inappropriate because
 of the lack of quantitative information available concerning sediment-
 water exchange phenomena and Lake Apopka ecosystem dynamics.  A more
 lengthy consideration was given to in situ drawdown studies.  Early in
 the study, a trip by boat was made around the entire Apopka shoreline
 in a search for a small embayment which could be separated from th-.-
 lake proper and drawn down.  No such embayment could !*e fniim1..  '.'c .ilso
 made economic estimates of isolating a small section of the lake by using
 large diameter pipes or earthen dikes.  These schemes were also found to
 be prohibitively expensive or technically unfeasible.

 In all,  five types  of contalncts were usrd for the lake simulations.
 Preliminary experiments were conducted in small glass aquaria and
 plexiglas cylinders.  For the full scale experiments, we used columns,
 tanks and pools.  The cr .amns (Figure 3) were constructed ot 0.6 cm
 thick plexiglas and were 0.3 m hy 1.2 m high.  Drainage portals were
 made by  drilling 2.5 cm diameter holes 2.5 cm from the bottom of each
 column.   The holes  were plugged  from the inside with a rubber stopper.
 through  which i glass tube  extended  to the outside.   Attached to the
 glass  tubes were plastic  tubes controlled by pinch clamps.

 The  tanks (Figure 4) consisted of 1.5 m  lengths of 1.2 m corrugated
 steel  pipe embedded In concrete.   Drainage ports consisting of 1.5.2 cm
 capped sections  of  5.1 cm pipe were  welded Into holes about 7.6 cm from
 the  bottom of each  tank.  Plankton netting (1 mm pore size)  covered
 each portal to prevent loss of the bedding material.   The concrete  was
 washed with sodium  trlpolyphosphate  and  the  metal  surfaces  of  the  tanks
 were cleaned  using  steel wool  and steel  scrub brushes.   All  surfaces
 (Inside and out) were  then covered with  two  coats  of  white  epoxy con-
 crete  enamel.  Brown pea gravel  (12.7 cm)  and builder's  sand  (2.5 cm)
 were placed In the  bottom of each tank to  facilitate  draining.

 A  final series of experiments was conducted  using  plastic pools with
 a diameter  of  2.4 m and a depth of about 0.5  m.  Submersible electric
 water  pumps were used  to etir  the water overlying  the  sediments.

 *he Lake Apopka sediment and water used  in the simulations were re-
 trieved as  follows;   A 3 hp (2237 J/sec) was  uoed  to  pump muck  from
 the Lake Apopka bottom into 55 gal (208 1) drums.  The pump was
operated from an 18  ft (5.5 m) boat powered by a 40 hp  (29828 J/aec)
outboard motor.  The pump was equipped with 7.5 m rubber intake
                                   13

-------
         1.2
                 ~K1\1\
                  77
              1\
                 1.2
                         //
              r
             0.46
             0.6
0.1 S
                 H2O
                MUCK
                SAND
Figure 3. Plexiglas column design (dimensions in meters)

-------
          0.1
               G.2
                           •1.2-
0.2
           -1.2-
                                                                        1.5
0.2
           I



Figure 4.  Steel tank design  (dimensions In meters)
                                      15

-------
and discharge hoses.  The intake hose had a weighted screen with a
0.64 cm mesh to prevent the access of larger pump damaging particles
such as snail shells.  Durinr. JrcdRlnR, the boat was used to pull
the hose throuRh the seJlnoru, providing continual intake of sediment
from a depth of 15-30 cm below the sediment-water interface.  When
the intake hose was not dragged through the sediment, the end dug
through the Buck sediment to underlying sand and clay material, and
brought up this material along with large amounts of water.

A wooden raft supported by eight 55 gal (208 1) drums was used to sup-
port five drums into which the dredged -nuck was discharged.  The water
depth ve were working in was about 1.5 ra.  A total of 2.000 gal '7.570 1)
of muck were retrieved in this manner and transported by truck to Gaines-
ville, where the drums were emptied by pouring into the simulations.
With the exception of special laboratory studies, the drawdown investi-
gation was conducted out of doors.

The peat was recovered from about 20 ra offshore  in about 1 m of water
in the same area of  the lane.  The material was dug up and shovelled
into plastic garbage bags.

Overlying water was  retrlevr.J Just offshore from  the Winter Garden,
Florida Municipal boat ramp.  A  gasoline powered  centrifugal pump was
used to pump the water directly  from the lake  into eighteen 55 gal  (208
1) drums on the back of the rental truck.  These  were emptied by pour-
ing over an  inclined bo.ird over  the sediment surface  in the tanks and
columns.  Refill water was obtained by carrying  the drums  and pumps  to
Gourd Neck Springs  In the research vessel and on  a rented  motorized
pontoon boat,  filling the drumj,  then  transporting these back to the
dock and punping the water  into  drums  inside the  truck.

Before obtaining the large quantities  of sediment for th<«  simulations,
studies were made of sediments  from  five locations  in Lake Apopka.
The results  (sec Section V) showed  that, with  the exception of  sediment
taken off  the  Winter Garden boat ramp, nutrient  concentrations, volatile
Hollds, and water content vnlucs for  representative  acdincnts were
fairly consistent.   Near Winter  Garden,  high phosphorus  levels  may  be
caused by  sewage disposal.  Decause of the  ea^e  of  access, we  recovered
sediment  for  the  simulations  fiom 200  m  offsnorc from the  Magnolia
County Park  on the  northeast  shoreline of  Lake Apopka.   Routine water
quality data was also obtained  from this area  of the lake.

The basic  study plan for the  simulations included:   (1)  a  period  of
equilibration and  baseline  data  gathering.  (2) sediment  exposure,
 (3)  refill.  aM (6) post-drawdown monitoring.   The parameters  measured
and  results for each are described In the  following sections.   A  sum-
mary  of  methods is  Included in the Appendix.   Special laboratory  studies
are  included where appropriate.
                                   16

-------
                                SECTION V

        TROPHIC STATE AND SEDIMENT CHARACTERISTICS OF  LAKE APOPKA


 This section summarizes the results of routine water  quality monitoring
 conducted on Lake Apopka through the duration of the  project.  Major
 parameters considered were primary productivity, algal bionass (chloro-
 phyll a)  and levels of nitrogen and phosphorus species.  These data are
 useful for comparison with the 1. ke simulation results and, coupled with
 previous  studies, provide a data base  for  the pre-drawdown trophic rjn-
 dltion of Lake Apopka.   This section also Includes the results of a
 synpntlc  survey of Lak« Apopka sediments,£0 evaluate  the potential for
 nutrient  leachability.

 Previous  studies have clearly shown that Lake Apopka  is greatly enriched.
 Brezonik  and Shannon. (1971)  found that  the  lake had the highest trophic
 state index (TSI) of 55 lakes studied in northern and central Florida
 and  suggested that agricultural  runoff  and  domestic waste coupled with
 the  lake's shallow depth (Z= 1.3 m) wera the primary  factors responsible
 for  its presently degraded conditions.

 For  the purposes of biological and  chemical water quality monitoring, we
 selected  three  stations in the area from which sediments were collected
 for  the simulations.  The stations  were  located 40, 165, and 330 m off-
 shore (stations  1-3,  respectively)  on a  line extending east from the
 Magnolia  County  Park.   The results  of the chemical analyses are shown
 in Table  3 and  represent grand means for  all stations and all sampling
 times.  No major seasonal trends  could be discerned and differences in
 results between  the individual locations were minor and inconsistent.
 Temporal  variations in  water  quality  are sporadic rather than seasonal
 and  reflect  algal  bloom-crash phenomena which occur throughout the year.

 The  results  in Table  3  compaio well with two years of data 'Table 4)
 obtained  from composite  samples of  the lake (Brezonik and Shannon, 1971)
 and  Indicate that  the lake contains extremely high levels of both organic
 and dissolved inorganic nutrients.  High COD,  Kjeldahl nitrogen and
 total phosphorus  values  reflect the abundance of algae.  Secchi disk
 readings  In Lake  Apopka ranged from 15-30 cm and averaged .-bout 22 cm.
These arc among  the lowest transparencies we have measure  in Florida
 lakes and  Indicate  continuously high  turbidity from profuse algal growth.
 By comparison, Secchi disk transparencies of 2-3 m are usually cited  as
 the critical values separating oligotrophic  and  eutrophic lakes.   High
daytime pH and dissolved oxygen values also  reflect  intense algal acti-
vity.  Among the  inorganic nitrogen forms, ammonia  is  always higher than
nitrate and in spite of the continually high algal  concentrations,total
inorganic N Is never depicteJ.  In fact, the minimum concentration found
 (0.4 mg N/l) during the two years of monitoring  exceeds the often cited
critical spring maximum figure for temperate lakes  (0.3 mg  N/l).   Simi-
larly, orthophosphate is never depleted, and the  minimum value  found
 (0.09 mg P/l) is far in excess of the critical concentrations cited by
limnolegists for algal bloom stimulation. Clearly  inorganic nutrient
                                  17

-------
TABLE 3.  SUMMARY OF WATER CHEMISTRY DATA FROM LAKE APOPKA MONITORING,
          1972-19/4
Parameter
Dissolved 02 (ng/1)
pH
Alkalinity (mg/1 as CaC03)
Total Organic N (mg/1)
Participate Organic N (mg/l)
Ammonia (mg N/l)
Nitrate 
-------
TABLE 4.  SUMMARY OF WATER QUALITY DATA FROM LAKE APOPKA FOR 1969-1970"

Parameter
Seech 1 Disc (m)
Primary Production
(rag C/m3-lir>
Chlorophyll a_ (mg/m3)
Dissolved Oxygen (mg/1)
pH
Alkalinity (as mg/1 CaC03)
Conductivity (umho/cm)
Total Organic Nitrogen
(ng/1)
Ammonia (mg/1 of N)
Nitrate (mg/1 of N)
Ort.iophosphate (mg/1 of P)
Total Phosphate (mg/1 of P)
Na+ (mg/1)
K+ (mg/1)
Mg+ (mg/1)
Ca* (mg/1)
Mn+ (mg/1)
Ci~ (ng/1)
SO}2 (ng/1)
F- (mg/1)
COD (mg/1)
Average
0.22
337
60.4
10.14
8.85
140
315
4.45
0.27
0.15
0.065
0.380
14.33
3.73
14.3
S5.3
7.0
23.5
16.3
0.41
159
Range
0.15 - 0.30
231 - 521
34 -73
9.40 - 11.1?
8.42 - 9.57
126 - 148
304 - 324
3.45 - 5.74
0.08 •• 0.58
0 - 0.26
0.004 - 0.110
0.239 - 0.500
10.00 - 18.00
1.20 - 5.00
12.9 - 16.0
25.0 - 100
S.O - 9.0
21.3 - 27.5
10.2 - 20.0

113 - 191
aData summarized from Breconik and Shannon (1971)
                                  19

-------
concentrations  do not limit alR.il  production tn Lake Apopka  at  tlic pre-
sent time.   More. likely, pliysic.il  factors, i-sppcJ.il ly  llf.ht.  .-in- more
Inport.-iiil  In control I Inn .'» \ltl.'  rr.idm-lltin.  I" siim-i.it/,  tin- « iit-mli-.il
d.ita indicate tli.it l..ikc Apopka  It  hvpi-ioiit rophlc .ind has highly degraded
w.i tor qual Ity.

Primary production In Lake Apopka  roust stonily occmrfd .it extremely
hlfih levels,  r.arbon-14 uptake  rates and chlorophvl 1  levels  t'.>r 1972-71
arc Plotted  in  KlRun- 5.  Avenue  chlurophvl I  .1 during I9/J-/1 was 84
mR/m\ .1 value  about 40 percent higher ili.ia the IV<>9-70 d  -K> mji/m1 are
usually cited .is the dividing point  ln.-l'vv.-n -.-I iy'.ot ruphli.- .uul i-iitroplili-
lakes.

Althougli there  are SOIHL- tcmpnr.il  fluctu.it Ions In bi>:li  c'lloroj-liyl 1 J .uul
productivity v.i lues, seasonal ln-iuls are not pronmnici-J.   Tlio mlninun
chlorophyll  a level otcurroc!  In iVK-lier ('•(> nn/m1) .mil Llie n.iiinium pro-
ductivity  occu.-n-d  In Miy and Dfi. ember < about !).«.? ••   "Vn'-hr).  Ilixiruni
productivity oiLiirn-d In .lime  (U.^l; (./n'-hr) .ind  ill-  me in over the
period of  measurement w.is 0.4J« C/p'-lu, a  hli-.h i.ite  .ilso  liulic.it I vt- of
hy pt1 re u trophic  condlt lens.

He.isur«:ment  of  total communliy  m..-t abol Ism iislnj-. dLurn.il  oxyKfii riianpos
(Odum nnd  Hoskln. I9'i8) also  showed 1-nko Apopk.i to be  extremely produc-
tive.  Ft Run* f> shown tvpiral (ilurnal ili.murs i'i ilissulvi'd nxvijen and
tcmpcraturo  as  me.-iHiired  in August , 197.J.   Ihe .-ompiiteil prndm 1 1 vl ty
va);-.04 for four ilan-s in  l')7)  are within  lli.-  raiino  for
some of  the  world's most prod-ict ive systems durliiK favorable periods
(Odum 1971).  Tin- met.-ihullc .ictivitv of Lake Apoi'ka  is largely dnu to
the activity of blue-Kl'eon II!R.IO  l'llcrocvst_is sp..  lj»Ji£l»y..i 'M'-. Osclll.i-
t^L1JLsP" and  A.I1il.l.l:l«llH sl'-)-   I'loJnctlvliy levels In Lake A|>.'pk.i com-
pared closely to values  found in  tropical  K-is. African lakes (Mel ads
and Ktlh.ira 1971).  These  la's**  wore also  shallow  (7.  loss  ili.m  2 m) anil
were dominated by cyanophyti-s (m-ilnly  Sj^tjriiljji;!. sp.).
An  Initial uurvt-y of  Lake Apnpk-i sediments was  mado for the purpose of
determining the |ihynlcnl  and i:hfnical rliaracter 1st ics of muck  from five
lake loratlons.  This survey was done to ev.iluato the variability of
sediment conditions within the lake and deternlr.r iin- di-Rree to  which
flcdlncntn collectfd  for the larRe scale drawdown slmulat ions reflected
the average lake sediment conditions.  The  results shown In Tab If 6
Indicate that muck sediment l» rlrh  in nutrients nnd organic matter.
The high water content reflects llio hit;; ly  f l-»cc-ulent nature of  the
Buck.  No definite sediment -water Interface  exists In tin-  lake and the
muck freely mixes with ovi-rlylnR water.
 In order to as SPSS  Hcdtmcnl nutrient  role.-isc potential, ^M'.ckman
 enmpleB wore taken  froa points over the  entire lake (Fir.nro  7)  and
 examined for Interstitial water nutrient content.  Table  7 shows that
 the lntcrsttti.il water Is quilt: rich  In  amm-.ii.i and orthophotph.ite,
 which Is released to  provide alR.il nntr ii-nf when mix I tip,  of  the sedi-
 ments and water occurs.  Avcrar.c orthophnqph.ite in the  lnterstlti.il
 water of all the samples was 3.23 mR  I'/l. while for murk  sediment alone
                                     20

-------
IUU
90
80
~ 70
i**
J 60
^
O
3 300
1
p
cr
rt
PRIMARY 1
o
o

' *•> X
\ ' " ^ ^ X

' \ ^""-^ /^ -"''
\ /' "
v/
V
	 . PRIMARY PRODUCTIVITY


 JUNE     JULY     AUG     SEPT    OCT      NOV     DEC      JAN     FEB   MAR

                         1972                                       1973

Figure 5.  Chlorophyll £ and  carbon-14 productivity at  Station 3,
          Lake Apopka

-------
                    DISSOLVED OXYGEN


                    TEMPERATURE
                                                                         25
    M
0600
1600
 14 '




 12-
zlO'
tu
o

is
o
to
CO
                                                      .30
                                                         ID
                                                         ce.
                                                                            ee
                                                                            UJ
                                                                            a.

                                                                            UJ
                                                      25
                                                                         .20
    M
0600
                               HOUR OF  DAY
1800
      Figure 6.  Diurnal changes of dissolved oxygen and temperature,  Lake

                 Apopka, August, 1972
                                       22

-------
                       TABLE 5.   COMMUNITY METABOLISM  IN LAKE APOPKA
Station and Chlorophyll Cross
Date a (ag/a2) Production
(g 02/«.2-day)
8/18/72
Station 3 74 34
10/30/72
Station 3 46 6
12/5/72
Station 3 68 18
4/11/73
Station 3 121 38
Station 1 105 32
Hld-lake 130 38
North Shore 98 26
Community Solar Pe'R Percent
Respiration Energy Efficiency
(g 02/B7-day) (Kcal/r».2-day) 
-------
TABLE 6.  COMPARISON OF MUCK FROM DIFFERENT LOCATIONS IN LAKE APOPKA
a b
Location TON3 NH3-N Total P
Winter Garden 28.9 0.11 0.93
Boat Ramp
Sandy Point 23.2 0.10 0.41
Gator Islar. 25.4 0.10 0.32
Winter Garden Pier 28.1 0.15 0.62
Station 3 29.6 0.15 0.62
2 H20 Volatile
Solidsc
92.5 53.1
95.6 70.2
97.3 67.4
96.8 67.2
96.8 67.2

a      ng N/g (dry weight)




'»      OR l*/g Wry wlRht)



       percent of dry weight
                                     24

-------
         Apopka
         Beauclair
         Canal
                                            Sunshine State  Parkway
Figure 7.  Lake Apopka sediment sampling locations
                                     25

-------
TABLE 7.  NUTRIENTS IN INTERSTITIAL WATER OF LAKE APOPKA
Station bottom Typo
1
2
3
4
5
6
7
8
9
10
11
12
13
14
IS
16
17
IB
19
20
21
22
23
24
25
Peat Bottom
Hard Sand and Rock
(not analyzed)
Peat
Huck and Peat Mixed
Huck and Peat
Sand
Muck and Peat
Muck
Soft Mud
Muck
Muck
Sand
Muck
Muck
Muck
Muck
Sand
Muck and Peat
Muck
Peat
Muck
Muck
Muck
Muck
Peat
•Jiii iconiratjon i.i
Ortho-Phosphate
ing P/l ng P/g
div wt
1.40
-
1.40
2.50
2.30
1.30
2.70
3.40
S.OO
6.40
S.OO
1.70
4.20
5.80
3.00
6.20
1.10
1.75
3.50
2.50
4.80
6.30
0.60
1.30
0.003
—
0.001
0 02
0.001
0.0002
0.02
0.03
0.03
0.10
0.08
0.001
0.09
0.09
0.04
0.12
0.0001
0.02
C.03
0.002
0.07
0.09
0.004
0.02
i In t M r 1. t 1 1 1 .1 1 '..'a L e r
Ammonia
mg N/l
3.30
••
4.00
10.5
6.00
6.00
10.5
il.O
>13.0
>13.0
>13.0
10.5
>13.0
>13.0
13.0
>13.0
5.30
10.5
13.0
10.5
>13.0
>13.0
7.20
13.0
Not enough interstitial water
analyze
mg N/g
dry wt
0.008
—
0.003
0.08
0.003
0.001
0.08
0.11
>0.22
>0.21
>0.20
0.006
J-0.27
>0.2l
0.19
>0.25
0.0004
0.09
0.11
0.009
>0.19
>0.18
0.05
0.18
to
                             26

-------
the aveiage was 3.74 mg P/l.  Corresponding mean values for ammonia were
about 10 and > 13 mg N/l.  Muck sediments, therefore, have the greatest
potential for nutrient release to overlying waters.   They are more floc-
culent than other sediment types and hence more likely to be disturbed
by physical agitation.  Furthermore their Interstitial waters contain
higher concentrations of dissolved nutrient compared to the Interstit-
ial water found In peat, sand or mixed sediments.
                                  27

-------
                          SECTION VI

               EFFECTS  OF  DRAWDOWN ON SEDIMENTS
PHYSICAL EFFECTS

 Preliminary Studies

 For preliminary aquarium studies,  muck  sediment was obtained  from  the
 northeastern section of rhe lake by Ekman dredge.  Wet sediment was
 added to glass aquariums to produce depths of  I inch  (2.5 cm), 3 inches
 (5 cm)  and 6 inches (15 cm).  All  were  allowed to dry in the  sun for
 14 days.  During that tine, moisture loss, consolidation and  volatile
 solids  were aionitored.  Following  the two-week drying period,  the  aquaria
 were filled with eqaal volumes of  Gourd NecV Spring Water (uie principal
 gro-. ndwater source for the lake),  and water quality in the  aquaria was
 monitored for the next 28 days. The aquaria were maintained  at 25°C and
 received approximately 0.45 langleys/min (1024 foot candles)  of light
 during  a 17-hour daily photoperiod.

 Figures 8   and  9  show the moisture loss and extent of consolidation
 for the three depths of sediment.   The  original wet sediment  was 96.4
 percent water and the dry sediment contained 65.6 percent volatile solids.
 As might be expected, water loss  and consolidation were dependent  on
 the original sediment depth.  Deeper sediment  lost water more slowly
 and hence consolidated less in the same amount of time  thrn did  the
 more shallow sediments.  After cracking occurred, consolidation mea-
 surements were discontinued since bot'.i horizontal and vertical shrink-
 ing occurred.  No detectable decrease in volatile solids occurred  as
 a result of drying indicating that consolidation  is  due primarily  to
 water loss.  Although consolidation measurements were discontinued
 following refill, it was apparent that only  minimal  swelling  of  the
 consolidated sediments occurred.   Tin refill water quality  parameters
 are discussed in following section?.

 A second drying experiment was set up to test  the affect of subsurface
 drainage on consolidation rate.  Two 76 cm sediment  columns 7.5  ens in dia-
 meter were allowed to consolidate in the sun  for 24  days.   One column
 was provided with free drainage through a bed  of sand while the  other
 was allowed to lose water only by evaporation.  Figure 10  shows the
 consolidation rate in the drained and undrained columns.   One da«
 after tlie columns were filled with sediment, a zone  of clear water
 appeared above the settling muck  in both columns.   On the  second day
 this water was siphoned off and analyzed (NH3-N = 45 mg/l and o-POA-P -
 10 mg/ft).  The dashed lines in Figure  10  show the immediate effect of
 the removal of the supernatant on the rnte of consolidation and indicate
 that drainage of the  sediments from  the bottom provides for much better
 consolidation  than does surface evaporation alone.   The results of this
 experiment are summarized  in  Table   8   .   As in the aquaria experiment,
 no significant loss  in volatile solids occurred.  In the draii.ed column.
                                  28

-------
          100H
           75
         oc
         IU
         H
         z
         IU
         o
         oc
         Ul
                                                        15cm
                                                       DEPTH
     7.5cm
   DEPTH
                                                       2.5 cm
                                                     DEPTH
                        4     6    8   10    12

                           TIME (days)
14
Figure 8.  Effect of sun drying on moisture content of nuck sediment

          (aquarium experiment)
                                 29

-------
                           2.5 em
                          DEPTH
                                     7.5cm
                                    DEPTH

                            ®l            15cm
                              1®—@   DEPTH
                    !4    6    8    10   12
                             TIME (days)
14
Figure 9.  Consolidation of muck sediment upon sun drying (aquarium
          experiment)
                                30

-------
  -10
                        UNDRAINED COLUMN
 -25
                                 DRAINED COLUMN
                     8      12       16

                       TIME (days)
 i
20
24
FiRure 10.  Consolidation of a drained and undrainod column of muck

           sediment.  Daahcd lines Indicate decrease In sediment

                  dfiedfHy th%5cmoval of overlying water.  Initial
                  depth waa 76 en.
                             31

-------
TABLE  8.  SUMMARY OF PHYSICAL CHARACTERISTICS  OK  MUCK IN COLUMN
          EXPERIMENT
D.nto
5/31/72
6/01/72
6/02/72 prior to H^O
drawoff
6/02/72 after H2C
drawoff
6/06/72
6/08/72
6/12/72
6/14/72
6/22/72
% H20
96.4
88.5
81.0

71.3
63.8
60.'-,
52.1
45.8
35.4
itiil Col nnn
% Solids
3.6
11.5
19.0

28.7
36.2
39.6
47.9
54.1
o4.6
7. VolatiK-1
65.5
66.8
64.3

6^.8
66.9
67.5
66.3
66.9
 Date
 Undrained Column
H20         % Solids
5/31/72 (Initial
6/01/72
6/02/72 prior to H20
drawoff
6/02/72 after H20
drawoff
6/06/72
6/OS/72
6/12/72
6/14/72
6/22/72

a % volatile •
96.4
94.5
93.6

80.7

80.6
80.3
79.5
78.4
76.5
- wt - fired wt :[ lm
dry wt
3.6
5e
.5
6.4

19.3

19.4
19.7
20.5
21.6
23.5

                                                                f-S.5
                                                                6 -'..3
                                                                6.1.. 5

                                                                65.fi
                                                                66.
                                                                61,
                                                                66,
                                                                66
                                                                66.8
                                     32

-------
water content decreased from 96.4 percent to 35.4 percent after 24 days
of drying.  The. minimum value  for the undralned sediment was 76.5 per-
cent water.

Lake Simulations  (Columns and  Tanks)

For Initial lake  simulations,  60 and 120 cm of muck sediment were
placed In 19 plexiglas columns and 6 metal tanks.  Three columns and
one tank were not drawn down and hence served as controls.  The sedi-
ment drying periods ranged from 69 days to 175 days for those which
were subsequently refilled.  Some sediment In tanks or columns devel-
oped leaks and remained dry for as long as 336 days.

Figure 11 shows the percent consolidation defined as loss In volume as
a percentage of the original volume of the sediment.  With a drying
time of about 170 days, consolidation averaged 63 percent.  The aver-
age consolidation for a 70 day drying tine was 46 percent.  Although
there are higher and lower values, around 40 to 50 percent consolida-
tion appears to be the point where an equilibrium is established
between drying and rainfall.  One tank was followed for 500 days fol-
lowing refill and maintained 30 percent consolidation throughout the
period of observation.  The significant point is that once sediment
Is consolidated,  it remains In that state for a long period of time
following refill.  In Lake Apopka, such consolidation would Increase
the water depth In the areas with deep (e.g. > 2 m) layera of muck
that dried well.  The increased depth, as well as the more rigid struc-
ture of the consolidated sediments, would lessen the likelihood of
sediment suspension caused by wave and outboard motor agitation.

Consolidation is due primarily lo water loss.  Figure 12  shows the
average loss after 168 days of drying to be 33 percent.  The dried
sediments contained 25 perct \i less water 150 days after refill than
did the control sediment.  Apparently, the physical structure ol the
consolidated sediment Inhibits reabsorptlon.  Volatile solids (Figure
13) did not show significant changes, further indicating that the con-
solidation Is due primarily to water loss.

During our sediment sampling oft' the northeastern shoreline of Lake
Apopka, we found that an area of peat sediment extends fron the shore
some 100 meters out into thr lake.  The peat is apparently scoured by
wave action and Is not covered by muck sediment.  Blocks 60 cm deep
were pieced in each of 3 plexiglas columns.  The peat consists pri-
marily of snarled fibrous plant roots, apparently remnants of previous
rojted vegetation.  This material docs not consolidate upon drying.
The 3 test columns were sun dried for an average of 180 days and the
mean percentage consolidation was 7 percent.  Two control columns of
peat, each containing 60 cm, showed greater consolidation in spite
of the fact that no drying occurred.  The organic content (volatile
solids) of individual peat samples varies between 80 and 90 percent
of the dry weight.  In this 3 column experiment, no change in the
organic content of peat could be detected with drying times an long
as 189 days.  The water content of the peat also ranges between 80 and

-------
  so-



   40-



   30-



|?20



SiulO
o>

8-3
3S*-  0
  -20
i
  -40
                                               •-	-• CONTROL


                                               •         • TEST
              INITIAL     POST

                  DRYING
                                  30
                                           60
                                                    90
                                                             120
                                                                       390
                                                                                500
                                            DAYS AFTER  REFILL
    Figure  11.  Consolidation of muck sediment  (column-tank study)

-------
  100
  80
1.60
3)40
CC
111
$20-
                                                      	•CONTROL

                                                      	»TEST
             INITIAL   POST     30
—T—

 60
—r—
 90
                                                            120
                             150
                                   DAYS  FOLLOWING REFILL
  Figure 12.  Percent water in muck sediment  (column-tank  study)

-------
  100H
S 80-
o
O
U

s
s
s
   40
   20
                                               •-	• CONTROL
                                               •———•TEST
INITIAL     POaT     30
     DRYING
                                          6O       90       120      150
                                DAYS FOLLOWING  REFILL
  Figure 13.  Volatile  solids In muck sediment  (column-tank study)

-------
  90 percent.  The drying  results showed great variation  among  samples.
  Apparently, peat dries better  if  it  is loosely packed.  It also reab-
  sorbs moisture more  rapidly  if loosely packed.  In areas of the lake
  where extensive peat depusits exist, there will be little increase in
  water depth resulting from drawdown and drying.  Muck sediment on top
  of peat consolidates abou'. the- Hame  (38 percent in 151 days) as muck
  alone.  From our observations, this phenomenon does not occur commonly
  *n rhe lake.  The peat we observed was near shore and not covered by
  large amounts of muck sediment.  During drawdown, muck; which does cover
  the peat will probably move with the receding water Into deeper areas
  of the lake.  In order to deepen the lake, especially in the littoral
  zone, it may be necessary or at least desirable to remove some peat  by
  physical means during drawdown.

  Lake Simulations (Pools)

 For the final lake simulation experiments, approximately 30 cm (1  ft)
 of muck sediment was placed In each of 5 plastic swlinnlng pools 2.4 m
  (8 ft)  In diameter and 45 cm (18 in)  deep.  Two pools (a and y) were
 drawn down and dried for  4 months.  The other 3 pools (6.  A. c) served
 c« controls.   The major difference between the  pool and  the  tcnk and
 column  simulations was the fact that  the  water  In all the pools was
 continually stirred by subnerslble »,atcr  pumps.   Additionally,  the sur-
 facft  to volume ratio was  greater  in the pool  experiments,  thus  afford-
 ing greater wind mixing,   following refill of the  test pools,  some 17
 parameters were  followed  for  an additional four months.   The physical
 parameters for the sediment  Included  percentage water and  volatile
 solids.  Although  sediment height  (from which we  calculated  consoli-
 dation  rates  In  the  previous  experiments)  was followed,  the  data --re
 unreliable because  of sediment  movement cnuaed  by  forced  clrcularlon
 of the  overlying water IP the ••ntrol pools.  Changes oc.-urrlng as a
 reiiult  of  factors other than  sediment shifting could  not be detected.

 Figures 14  and 15 show the water and volatile solids  content of the
 sediment before, during and after diyit,R.  A  two wjy analysis of vari-
 ance  (see  appendix  for details) was used Co determine significant dif-
 ferences between the  test  and control pools.  Although the lines on
 the figures represent means for all results obtained, the data were
 statistically analyzed for differences between test and control pools
 by grouping the data  In three ways:   (1) data from all 5 pools  for the
 local tie  period: (2) data from 1 teat pool  (Y) and 2 control pools
 (B. A) for  the total  time  period and. (3) data from all 5 pools taken
 during the  last two montha of analysis.  The second and third groups
were used because leaks occurred In control c and In test a.  These
 sediments were Immediately transferred to new pools but were disturbed
 in the process.

The water content of the sediment  (Figure 14 ) shows a statistically
significant difference (P < 0.001)  due to drying when the data are
grouped as in (2) and (3)  above.  The difference between  test and con-
trol  for all pools for the total tine period was not  significant (P -0.05),
                                  37

-------
  95-
  90-
Ui

I
IU
  80
                                                                            CONTROL
                                                                            TEST
        BASELINE DATA  0
4      8      12
WEEKS OF DRYING
 4      8     12     16
WEEKS FOLLOWING DRYING
                                                                                20
24
     Figure 14.  Percent water in muck sedincnt (pool study)

-------
   70
 "
X
S
g
2
g
  60-
  40-
  30<

CONTROL
TEST

^
DRAWDOWN
BASELINE DATA 048
^
REFILL
17
12 0
r 	 1 	
4 8

	 1 	 1 	 1 	 T
12 18 20 24
                           WEEKS OF DRYING
                                                       WEEKS FOLLOWING  REFILL
   Figure 15.  Volatile  soldia In muck sediment (pool study)

-------
Again, it appears that once drying takes place,  the new structure  of
the sedlnent prevents reabsorption.  In the pool experiments,  drying
reduced the water content from about 95 to 86 percent.   In the tank
and column experiments, the water content -f the dried  sediment was
much lower, reaching a rough average of 65 percent.  The primary rea-
son for the difference is probably the method of drainage.  The columns
and tanks were underlain with sand and equipped with drainage  valves,
which were left open during the period of drawdown.  The test  pools,
on the other hand, were pumped or siphoned following rainfalls.  The
latter method is oovlously less effective and the drying sediments
retained some rainfall.  Furthermore, the columns and tanks were dried
for over a month longer than were the pools.  The experiment does show,
however, that even under less than optimal drying conditions,  signifi-
cant water loss occurs and reabsor-ption l'i not important.

The effect of drying on the volatile solids content of the sediment
was significant wi .1 data  irom 1  test pool  and 2 controls are consid-
ered  for the total tine period.   However, Figure 15 shows that volatile
solids decreased  In both test and control pools from around 65 percent
to abojt 50 percent.  The  decline was more rapid in the test pools than
the controls, but by the end of  the study both sets were approaching
the same volatile solids level.   It should be noted that  this  is  the
only  case we observed  in which volatile solids showed a slgnlficarr
decreare as a result of drying.   The rapid initial decrease upon  draw-
down  (most of the decline  occurred  in  the  first two weeks of sediment
exposure)  suggests  that  part of  the decline  could  be attributed to
heterogeneous and non-representative samples.  This suspicion  is  stren-
thened by  the data  In  Table  2, which show  there is a natural  in-lake
volatile solids  range  of 53.?.  to 70.2  percent of muck  sediment dry
weight.
BIOLOGICAL EFFECTS

 Aquatic Hacrophytes

 Nutrient additions and mismanagement of L.ikn Apopka have resulted  in a
 phytoplankton-water hyacinth based ecosystem.  Both types of  primary
 producers die, decompose, and release nutrients which insure  their
 survival.  Both overwhelm their competitors, emergent and su'omcrsed
 macrophytcs. by creating shade, turbidity, and a flocculent sediment.
 For drawdown to be effective in improving water quality. It will have
 to cause significant changes in the primary producer component of  the
 Lake Apopka system.  Submersed microphytes would be the ideal replace-
 ment.  They would provide a habitat for fish and pcriphyton,  compete
 with the phytoplankton for nutrients, and reduce mixing velocities at the
 sediment-water interface.  Their root structures would further aid in
 preventing sediment disruption.

 According to a recent survey (Chesnue and Barman, 19/4), the  most com-
 mon macrophyte on Lake Apcpka is the hyacinth and no submersed plants

-------
  are  present.  The  fourteen  species observed by Chesnut and Barman
  are  shown  below  In Table  9.
            TABLE  9.  AQUATIC VASCULAR PLANTS OF LAKE APOPKA
                      (FROM CHESNUT AND BARMAN, 1974)
                                              Panicum paliiJivagum
 Altcrnanthera phlUxeroidea                  Panicum hcmlcomon
 Lemna sp.                                    Scrlpus valldus
 £u£har advcna                                Typha do-aiTgensis
 Pontcderla lanceolata                        Cladlum jamalcensls
 Saglttarla land folia                        Hydrocotvle umbciTa'ta
 Saglttarla latlfolla                         Jusslaea            "
 We also observed alack of submersed vegetation In Lalr» Apopka.  and
 for that reason. Investigated the sediments as a potential  source  for
 submersed macrophyte regrowth.   Accordingly, In September,  1973,  17
 sediment samples were taken by  Pieman dredge at approximately  equidis-
 tant Intervals around the perimeter of tho lake.  Seeds present  In
 each sample were separated by sieving ar.d were Identified.  Table  10
 summarizes that data.  Several  species of both emersed and  submersed
 macrophytes were found, suggesting that the potential  for regrowth
 Hoes exist In the  sediment as seeds.

 In our own lake simulation studies,  however,  the only  two aquatic
 macrophytes which  sprouted on the drying sediment were cattail  (Typha
 ep.)  and water hyacinth (Eichornla craaslpcs).   Table 11 shows a list
 of plant species which  developed  on  dried sediments  In our  tanks and
 columns.   For the  tanks,  all  plants were harvested and weighed and the
 figures  in Table 11 represent  percent  contribution of each species  to
 total  wet  weight.   The  water  hyacinth is not  present in the list because
 of its short  survival time  on drying  sediments.   It appeared shortly
 after  draining  and  lasted only  a  week or two.  With the exception of
 TyPna,  (cattail), all  the species  listed  are terrestrial and apparently
 most germinated  from *eeds  blown  into the simulations.  Cupatorlum
 (dogfcnncl) and  Typha (cattail) accounted  for the majority of the
 blomass harvested from  the  tanks  and  columns.

The species list in Table 11 contains  plants which grew on muck sedi-
ments exposed for about 7 months  (non-refilled tanks and columns).
Sediments dried  for leas time had  fewer species growing.  The first
plant growth appears about  5 days after drying begins and consists of
blue-green eploellc algae.  After about 20 days, the algae Is replaced
                                   41

-------
 TABLE 10.   SEEDS FOttZD IN  17  SAMPLES OF  SEDIMENT FROM LAKE APOPKA
            (SEPTEMBER 5, 1973)
 1.
 2.
 3.
 4.
 5.
 6.
 7.
 8.
 9.
10.
11.
12.
13.
14.
15.
16.

17.
18.
19.
20.
21.

22.
23.
24.
Najas guadalupcnsis
Sambucus canadensis
Potamogeton sp.
Brasenia schreberi
Acnldia or Amaranthus sp.
Cladium jaaaicensis
Elchhornia crassipes
Cyperus sp.
Sclrpus sp.
Heteranthera sp.
Nymphaea sp.
Sagittaria lattifolia
  rrica cerifera
Rumex cripus
Phytolacca americana
Cicuta maculata or
Conium maculatum
Jussiaea leptocarpa
Anar.haris sp.
Cephalanthus occidentalis
Magnolia grand!flora
Proserpinaca palustris or
Hyriophyllua sp.
Pontederia  condata
Zosteia marina
Graminae
Southern naiad
Elderberry
Pond weed
Watershield
Water hemp
Sawgrass
Water hyacinths
Sedge
Bulrush
Waterstar grass
Waterlily
Arrowhead
Wax myrtle
Curly dock
Pokebcrry
Water hemlock

Water primrose
Watcrweed
Button bush
Magnolia
Mermaid weeH
Watermilfoil
Pickerelweed
Eelgrass
Unidentified  grass

-------
TABLE 11.  PLANTS COLONIZING DRIED SEDOTNT (EXPRESSED AS PERCENTAGE
           OF TOTAL DRY WEIGHT)
Plant Species
                                Tanks            Columns
                               1      2123456
Amaranthus autralis
Baccharus halimifolia
Celtls laevigata
Cocculus carolinus
Conyza bonariensis
Conyza canadensis
Conyza parva
Cynodon dactylon
Echinochloa waiter!
Eupatorlum capillifolium
Eupatorium leptophyllum
Phytolacca riglda
Sambuccus Simpson!!
Solidago sp.
Typha latifolia
Urtica urens
Unidentified
                               2Z
                                    .22
.01%   .(
                              86%   94X"
                                     5Z
                              iu
                            <.012
                            <.01Z
                                           X   X
                                               X
                                               X
                                                   Xa
                                               X   X
                                                            X   X
                                                            X    X
a
b
    X indicates presence only for columns
    Fupatorium leptopliyllum, Conyza canadensis and Conyza bonariensis
    were weighed together
                                   43

-------
by moss and actlnomy;etes.  These forms are slowly replaced  by  the  vas-
cular plants, which do not appear until about 60 days  after  drying
starts.  The first vascular plant to appear was Typha  (cattail)  on  peat
sediment.  About 2 weeks later Typha and Elchhornla (hyacinths)  sprouted
in the muck sediment.

The only plant which survived refill In the tank and column  experiment
was Typha.  Following refill, however, Chara braunii (stonewort), a
macroscopic green benthlc alga, grew on the sediments  of 1 tank (69
day drying period) and 3 columns (175 day drying period). The  Chara
remained viable In the tank for over a year, at which  time the  tanks
and columns were disassembled.  Transplantation of the Chara Into
other columns or aquariums is easily accomplished by sprigging.

The same general pattern of plant succession occurred in the pool
experiment.  The biomass of the plants harvested from the test  pools
following 4 months of refill and the control pools arc shown in Table
12.  One test pool supported equal growth of Typha and Chara while the
other contained primarll> Tyjrtia.  Over 2 kg of Chara was also harvested
from one of the control pools.  This was surprising, for Chara has not
been reported as growing  in Lake Apopka.  Apparently, the laboratory
simulations favor  the growth of Chara. and we  cannot state positively
that drying is  the sole factor responsible  for  its growth.  However,
the control pool  In  which Chara grew was the pool which  sprung a leak
and partially dried  before the sediments could be  transferred to a new
pool.   Perhaps  the partial drying stimulated the growth  of Chara.
        TABLE 12.   PLANT BIOMASS HARVESTED FROM POOL SIMULATIONS.
    TEST POOLS HARVESTED FOLLOWING DRAWDOWN AND FOUR MONTHS OF REFILL
 Pool                           Plant Harvest (kg)
                      Chara            Typha          Other
Test (a)
Test (T)
Control (6)
Control (A)
Control (c)
6.97
0.61
0
0
2. 59
6.52
16.9
0
0
0
0.20
0.42
0
0
0

-------
  One  Impediment  to  the establishment of Chara and Typha In Lake Apopka
  Is the water hyacinth.  According  to Chesnut and Barman (1974), as well
  as our field observations, mats of hyacinths can flatten and destroy
  stands of emergent  littoral vegetation.  The mats also shade submersed
  vegetation.  A  continuing hyacinth removal program (proferrably mechan-
  ical) may be necessary to prevent  the destruction of cattail stands
  which will become established during and following drawdown.  Protec-
  tive fencing measures will probably be ineffective due to the large
  hyacinth blooass and wind forces Involved.  The cattail would occur
  In the shallow water areas of the lake (<1.3 m).  Chara. should It
  become established as a result of drawdown, might grow in deeper water.
  There, it would less likely be adversely affected by hyacinths, which
  tend to blow into shore.  Vhen hyacinths are in deeper water, we have
  noticed they occur as small clumps, which would not significantly
  shade submersed plants.   The establishment of Chara.  however, is not
  certain.   Two slRnlficant factors affecting its growth are water tur-
 bidity and caiciua content.   If turbidity can be kept low  during refill
  its growth seens likely.  The calcium content of Gourd Neck  Springs
 water (the primary means by which the lake will be  refilled)  is moder-
 ately hard (Ca* 2S ppm). which should be adequate for Chara  growth.

 Benthtc tnvortebratca

 The Lake  Apopka benthic  Invertebrates  sampled by the  Florida  State
 Board of  Health during their 1962  to  1964 survey (undated  report)
 Indicated oxygen stress.   The  predominant forms were  sludgcworros  (Oli-
 gochaeta), bloodworms (Chlronomidac) and  phantom midges  (order  Diptera,
 family  Culicldac).   These  three  forms were  found consistently in  the
 muck  sediment.   At  several muck  bo-.;om stations, no Invertebrates were
 found.  Both sludgewonas and bloodworms contain erthyrocruorln, a hemo-
 globin-like pigment which allows them  to  live at low  levels of dissolved
 oxygen  for extended  periods of time.  When  they are the only  forms
 found  (low diversity), oxygen stress Is usually  the reason.  The phan-
 tom nidges are capable of vertical migration and hence are considered
 poor  indicators  of water quality.

 Other benthic invertebrates found In the early  1960's included mayflies
 (class Insecta, order Ephcmeroptera), scuds (class Crustacea, order
 Anphipoda), and  clams and snails (phylum Mollusca, Classes Pelycypoda
 and Gastropoda).  None were found frequently and they usually were not
 present In muck sediment.  These organisms, taken as a group, indicate
 adequate dissolved oxygen.

 In 1970 and 1971, Florida Technological University conducted  a eurvey
 of the benthic invertebrates of the Oklawaha chain.   Their raw data is
 included as Appendices B-l to B-4 in the Florida Department of Pollu-
 tion Control's Restoration Plan (1972).  The Apopka  survey consisted
of a one year seasonal sampling of about  30 stations.   Again,  the pre-
dominant forms were bloodworms, sludgcworms and phantom midges.   A few
 leeches (phylum Annelida, class Hirudlnca),  which also can  tolerate
 lov dissolved oxygen, were found.  Scudo  and snails  were occasionally

-------
observed.  A total of 2 mayflies, the least pollution tolerant  form
found, were recovered from the 30 stations for the 4 sampling periods.

Samples of flocculcnt muck and peat from the northeastern section of
Lake Apopka taken for simulations In this study contained no benthic
invertebrates.  The only signs of life were shells of long dc.iu snails
(mostly Vlviparus sp.), which were abundant in tauck and peat and pro-
bably predate the 1947 hurricane.  In order to minimize sediment per-
turbation, we did not monitor benthic inveitebrates during the course
of the experiment.  We did observe, however, that prior to drawdown,
the plexiglas columns and tanks were rapidly colonized by bloodworms.
Possibly in our initial examination of the sediments for invertebrates,
eggs or very early instar larvae passed through the U.S. Standard No.
30 sieve (595 micron mesh) and were therefore missed.  It is more
likely that adult female chironomids ]=iid eggs on the surface of the
tank and column water covering the sediments.  The simulations were
near a sewage treatment plant oxidation pond, where large numbers of
chironomids thrive.

About 3 months after the control simulations were constructed, the
trophic structure shifted from a phytoplankton dominated system to one
in which comparatively large biotnasses of zooplankton were supported
by cpipulic algae.  This shift was probably due to the cessation of
nixing, which, in the lake, keeps the phytopl.inkton supplied with nu-
trients by accelerating sediment-water exchange.  The same phenomenon
occurred in the test simulation  following refill.  The green filamen-
tous alga, Oedogonium. was the predominant cplpellc growth form.  In
one control tank, this alga covered the bottom to a depth of several
centimeters and supported a community of diving beetles  (order Colc-
optera,  family Dytisi iclae), dr.igonfly nymphs  (order Odon.ita), and an
occasional water  scorpion  (order llemiptera,  family Nepldae, genus
Ranatra).  These  unusual biotic  conditions, which we  felt were the
result of  the lack of mixing, were responsible for  the disassembly
of the tank and column experiments and the assembly of the mixed pool
simulation.

The lake sediment placed in the  pool simulations also contained no
living benthic invertebrates at  the start of  the experiment.   Follow-
ing drying and the 4 month refill period,  the benthic invertebrates
were  recovered from  the test and control pools.  Recovery was  by dip
net,  sieve and siphon.  The results are shown  in Table 13.  Although
the figures in Tablel3arc not on a quantitative areal ba^is,  they  do
represent  the results of equal sampling effort per  pool  and hence can
validly  be used  to compare test  versus control benthic  invertebrate
populations.  The diversity  indices presented  in Table 13  are  Shannon-
Weaver  (base  2)  family diversity indices.

In  interpreting Table  13,  probably  the most  significant  point  Is  that
diversity  in  the  test  pools was  higher  than  in the  controls.   Since
the number of replicates was  so  small, no  statistical analysis was
possible.  The diversity of  control  c  (1.94),  however,  is  probably
not significantly different  from that of  test pool  Y (2.14).   Control
                                    46

-------
TA'JLE 13.  BENTHIC INVERTEBRATES POPULATING POOL SIMULATIONS FOLLOWING DRAWDOWN AND REFILL.
           :iUMBER EXPRESSED ON THE BASIS OF EQUAL SAMPLING EFFORT PER POOL


Phylum Annelida
Family Tubiflcldae
Phylum Arthropods
Class Insecca
Order Dlptera
Family Chlrononidae
Order Ephemeroptera
Family Caenldae
Family Leptophleblldae
Older Hemlptera
Family Notonectidae
Family Nepldae
Order Odonata
Family Coenagrionidae
Family Libellulldae
Phylum Chorda ta
Class Amphibia (unidentified
tadpoles)
Total
Number of Families
Diversity Index
Test
o





11

1
1

IS
3

22
16

11

80
8
2.55
Pools Control Pools
Y B A e

1



15 41 8

1 2
2 12 3 7

1362
1

8
13 4 2

3

29 34 52 20
7444
2.14 1.72 1.05 1.94

-------
 pool c, as explained earlier, developed a leak and  the sediments had
 to be transferred to a new pool.   As a result  of  the  leak and  water
 loss during transfer, the sediment in c underwent some drying.   In
 reality,  therefore,  control  pool  L night be considered intermediate
 between a tost and a control  pool.  This hypothesis Is strengthened
 by the fact that ClKijra_ grew only  In control  pool  c  ••:"! tlio tun  tc"U
 pools, a  and Y.  No  macroscopic plants grow In ultnur  of  the oilier
 two control pools (3 and A).

 The presence of the  cmeraed  and submersed plants  provided  a habitat
 suitable  for a greater diversity  of benthlc  invertebrates  in the test
 pools. The predominant organisms In all control  pools were bloodworms
 (Chlronomidac).  In  the test  pools, dragonfly  nymphs were tin: nose
 numerous  forms, with air breathing hemipterans a  close second  in test
 o.   Since the pool simulations wore located  in Cainusvllle, the ben-
 thlc fauna which develops following drawdown m.d  refill  in Uike Apopk.i
 might  be  slightly different.   If  submersed  aquatic  plants, either an
 alga such .is Ch.ira or vascular plants,  develop, houe/cr. an Increase
 in  benthlc Invertebrate diversity and blomass  can be expected.   Although
 we  did not measure biomass,  the combined weight of  the t/ulpoK-s In  the
 test pools easily surpassed  the combined biomass  of the  invi-rti.-br.Ttos
 recovered from the control pools.   The  shift In primary producers from
 phytoplankton to macroscopic  plants should be  evident  all  the way up
 the food  chain.   Fish should  respond positIvely to  the Increased bot-
 tor stability,  the increased  abundance  of  their invertebrate food
 source, and the flngcrllng habitat  provided by the  plants.
CHEMICAL EFFECTS
Effect of sediment drying on major nutrient forms and general chemi-
cal characteristics of the sediments appear to he ntnim.il.  Although
variations occurred both temporally in a given simulation and between
dried and undrlcd sediments, they were generally no greater than those
attributable to sampling variations resulting from sediment hetero-
geneity.  Figures 16 shown measurement of total phosphorus and total
nitrogen in the column and tank simulations.  The apparent reduction in
phosphorus in the dried sediments is not so significant as it first
appears In Figure 16.  Phosphorus levels within a pool were found to
vary by about + 20 percent nreally.  The arcal variation In total nitro-
gen was about + 21 percent, making the reduction below control levels
130 days after refill barely significant.  Free ammonia was highly
variable in the column and tank sediments (Figure 17) and represented
only about one percent of the total nitrogen.  Free ammonia levels were
always less than 0.4 mg/g dry wt. whercns TKN averaged 30 mg/g dry we.
of sediment.  Ammonia in the undricd sediment (control) appeared to
rise slowly during the period of study.  The drying (test) sediments
showed higher ammonia levels during most of the drying period.  Ac
the end of the dr&wdovn period, however, the ammonia concentration of
the dried sediments was below that of the control sediments.
                                   48

-------
  
-------
     I.OOi
     0.80
LU
ce
a


o
o
      0.60
,5    0.40
      0.20
                      20
40
                                         TEST


                                         CONTROL
 60          80



DAYS  OF DRYING
                                                                    100
                                              120
140
  Figure 17.  Changes in ammonia nitrogen in column and tank sediments

-------
  Table U shows results of sediment cores taken from a control  and dried
  sediment tank.  No significant pattern of stratification in TON and
  total P can be seen for either series ot cores.   Data for the  dried
  cores gives further weight to the conclusion that drying results pri-
  marily in physical changes.  The averaKe TON increased  about 15 per-
  cent in the dried sediment core, but  total P stayed about th-.>  same.
  If substantial oxidation had occuvrcd,  the phosphorus and perhaps  the
  nitrogen would act as conservative substances and their  concentrations
  on a dry weight basis would have Increased.

  Sediments from control and test  tanks have also been  examined  to de-
  termine the forms (speclation) of nitrogen and phosphorus.  Fractlona-
  tion of sediment  phosphorus was  done according to a modified Chang and
  Jackson procedure for soil  analysis (Peterson and Corey,  1966).  To
  avoid changes  in  sediment  characteristics  the studies were done on
  fresh wet and  dried  sediment  rather than oven-drleu sediment.   Inter-
  stitial water  was  removed by  centrifugatlon.  Nutrient analyses were
  performed on the  supernatant, and  nutrient fractionation  tests were
  done  on the centrltuged sediment.

  In general  the  fractionation  results show  little qualitative differ-
  ence between dried and undrled sediments with respect to nitrogen and
  phosphorus forms.  Preliminary experiments with fresh Lake Apopka
  sediment indicated the presence of phosphorus in only one fraction;
 analyses on surface sanples from three tanks  (B,  F and J) likewise
  indicated Inorganic phosphorus occurred in the reductant soluble frac-
  ion.  No phosp.iorus was found in the "easily soluble" phosphate, alum-
  inum phosphate, iron phosphate, calcium phosphate, or occluded  phos-
 phate fractlono.  Thus inorganic  phosphorus in Lake Apopka sediment
 is apparently bound entirely In the reductant soluble form (mainly
 ferric phosphate coated with ferric hydroxide).  Organic phosphorus
 ranged trom 0.12 to 0.37 mg/g in  the three sediments,  or from 32 to 56
 percent of the  total P.  Results  for the pool sediments  Indicated the
 same simple P speciation both before and after drying.  Table 15 pre-
 sents results of metal analyses on sediments  used  In the pool slnula-
 tions.  Results verify the low marl character of the sediment:  calcium
 content was  5 percent or less and some of  this was undoubtedly  tied up
 In snail shells.  The Iron levels are moderately high  and most  of  the
 Inorganic  P  was found to be sorbed to  the  iron hydroxides.

 Sediment  interstitial water  levels of ammonia and  orthophosphatc
 were  measured on three occaslona  in the  pool  simulations;  shortly
 after they were constructed  and filled,  at  the end of  the  drying per-
 iod and  three months  after  refill.   Interstitial water was separated
 from  the sediment  by  centrifugatlon and  nutrient analyses were  per-
 formed on the supernatant using the Auto Analyzer  techniques described
 In the appendix.   Results are  summarized in Table 16.  Pore water of the
Lake Apopka sediments  in the pool study  initially contained 0.1
 to 2.6 mg/1 of orthophosphate-P and from 0.3 to 1.5 mg/1 of anmonia-N.
This material would be  readily transported  to  the overlying water when
the sediments are suspended by wind driven currents and wave action.
The high initial variability among  the pools probably reflects dlstur-
                                    51

-------
TABLE 14.  NITROGEN AND PHOSPHORUS DISTRIBUTION IN SEDIMENT CORES FROM
           TANK EXPERIMENT
Control (D)
Depth Water TON Total P Water
~Cm H mg/g-dry wt *
Surface 97
5 98
10 95
15 91
20 90
50 88
Ave. 93
Surface
5
10
15
A •»
20
25

Ave.

Surface 97
5 95
10 92
15 90
20 85
50 83
Ave 90
August 11, 1972 (PreJravdown)
32 0.46 697
32 0.55 97
27 0.56 93
27 0.43 92
36 0.46 92
35 0.50 87
36 0.49
September 18. 1972






October 20. 1972
29 0.57
35 0.43
32 0.51
33 0.46
31 0.50
30 0.49
32 0.49
93
75
74
76
79
80
80
77


59
65
74
73

_
68
Dried (B)
TON Total P
mg/g-uTy wt
29
26
25
29
23
26
26
30
29
24
29
29
30
28


26
32
32
30

—
30
0.31
0.35
0.37
0.26
0.28
0.30
0 3'
0.38
0.35
0.31
0.30
0.32
0.36
0.33


0.35
0.25
0.30
0.36

	
0.32
                                     52

-------
 TABLE IS.  METAL ANALYSES ON SEDIMENTS FROM LAKE APOPKA POOL SIMULATIONS
Pool
Tea; a
Control B
Test Y
Control A
Control c
Mg
mg/g
4.18
6.63
5.10
3.92
4.77
Mn
Pg/g
3
4
5
4
4
Fe
ng/g
.13
.25
1.08
.30
.25
Ca
mg/g
19.5
33.2
51.0
27.4
45.3
 TABLE 16.   INTERSTITIAL WATER NUTRIENT CONTENT OF APOPKA POOL SIMULA-
            TION J?
Pool
Test a
Control B
Test Y
Control A
Control c
Beginning
NH3 OP
0.28
1.54
0.95
1.33
0.84
0.14
2.6
2.2
0.64
0.94
Middle
NH3 OP
.21
.20
.20
.20
.30
.13
.60
.41
.20
.16
End of
NH3
6.4
7.3
5.1
7.8
9.0
Study
OP
.15
.63
.20
.20
.97
a  ResultR in mR/1 of  pore water
                                   53

-------
barces resulting from the collection,  transport ami addition of the
    -^^^
    i,.
had  ncrea cd soLwhat in all the waters  while orthophosphate changed
SeviaS little from predrawdown levels.   The final orthophosphate
anTammonia levels fo? the dried sediment pore "««;.£" -"""J^
™wer than the controls, but considering  the varlabli  ty a.onR the
nools the significance of these differences is  questionable.  Al though
5e results In Table 16 indicate slight if any differences between
i^S "nlJlS sediments with respect  to nutrient  levels  in pore
«[«   It should be noted that dried sediments have a lower  water con-
te" and are less likely to be suspended  by turbulence than  the «n-
5r led sediments.  The potential contribution of  nutrients  fr«. pore
water of driec. sediments to th, uwrlyln. lake water  thus  should be
lower than that  from undricd sediments.

A further question  that arises concerning the effectiveness  of scdl-
« SrJIn!  is the  extent of nutrient  leachabllity  rom dr ed nn
undried sediments.  A  crlorl arRuments can be made in either of  two
directions    If  substantial oxidation  occurs as n result of  sediment
oe    ation.  then organic N and  F  forms  should ^1^«^i;;i^1-
falc inoraanic  forms,  and  reduced N  should be oxidized to nltrati.
wMcJlI reaSily leachable.  On  the other hand, oxygenated sed  nents
 are generally  considered  less  susceptible  to nutrient rele.is, than
                         Formation  of an  oxldi.ed crust at th,    i-
                          in control and dried  systems  ,	
                          nutrient leaching from  dried and undrled soot-
                           .. .  _/- ..u_ »»_u ,.n.i nnni dnt.i directly in
    order to obtain direct data on the sorptlvc-lenching  properties of
           --                             Kra sss:
  a  some of the variation also results  from differing water content
     (?.e. ml interstitial watcr/g sediment) present  Initially.

-------
water  content of NH3  and  ortho P. and another portion was dried under
a heat lamp  to  a degree simulating natural drying for several months.
Leaching and exchange measurements were made on dried and undrlcd
aliquots of  the sediment.  The  results (Table 17) Indicate an increase
in  leachable and exchangeable ammonia.  Thus drying would seem to cre-
ate the potential  from greater phosphate release.  However, this
result must  be  Interpreted  cautiously with respect to the effect of
drawdown on  the lake  proper,  The leaching experiments were conducted
under  drastic mixing  conditions, whereas the net effect of drawdown
drying will  be  to  decrease  sediment suspension by creating a stable
sediment structure.   The  lack of an Increase In leachable IIH} as a
result of drying perhaps  resulted from some ammonia volatilization
during drying.

Table  18 presents  results of several sets of leaching experiments done
on  sediments from  the  pool  simulations before and after drawdown find
refill.  Before drying all  pools had a rather uniform tmmonla leaching
rate (0.35-0.40 mg/g).  The somewhat Increased NH3 Jeachobllity in nil
sediments after the drying  period probably reflects disturbance of the
sediments in setting  up the pools (hence depressed initial values).
The post drying results Indicate lower NH3 leaching from the dried
sediments (means «• 0.47 and 0.57 mg/g) for the two dates compared to
the controls (means •  0.66  and 0.94 mg/g, respectively).  However,
the range of control  values indicates that the differences may be
less than those suggested by the means.  For phosphate the initial
leaching rates  were more variable (0.02 to 0.06 mg/g) than for Ml3.
After  drying leachable ortho P was lower in all pools, again proba-
bly an artifact of initial  sed
-------
                                                     -vr»\-a
TABLE 17.  SUMMARY OF LAKE APOPKA. SEDIMENT SYNOPTIC STUDY
                                    OP                      NHj
Fresh brfed Fresh
Interstitial Water 3.09 — 9.6
Content
Leached with Deminera- 0.106 0.315 0.88
lized Water
Exchangeable N or rb 0.24 0.42 .'.58
Dried
	

0.82

1.47
      T-esults arc mean values  (in FR N or p/l)  for 25 sanples collected
      around Lake Apopka
b     Amount of nutrient  leached  into  IN XH^Cl  (for phosphate or  2N KC1
      (for ammonia)
TABLE  13.  NUTRIENT LEACHABILITY OF  APOPKA POOL  SIMULATION  SEDIMENT*8
Pool
Test a
Control B
Test Y
Control A
Control c
Before
NH3
0.40
0.39
0.37
0.38
0.35
Drying
OP
0.02
0.06
0.06
0.05
0.05
Att-.T
NH3
.45
.65
.50
.95
.45
.:cfili
OP
0.007
0.015
0.010
0.015
0.010
bnd of
NH3
0.49
1.20
0.66
0.96
0.66
Study
OP
.003
.002
.015
.004
.003
       Results in mg H or P/B dry weight sediment

       Experiments conducted by placing -vfl.Sg (wet wt) of centrifuged
       sediment (to remove interstitial water) Into 10 ml deionized
       water and mixing for one hour.
                                    56

-------
                              SECTION VII

            EFFECTS OF DRAWDOWN ON OVERLYING WATER QUALITY
PHYSICAL - CHEMICAL EFFECTS

F*>r the initial lake simulations, muck sediment (60 and 120 CD deep)
was placed In 19 plexiglas columns and 6 metal tanks.  Three columns
and one tank served as controls.  Following a drying tine ranging from
69 to 175 days (dried a mean of 117 days), 3 tanks and 10 columns were
refilled with Gainesville ground water, which is chemically similar to
Gourd Neck Spring water.  A variety of routine water quality measure-
ments were made on the tanks and columns at weekly to monthly Intervals.
Measured parameters included physical conditions, such as temperature,
turbidity and conductance, and chemical conditions, such as dissolved
oxygen, pH, major ions and all  X and P forms.  Of primary interest
with respect to the assessment of drawdown feasibility as a lake renewal
technique are the levels of nutrient forms and related parameters in the
water following refill of the test simulations.

Orthophoaphare levels in the waters of the experimental columns and
tanks (Figure 18) exceeded control levels by relatively small amounts
(0.01 - 0.03 mg P/l) for the first 90 days after refill, but after ISO
days of refill, water above the dried sediment bccate highly enriched
(average 0.131 mg P/l), while the centals remained low (average 0.03
mg/1).  Reasons for this increase are not entirely clear, but as dis-
cussed later, the lack of adequate mixing in the tanks led to dense
growths of filamentous algae on the sediment of some of the test tanks,
and the conditions in the tanks after the onset of the.ie growths are
not considered representative of in situ lake conditions.

Nitrate (Figure 19) in the refill water, on the other hand, remained
lower In the test than in the control simulations, even after ISO days
of refill.  TV-tal nitrogen (Figure 20) was also slightly lower In the
cest tanks just after refill, and later (after 90 days of refill) the
control tank water increased markedly in total M compared to the test
tanks.  Total N levels in both test and control tanks remained above
1.5 mg/1, indicative of eutrophic conditions.  A pulse of ammonia nitro-
gen to almost 0.7 mg N/l (Figure 2.; occurred following refill of the
of the test tanks, but the level dropped ir.pidly and amnonia remained
depleted after 60 days of refill.  The ammonia levels in the controls
declined almost linearly during the study period, eventually becoming
depleted by day 180 of the refill period.

Dissolved oxygen and pH changes In the tanks are shown in Figures 22
and 23.  There were also wide diurnal variations in oxygen levels not
reflected in Figure 22.  In general, test simulation oxygen levels
were at or above saturation and exceeded control levels after 30 days
of refill.  This condition remained until 120 days cf refill.  The pH
levels varied between about 7.5 and 9.2, and generally followed the
trend in dissolved oxygen level, suggesting photosynthetic activity
as the main cause of the changes.
                                   57

-------
   0.13
   0.11
   0.09
ui
   0.07
Q.
to
O

Q.
   0.05
o:
O
  0.03
  0.01
                           CONTROL
                           TEST
             INITIAL  POST     30
                      DRYING
 60       90      120

DAYS  FOLLOWING  REFILL
150
        figure 18.  Changes in orthophosphate in column and tank simulation
                   water
                                      58

-------
 0.18
0.17
0.15
0.13
                                                              CONTROL
                                                              TEST
         INITIAL  POST     30       60       90      120      ISO
                  DRYING
                                      DAYS  FOLLOWING REFILL

    Figure 19.  Changes in nitrate in column and tank simulation water
180
                                 59

-------
   4.
z
o
	 CONTROL
•	 TEST
O
Of.

t  2J
<
o
           INITIAL  POST     30
                   DRYING
                         60       90      120      150      180

                          DAYS  FOLLOWING REFILL
      Figure  20.  Changes in total nitrogen in column and tank simulation
                 water
                                   60

-------
Ul
o
oc.
o
    0.8
    0.6
    0.4
   0.2
                                               	'CONTROL

                                               	-TEST
            INITIAL   POST     30

                    DRYING
  >         i


 60       90     120      150



DAYS FOLLOWING REFILL
180
      Figure 21.  Changes in ammonia in column and tank simulation water
                                      61

-------
  16
X
o
a
HI
   4
                                           	. CONTROL
                                           	• TEST
           INITIAL  POST      30      60
                   DRYING
90
120      150
                                        DAYS FOLLOWING REFILL
     Figure 22.  Changes in dissolved oxygen in column and tank simulation
               water
                                     62

-------
2
           •	CONTROL
           	• TEST
       INITIAL   POST     30
                 DRYING
60
90      120      150
                                   DAYS FOLLOWING REFILL
 Figure 23.  Changes In pH In column and tank simulation water
                               63

-------
Similar measurements were made on wacer quality in the test and
control pool simulations and Figures 24 to 32 summarize the results.
For purposes of this report the data for the test pools and the  con-
trol pools were averaged and the graphs show the mean values vs  time
for the two treatments (test—dried sediment—and control—undried
sediment).

Water temperature in the pools reflected the changing season and short
term weather changes, with a general trend of increasing temperature
in the weeks following refill (which occurred on Jan. 15, 1974).
Control and test pool temperatures closely paralleled each other but
there was a distinct trend for slightly warmer water (by about 1"C) in
the control pools.  This evidently resulted from increased turbidity
levels in the control pools (Figure 24), which caused greater ahsor-
bance of solar radiation.  Comparison of Figures 24 and 25 indicates
that maximum temperature differences between the test and control pools
coincided with maximum differences in turbidity.  Drawdown had a marked
effect on turbidity levels in the pools; values in the test pools ranged
from less than 1 JTU to about 4 JTU in the weeks after refill.  Turbid-
ity In the control pools was consistently higher than in the test pools,
and large fluctuations in the control pools reflected periodic sediment
resuspension.  The mean turbidity for the control pools was nearly 8
JTU and the range was 2 to 17 JTU.

Alkalinity (Figure 26) and pH (Figure 27) levels in the test and con-
trol pools shoved relatively large temporal variations during the weeks
following refill,obscuring any differences which may have existed be-
tween the two sets.  Variations  in alkalinity  resulted primarily from
evaporation during dry weather plus the dilution effect of periodic
rainfall.  Variations  in pH also  resulted from these  factors plus bio-
logical reactions such as photosynthesis and respiration.  In general
there was a trend toward higher  pH  in both  sets of pools  in the weeks
following refill with  initial values less than 8.0 in both sets.  The
pH of the test pools approached  9.0 by the  end of the study, while the
control pools  fluctuated between 8.2 and 8.6.

Ortho and total phosphate levels  (Figures 28 and 29) exhibited similar
trends in both control and test pools, with orthop'.iosphate variations
being especially close.  No initial spike of either ortho or total
phosphorus was noted in the test  tanks upon refill.  Concentrations
of orthophosphate in both sets of pools ranged  from moderate to high
during the monitoring period.  Figure 28 shows  a slight tendency for
lower orthophosphate in the test  pools, but this trend was not statis-
tically significant.  Except for  two brief  peaks of 0.11 and 0.060
mg P/l, orthophosphate remained  less than 0.050 rag F/l in the test
pools and averaged 0.028 mg P/l  (excluding  the two peaks) and 0.038
mg P/l Including all the data.  The mean for the control pools  for all
dates was 0.044 mg P/l.

Total phosphate levels  in both sets of pools started at about 0.080 mg
P/l just after refill and showed  a trend toward gradually  increasing
values in both sets  throughout the monitoring  period.  Control  levels
were higher than  test pool levels on all but one date and  the differ-
ences appeared to increase as the post drawdown monitoring period
continued.  A maximum of 0.21 mg P/l was reached in  the  control pools

                                    64

-------
                  20 -I
Ok
Ul
                                                                                         	 TEST

                                                                                         	 CONTROL
                                                 WEEKS FOLLOWING REFILL
                   Figure 24.  Weekly changes in turbidity in pool simulations

-------
   24 -
   22 -
o°
\.s
IU
g.H
   18 H
   16 -
    14
                                                                     CONTROL  —

                                                                     TEST
-T	\	1	1	1	1	r
  8        8       W       12       14      16      18
 WEEKS FOLLOWING REFILL
                       T
                        4
20
      Figure 25.  Weekly changes in temperature  in pool simulations

-------
o
u
co

<
     200
     175
     ISO
     125
     100
      75 .
                            TEST


                            CONTROL
                                      B              12


                                           WEEK
    Figure 26.  Weekly changes In alkalinity in pool siaulacions
16
20

-------
9.0-
8.6-
8.2-
                               WEEKS FOLLOWING REFILL
   Figure 27.  Weekly changes in pH in pool  simulations

-------
                 0.11-
             -N  0.085-
             o.

             (9
                 0.06-
S
             0.
             to

             § 0.035

             a.
             o
             cr
             o
                 0.01 -
                                                                                      TEST


                                                                                      CONTROL
                                     «              8              12



                                                  WEEKS FOLLOWING  REFILL



                         28.  Weekly changes In orthophcsphate  in pool simulations

-------
O.21-
018-
0.12-
O.09-
0.06-
                     I
                     4
       8              12
WEEKS FOLLOWING REFILL
                                                                               CONTROL 	
                                                                               TEST
 I
16
20
  Figure  29.  Weekly changes in total phosphorus in pool simulations

-------
  by the 15th week.   The test pool maximum occurred  simultaneously  but
  reached only 0.16  mg P/l.

  An initial spike in the aqueous ammonia  content occurred  in  the test
  pools when they were refilled,  but  this  spike  rapidly dissipated  (Fig-
  ure 30).   Consequently, the initial ammonia  level was higher in the
  test  pools (0.45 mg N/l)  than  in the controls  (0.33 mg N/l), but  the
  control pools maintained  high  levels (0.3-0.4  mg N/l) until the seventh
  week  after refill,  whereas  the  ammonia in the  test pools dropped  to
  less  than  0.1 mg N/l by the second  week.  Two  mechanisms appear respon-
  sible for  this rapid decline:   readsorption of ammonia by the sediment—
  probably by  a layer of  benthic  algae and bacteria—and volatilization
  of  ammonia to the atmosphere under  the slightly alkaline pH conditions
  (Bouldin e^ al_.  1974).  Thorc was not a sufficient increase in plank-
  tonic algal  biomass during  this  period to account for the rapid ammonia
  loss  (see  later  discussion  of chlorophyll data).  Ammonia in the test
  pools subsequently  fluctuated between 0.05 and 0.2 mg N/l, showing at
  first a slowly increasing trend  to a maximum of 0.2 mg/1 by the ninth
 week and then  rapidly declining and  remaining below 0.1 mg/1 until the
 end of the monitoring perioj.  Ammonia in the control pools declined
 ripidly after  the seventh week, and test and control levels were equal
 by the ninth week.   However, the decline in control pool ammonia was
 slower thereafter,  and the control pools exhibited significantly higher
 ammonia levels compared to the test pools for most  of the remainder of
 the monitoring period.

 Nitrate levels in the pools were temporally highly  variable (Figure 31),
 but nc initial spike was noted in the test pools following refill.
 Evidently some organic oxidation occurred in the drying  sediments,
 releasing ammonia from organic N forms,  but  under the drying  conditions
 used for these simulations nitrification  of  the released  ammonia did
 not occur.   This is probably an advantage since nitrate  is  more easily
 leached from sediment upon refill than is ammonia,  and conversely  am-
 monia  is more easily lost from solution by volatilization and sediment
 resorption  than is  nitrate.   The relatively  large and rapid oscillations
 In  nitrate  levels in the pool waters after refill preclude detailed
 discussion  of the data,  but  it  is clear from  Figure  31 that nitrate was
 generally higher  in  the  control  pools than in the test pools during the
 first  nine  weeks  following refill.   Nitrate levels in both sets of
 pools  showed  similar fluctuations thereafter, but the maxlmun values
 associated  with two  short-lived  peaks in nitrate were higher in the
 control  pools.

 Interpretation of the  temporal variations In inorganic nutrient forms
 is a complicated matter.  Obviously  the variations are the result not
only of sediment water interactions but of seasonal and short-term
changes in benthic and planktonic algal growths, as well as periodic
nutrient inputs to the open pools from rainfall and dry fallout.  It
can be concluded that sediment drying did not v.ause any spikes of
orthophosphate and nitrate in the refill water and that an initial
ammonia spike was short-lived with values rapidly falling below the
control pool levels.  Furthermore, inorganic nitrogen levels were
                                   71

-------
  0.5-
  0.4-
  0.2-

  0.1-
                     T
                      4
-T	1	""
 8              12
 WEEKS FOLLOWING REFILL
T"
 16
20
Figure 30.  Weekly changes in ammonia in pool simulations

-------
C9
Ul

Of

Z
    0.40-
    0.32-
    0.24-
    0.16-
    0.08-
    0.00
                                        8
                                                      —T
                                                       12
T
 16
—T
 20
                                   WEEKS  FOLLOWING REFILL
    Figure 31.  Weekly changes in nitrate  in pool simulations

-------
 generally lower  in the test pools compared to the control.  Inorganic
 phosphate was nearly the same in both sets of pools, b.it total phosphate
 was definitely lower in the test pools.  Total organic nitrogen (not
 graphed) was also lower in the test pools, but this conclusion is based
 on limited data.

 Dissolved oxygen values (Figure 32) in the test pools were, with few
 exceptions, higher than those in the control pools, and both sets were
 generally above saturation values.  By the end of the monitoring period,
 both sets had dissolved oxygen values considerably higher than satura-
 tion, implying high rates of photosynthesis.   The generally higher
 oxygen values in the test tanks may be explained by their lower turbid-
 ity values.  The high turbidities in the control pools reflect suspen-
 sion of the flocculent sediment; consequently sediment oxygen demand
 Is likely to have been a more important factor in the control pools
 than in the test pools with their consolidated sediments.

 The data shown in Figures 24 to 32 were subjected to statistical
 analysis in order determine the significance  of  apparent  differences
 between the two sets  of pools.   Analysis of variance was  performed
 (Table 19)  for 14 parameters on three combinations of the pool water
 quality data:   1) total time following refill using all pools,  2)
 total time  following  refill excluding two pools  (a and e) that  devel-
 oped  leaks  and were  inadvertently partially dried (see Section VI),  and
 3)  last two months of  monitoring period  using all pools.   The effects
 of  the partial drying  and  transfer operations on  conditions in pools a
 and e  were  felt  to have been  eliminated  by that  time.   Hence  this  treat-
 ment  probably  represents  the  best  comparison  of  refill water  quality
 in  the dried and  undried  pools.  The results  in Table  19  corroborate
 the qualitative Judgements  made  earlier  in describing-Figures 24 to  32,
 and indicate that  among the nutrient forms only  total  phosphorus and
 ammonia were significantly  different in  the two sets of pools.  Temp-
 erature was significantly different  during the last  two months  for all
 pools,  and  turbidity Just missed being statistically different  (at the
 95% confidence level) using the  same sampling period.   However, turbid-
 ity differences were highly significant  for the total  refill  time when
 the two  leaky pools were excluded.   Dissolved oxygen differences were
 significant for both of the latter two data treatments  but were not
 different considering all pools  for  the total time.  Only one parameter
 (total phosphate) was significantly  different for all pools and san-ple
 times, but seven parameters were significantly different when the two
 leaky pools were excluded, and six parameters showed significant dif-
 ferences using all pools for the last two months of monitoring.

A laboratory experiment was conducted to determine the effect of drying
extent on refill water quality.  Six small  (15 cm x 25 cm x 20 cm)
aquaria were filled with fresh Lake Apopka sediment to a depth of 6.5
cm.  One aquarium was then filled with Gourd Neck Springs water and
served as a control, while the other five were subjected to accelerated
drying conditions under 150 watt spotlights.  Aquaria remained under the
lights for varying periods in oraer to achieve a  gradient of dryness;
as each aquarium was removed from the lights,  it was covered with

-------
UJ
CD

X
o

Q
UJ


O
V)
Vi
18


16



14



12



10



 8



 6



4'


2-1
                                 TEST

                                 CONTROL
                     4          8          12         16

                         WEEKS FOLLOWING REFILL
                                                           20
Figure 32.  Weekly changes  In dissolved oxygen in pool  simulations
                                75

-------
TABLE 19.  ANALYSIS OF VARIANCE RESULTS FOR WATER QUALITY PARAMETERS
           IN POOL SIMULATIONS
Parameter
Sampling Periods
Total Time
(All Pools)
PH
Conductivity
Turbidity
Dissolved Oxygen
Temperature
Alkalinity
Ammonia
Nitrate
Orthophosphate
Total Phosphorus
Percent Water
Volatile Solids
Oxygen Saturation
Chlorophyll a_
Adenosine Triphosphate
Bacterial Counts
0.32
0.29
0.44
0.19
0.08
0.59
0.98
0.27
0.79
o.ooia
0.15
0.83
0.26
0.17
0.95
0.47
Total Time
(Excluding Two)
0.75
O.OOP
0.001a
0.001a
0.40
0.001 a
0.0013
0.48
0.76
0.30
0.001s
0.001 a
0.04
0.87
0.10
0.81
Last Two Months
(All Pools)
0.16
0.14
0.07
0.04a
o.onj
0.71
0.03a
0.10
0.37
0.02B
0.0013
0.67
0.09
0.03a
—
™
     Denotes significant difference (at or greater than 95% confidence
     level) between test and control pools.
                                   76

-------
  aluminum foil  to  prevent  further  evaporation.  The least dried sediment
  had  only slight cracks  in its  surface while the most dried had developed
  cracks down  to the bottom of the  aquarium and had separated into indi-
  vidual hard  cakes.  After all  aquaria nad been removed from the lights,
  they were  filled  with fresh Gourd Neck Springs water by carefully si-
  phoning  water  down the  sides of the containers to minimize sediment
  disruption.  Air  spargers were placed in the aquaria to keep the water
  mixed and  aerated and samples  were taken from each aquarium twice a
  week for a one-month period for turbidity and inorganic nutrient
  analyses.  Results of the monitoring are summarized in Figure 33.

  The  extent of  drying is shown  by the percent water in sediment samples
  taken prior  to refill.  The control sediment (//I) was 96.6 percent
 water, the least  dried sediment had a water content of 91.6 percent,
 and  the most dried had only 16.1 percent water.  It is obvious that
 drying had little effect  on organic content.  Even the driest sediment
 had a percent volatile solids only 1.2 percent lower than the control,
 and  the range  for all six aquaria was 67.2 to 69.0 percent.   Of course,
 under slower, natural drying conditions microbial oxidation may result
 in somewhat larger volatile solids decreases,  but the results of our
 larger simulations also  indicate miiior  losses.

 The effect of drying period on inorganic nutrient levels  in  the refill
 water was different  for  each of the three nutrient forms.   Drying en-
 hanced nitrate release from the sediments in that all dried  aquaria
 had higher nitrate levels than the control.   However, initial  nitrate
 release was inversely related to  the  extent  of  drying,  with  the least
 dry sediment  yielding nitrate spikes  greater than 2 mg N/l and concen-
 trations  then declining  with time. The more highlv dried sediments
 showed lower  initial  spikes but increasing nitrate levels with time
 following refill.  Apparently a small amount of drying enhances nitri-
 fication  but  extended  drying inhibits  further nitrifying activity.
 When  the  highly dried  sediments are reilooded they provide a better
 substrate for nitrifieis than the undri«d  muck, and consequently nitrate
 levels  eventually  increase.

 The opposite  results are seen for  ammonia in the refill water.  Slight
 to  moderate drying depressed ammonia release, but  extended drying
 caused sufficient  organic  matter decomposition to  produce elevated
 ammonia levels  in  the refill water.  However, only the most highly
 dried sediment  (which is unlikely  to be achieved in the environment)
 produced  excessive ammonia levels.  It should be noted that the control
 ammonia levels  were much lower  than values found  in  Lake Apopka.  The
 refill water  (from Gourd Neck Springs) is naturally low in ammonia.
 Under the relatively quiescent  laboratory conditions, a defined sedi-
 ment/water  interface was maintained, thus minimizing ammonia release.

 Orthophosphate  levels in all aquaria showed Increasing concentrations
with  time following refill.  Extensive drying led to elevated phosphate
 releases compared  to the control sediment, whereas the slightly to
moderately dried sediments had orthophosphate levels somewhat lower
 to slightly more than the control.   Turbidity levels in the refill
                                  77

-------
01
01
£L
 (9
100
50
0


1.0
0.0
1.5

1.0

0.5
0.



A





C


II


















X H20




1 4
NO~-N



1 2 3


II



4
™ ^


n
5 6


II


5 6
E ORTHO P04-P
•




II








||













B


E




—





i —





%


V


OL


.1


SC


(LIDS


123454
NH3-N
(x10~



.. i.


-

>
1



1 23456
F TURBIDITY


•




III III III

III




0
rr 04
65 0
cc
Hi
C-
60
1.0
-s.
Z
19
0.5
0.0
3.0

2.0
ri
-)
1.0
0.0
           123456

                              LEACHING STUDY
  Figure  33.  Effect of drying period on sediment and refill water quality
             In  laboratory  (aquarium) experiment.  Numbers on abscissa
             represent different aquaria  (dried to varying extents).  The
             three vertic.il  lines  In C-F  for each aquarium represent analy-
             sis Bade one,  two, and three weeks after refill, respectively
                                      78

-------
  waters of all aquaria (Including the control) were lower than tvpical
  values for Lake Apopka.   Slight to moderate drying decreased the
  already low control values, but extensive dr ing led  to elevated  tur-
  bidity.

  The results in Figure 33 indicate there is an optimum level  of drying
  in terms of etfects on water quality.   Too little drying stimulates
  nitrate release and more seriously will not  consolidate the  sediments
  ana stabilize the  interface.   On jhe other hand  extensive drying  tends
  to release  more ammonia  and phosphate,  at  least  in short  term tests.
  than docs undricd  sediment.
 BIOLOGICAL EFFECTS

 Biological parameters measured in water overlying the sediments in the
 test and control columns and tanks included primary production and
 chlorophyll a.  Production was measured by the C-U uptake technique
 (Parsons and Strickland.1968).  As mentioned earlier, the primary pro-
 ducer component of the control columns and tanks shifted from phyto-
 plankton to periphyton.  Shortly after refill, the same phenomenon
 occurred in the test columns and tanks.  This shift frow plankton to
 periphyton in the controls, which was clearly unrepresentative of
 Lake Apopka, necessitated abandonment of the tanks and columns and the
 construction of the continually mixed pool simulations.   Figure 34 con-
 tains typical control tank data showing the rapid decline in C-14 pro-
 ductivity and ci.lorophyll a levels,  both indicating crashes In the
 pnytoplankton populations.   Sixty centimeters of muck sediment was
 placed in the control iank on July 29, 1972.   A dense mat of Oedogonium
 sp.  dominated the tank by late  November,  1972 and remained  until  the	
 end  of the experiment.   The test  tank had  the same amount of sediment
 added to ic on July 2S  and  was  drawn  down  on  August  11 for  approximately
 2  months.   Refill occurred  on October 19,  1972.   After the  test tank
 was  refilled,  a growth  of Chara rapidly covered  the  dried sedimentH
 and  apparently outcompeted  the  phytoplankion  for nutrients,  as did the
 Ocdogontun ln  the control tank.   Chlorophyll  a  levels  an-J C-l« primary
 production remained well below  values measured during  the same months
 in Lake  Apopka.   In December, for instance, the  chlorophyll  level  In the
 control  tank was  about  10 mg/m3 and only slightly over 20 mg/m3 in the
 test  tank.   In  the  lake, the  chlorophyll level was about  70 mg/m3.
 The  primary production value  for  the  lake at that time was about 350 mg
 C/mVhr, as compared  to only  20 mg C/m3/hr for the test and control
 tanks.  With our controls behaving so  differently from Lake Apopka,
 it was impossible to determine whether or not the low chlorophyll and
 productivity values in tne test lar.k  refill water were a result of draw-
 down and sediment drying or simply a simulation anomaly.   This was one
 of the major reasons for changing to the mixed pool simulations.

 In an effort to determine whether or not dried sediments mixed In water
 stimulated the growth of algae,  a series of algal bloassays  were per-
 formed.  In  discussions of drawdownasa lake restoration technique,  it  is
often hypothesized that sun drying will, through oxidation,  convert
                                   79

-------
  100-1
   80-
E
a
   60-
u
i
-|  40-
a.
O
c
3
   20-
































t
1
t
1
L
to a CONTROL D
1 O REFILLS
	 P.P. SOLID LINE
	 CHLOROPHYLL A








^^
t'ta/fc
1 <3
I ' \
1 J \
1 9 b

! • \
1 A -/I A
' 1 * / \ sA \
«L-. I ' * $^ NVA ^^^^^x)
\ I/ 1 '*WI\' \ \
\ ' l f {4 ^9 <*** b
\J»-^_A^ ^><2!2^Ix>*®""?!^z^%'®






500



400
T)-
a
•I
a
r300 5
O
O
c
q
<
H
-200 <
3*
J
3
- ^
-100 **




-0
          AUG.
SEPT.     OCT.
 NOV.      DEC.     JAN.     FEB.
TIME
   Figure 3A.  Primary productivity  and chlorophyll a in control and
               test tanks
                                     80

-------
organic nitrogen and phosphorus compounds in the sediment to inorganic
forms, which upon refill might be released and stimulate algal growth.
In our pool experiments, we monitored algal growth following refill.
Those results (described later) can be considered more representative
of what might happen in Lake Apopka itself.  The algal bioassay data
reported here represent "worst case" results.  Because the dried sedi-
ments are almost rock hard, very little mixing with the overlying water
occurred in the pool simulations or would occur in the lake.  In our
bioassay, however, we ground the hardened sediment to a sand-like con-
sistency and then used a magnetic stlrrer to mix the sediment with
deionized water for 24 hours.  Obviously, no such mixing forces are
present in Lake Apopka.  Prior to grinding, the sediment was exposed
to sunlight In a drained plexlglas cylinder for one month.

Two mixtures of dried sediment were used:  0.5 g/100 ml deionized
water and 1.0 g/100 ml.  Fresh wet sediment was also mixed with
deionized water to yield 0.5 and 1.0 g/100 ml deionized water on a
dry weight basis.  A fresh sediment water content of 95 percent was
used in making those calculations.  The mixtures were ccntrlfuged to
remove suspended solids and the supernatants were Mlllipore (0.45 um)
filtered prior to testing.  Three 100 ml filtered replicates of each
dried sediment solution and duplicates of the wet sediment solution
were inoculated with Selenastrum capricornutum to produce an initial
cell concentration in the test flasks of 1000 cells/ml.  Media and
test procedures followed those recommended by the Environmental Pro-
tection Agency (1971a).  Cell countb were made daily until the increase
in cell number was less than 5 percent per day.  The results are plotted
in Figures 35 and 36.  At 0.5 g/100 mi of deionized water, dry sediment
supports less growth than does wet sediment.  At 1.0 g/100 ml. however,
slightly higher populations of Selcnastrum were reached with the dried
sediment mixture.  The population of Selcnnstrum grown In algal assay
procedure medium exceeded all test replications, reaching almost 5 x
106 cells/ml after 12 days.  Possibly our initial sediment inoculations
were too small.  This experiment did indicate, however, that dried
sediment, even when thoroughly ground up and vigorously stirred with
water, does not support a significantly higher algal blomass than does
a comparable amount of wet sediment.  Rather than continue this rather
unrealistic line of experimentation, we set up the large mixed pool
simulations and followed algal populations in them.

Phytoplankton sampling was begun in early April, 1974, some 5 weeks
following refill.  All pools were sampled on an approximate weekly
basis until June 7, 1976.  Using the methods recommended by the
Environmental Protection Agency (1973), Shannon-Weaver generic diver-
sity indices were calculated and are shown in Figure 37.  Although an
analysis of variance shows that there is a significant differece (P •
0.07) between test and control tank diversity (test being higher), the
interpretation of the phytoplankton data is complicated by several
factors.  The refill water added to the test pools was originally
free of algae and therefore required some time to develop a flora,
which has as Its origin resistant algal forms in the dried sediment
and airborne algae.  Although Gourd Neck Springs water is also algal
                                    81

-------
  10*-
u
U
   10*-
             o


        Figure '
T
                                     O.Sgm wo* ••dlm«nt/1OO ml
                             , _	1.09111 wot a«dimont/1OO ml
                                   T
                                   4
                      T
                       S
"T
 e
T
 7
T
 8
123

                   TIME IN DAV9

 Selenastrum growth In wet sedinent suspensions

                     82
T
 0
                                                    10    11

-------
   io'H
w

.j
HI
U

  104H
     •

                                    O.8 gm dry sadlmcnt/lOO ml

                                 — 1.0 gm dry Mdlm«nt/iOO ml
          "T
           0
T
 2
3
T
 4

T
 e
"T
 7
T
8
                                                                    10
                                                                          II
                                    TlfiflB IN DAYS

         Figure 36.  Sclenaatroa growth In dry sediment suspenolons
                                      83

-------
free, It mixes limedlately with the algal population Indigenous to
Lake Apopka.  Furthermore, the test pools received no nutrient  input
other than that provided by rainfall and sediment-water exehanfie.   in
the lake, the water overlying dried sediment will exchange ireely  with
water overlying wet unconsolidated sediment.  Furthermore, nutrient
sources other than the sediment and rainfall will exist following  draw-
down.  The most important nutrient source will probably be runoff.
Figure 37 shows that 10 weeks after refill, the diversity indices  for
test and control tanks are quite similar.  It was mentioned earlier
that turbidity in the test pools was lower than in the control  pools.
Hopefully, increased water clarity would occur in the lake following
drawdown and favor the establishment of submerged macrophytes which
would out compete the phytoplankton for nutrients or exert a negative
influence on the floating algae.  Again, however, the presence of
unconsolidated sediments will add suspended solids and nutrients to  the
overlying water as the lake refills.  At this time, no quantitative
estimate of the  Influence of consolidated sediments on refill water
quality  can be made.  Table 20 provides further detail on the phyto-
plankton results.  Averages of the  test versus control pool means
presented in Table 20 indicate the  test pools had lower total numbers
of  algae, a lower percentage of blue-greens, and  the same number of
blooms  (defined  as 500/ial or greater  for one species).  Chlorophyll
a was measured weekly for 36 weeks  following refill  (Figure 38).  The
Analysis of variance showed  that  the  test  pools had significantly
lower chlorophyll values than  the  control  pools when only the  last
two months  were  considered.  Thirty-six weeks after refill, however,
chlorophyll levels were  similar  in the  control and  test  tanks.  Con-
sidering all  the data relating directly or indirectly  to  phytoplankton,
it  appears  unlikely  that a lake-wide  pulse of phytoplankton growth
will occur  following refill.

In addition to phytoplankton and chlorophyll  a,  the pool simulations
were monitored for sediment-water interface adenosine  triphosphate
 levels  and  total bacterial counts.  These  parameters were chosen  to
 determine whether microbial metabolic activity was higher in drying
 sediments.   The data for ATP and total plate counts fluctuated greatly
vith time for both test and control sediments.   No trends were apparent.
 These results strengthen the conclusion that significant breakdown
 of organic matter in drying sediment does not occur

 Benthic invertebrate and microphyte populations were also determined
 and are presented in Chapter VI.
                                     84

-------
8
CO



z  3



>-



co
                0 .
                      TEST a
                                          6           7           8



                                         WEEKS FOLLOWING REFILL
                                                                           10

-------
                   TABLE  20.   POST DRAWDOWN POOL SIMULATION PHYTOPLA1IKTON  RESULTS
Date


4/12/74
4/20/74
4/24/74
5/02/74
5/10/74
5/31/74
6/07/74
MEAN
Test Pools

Count*
1,953
10,206
7,268
23.942
3,864
4,990
12,060
9,183
0
Bb
0
8
4
10
2
3
2
4
Y
BG
20
27
16
55
53
39
3
30
Count
4,792
8,306
3,410
10,465
1,967
1,081
510
4,362
B
2
2
2
4
1
0
0
2
BG
66
5
3
9
1
6
18
15
B
Count B
6,379 2


3,540 3
5,865 3
4,068 2
5,580 2
3,633 2

BG
9


45
17
32
44
29
Control Pools

Count
2,742
3,680
6,442
18,170
29,120
11,146
47,992
17,041
A
B
1
2
2
6
3
2
16
5
c
BG
60
36
19
51
18
55
27
38
Count

19,090
40,883
8,970
6,323
1,326
3,060
13.275
B

1
1
2
3
0
1
1
BG

3
1
33
74
37
55
34

a     Count in organisms per ml



b     Number of blooms,  defined as 500 or more cells' of one  species  per  ml



c     Percent of total count belonging to the phylum Cyanophyta

-------
30-i
25-
20-
10-
 5 J
        23    24    25    26    27    23    29    30    31    32    33

                                  WEEKS FOLLOWING REFILL
34    35    36
   Figure 38.   Chlorophyll concentrations in test and control pool simula-
               tions following refill

-------
                             SECTION VIII

                              REFERENCES

American Public Health Association.  1971.  Standard methods for the
examination of water and wastewater.  A.P.H.A., Washington, D. C.
874 p.

Bouldin, D. R., R. L., Johnson, C. Burda, and C.-W. Kao.  1974.   Losses
of inorganic nitrogen from aquatic systems.  J. Environ. Quality, 3:
107-114.

Brezonik, P. L., V. H. Morgan, E. E. Shannon and II. D. Putnam.  1969.
Eutrophication factors in north central Florida lakes.  W.R.R.C. Pubi.
5, University of Florida, Gainesville. 101 p.

Brezonik, P. L. and E. E. Shannon.  1971.  Trophic state of lakes in
north central Florida.  W.R.R.C. Publ. 13, University of Florida,
Gainesville. 102 p.

Burgess, J. E.  1964.  Summary report of Lake Apopka.  Florida State
Board of Health. 23 p. (mimeo).

Chesnut, T. L. and E. H. Barman, Jr.  1974.  Aquatic vascular plants
of Lake Apopka, Florida.  Florida Scientist 37_(l):60-64.

Department of the Army.  1967.  Report of subsurface investigations,
Lake Apopka.  Jacksonville District Corps of Engineers  (mimeo).

Environmental Protection Agency.  1971*. Algal assay procedure bottle
test.  E.P.A., Corvallis, Oregon. 82 p.

Environmental Protection Agency.  1971b. Methods for chemical analysis
of water and wastes.  E.P.A., Cincinnati, Ohio. 312 p.

Environmental Protection Agency.  1973.  Biological fi^ld and labora-
tory methods for measuring the quality of surface waters and effluents.
E.P.A.. Cincinnati, Ohio.  Approx.  120 p.

Florida Department of Pollution Control.  1972.  Lake Apopka restora-
tion project.  F.D.P.C., Tallahassee, Florida.  Approx. 140 p. * Appen-
dices.

Florida Game and Freshwater  Fish  Commisson  (W. Wegener, V. Williams,
and D. Kolcomb).   1972-1973  Annual  Progress Report.  Water level
manipulation, Lake Tohopekaliga drawdown.  F.G.F.W.F.C., Tallahassee.
Approx. 100 p.

Florida State Board of Health.   1964.  Summary report of Lake Apopka.
Florida State Board of Health, Jacksonville.   19 + p.
                                    88

-------
  Florida Stace Board of Health.  Undated.  Biological, chemical, and
  physical study of Lake Apopka. 1962-1964.  Fla. State Brd.  of Health,
  Jacksonville.  56 + p.

  "•orbes, R. B.  1968.  Water quality studies, Zellwood Drainage and
  Wcter Control District.  Central Florida Experiment  Station Report.
  11 p. (mlmeo.)

  Heaney, J. P., A.  I. Perez and J.  L.  Fox.  1971.   Nutrient  budget
  within the organic soils area north of Lake Apopka.   Report  to the
  the East Central  Florida Regional  Planning Council.   University of
  Florida, Gainesville.   18 p.

  Henriksen, A.   1965.  An automatic method for  detenning nitrate and
  nitrite In fresh and saline waters.   The Analyst,  90(1067): 83-88.

  Huffstutler,  K. K.,  J.  E.  Burgess  and B.  B.  Glenn.   1965.  Biological
  physical and  chemical  study of  Lake Apopka,  1962-1964.  F-arlda State
  Board  of Health, Jacksonville.   78 p.  (mimeo.)  Also sec Florida State
  Board  of Health, undated  report.

  Lane,  E.   1966.  Lake Apopka 's  lingering death.  Orlando Sentinel
  article, April 3.

  Melack,  J. M. and P. Kilhara.  1974.  Photosynthetic rates of phyto-
                          alkaline- saline lakes-  Limnol. Oceanogr.
 Mortimer, C. H.  1941.  The exchange of dissolved substances between
 mud and water in lakes.  Ecology 29; 280-329.

 Murphy, J. arl J. P. Riley.  1962.  A modified single solution mehhod
 for the determination of phosphate in natural waters.  Analyt. Chim.
 Acta ^7 ! 31~36»

 Odum, E.  P.   1971.   Fundamentals of Ecology.  •„.  B.  Saunders Co.
 Philadelphia.   574  p.

 Odum, H.  T.  and C.  M. Hoskin.   1958.   Comparative studies on the  meta-
 bolism of marine waters.   Publ.  Inst.  Mar.  Scl.,  Univ.  of Texas,  5:16-
 MU»                                                               "~

 Perez,  A.  I. 1972.   A Water Quality Model  for a Conjunctive  Surface
 Troundwater  System.   Doctoral  Dissertation,  University  of Florida
 Gainesville.   489 p.

 Peterson,  G. W.  and  R.  B.  Corey.   1966.  A modified Chang and  Jackson
 procedure  for  routine fractionation of inorganic  soil phosphates.  Soil
 Sci.  Amer. Proc.  30:63-65.

 Schneider, R.  F.  and  J. A.  Little.  1968.  Characterization of bottom
 sediments and  selected nitrogen and phosphorus sources  in Lake Apopka,
 Florida.  U.S. Dept.  of Interior Report, F.W.P.C.A., S.E. Water lib.,
Athens, Ga. 42 +  p.  (mimeo.)
                                   89

-------
Shannon, E. E.  1970.  Eutrophtcatlon-Trophic State Relationships in
North and Central Florida Latas.  Doctoral Disseration, University of
Florida, Gainesville.  258 p.

Sheffield, C. W. 1967.  Report on loose unconsolidated sediment samples
from Lake Apopka.  Orange County Pollution Control Depart, (mimeo.)

Sheffield, C. U.  1970.  Effects of Agricultural drainage waters.  A
paper presented at the Irrigation and Drainage Specialty Conference,
Miami Beach.  23 p. (mineo.)

Sheffield, C. W.  1974.  Personal communication.

Sokal, R. R. and F. J. Rohlf.  1969.  Biometry - The Principles and
Practices of Statistics in Biological Research.  W. H. Freeman and Co.,
San Francisco.  776 p.

State of Florida, Department of Air and Water Pollution Control.  1970.
The Lake Apopka - Oklawaha River Basin restoration project.  State of
Florida, Tallahassee.  10 p. (mimeo.)

Strickland, J. D. H. and T. R. Parsons.  1968.  A practical handbook
of seawater analysis.  Fisheries Res. Board of Canada, Bulletin No.
167.  311 p.

Tecknicon Corporation.  1969.  Ammonia in water and wastewater.  Tech-
nicoa Industr. Heth. Bull. 19-69w, Tarrytown, New York.

Yorton, R. 1971.  The Effects of Amino Acids and Organic Color on
Automated Nutrient Analyses of Natural Waters.  M.S. Thesis, Universi-
ty of Florida, Gainesville.
                                   90

-------
                               SECTION IX

                                APPENDIX

                                 METHODS
 Whenever possible, the methods used followed -hose described in either
 Standard Methods for the Examination of Water and Wastewater (A.P.H.A.
 1971), Methods for Chemical Analysis of Water and Wastes (E.P.A., 1971b)
 or Biological Field and Laboratory Methods for Measuring the Quality
 of Surface Waters and Effluents (E.P.A., 1973).  The following is a
 summary of methods used for water, sediment and statistical analyses:

WATER ANALYSES

 1.   Algal Assay - Performed according to the U.S.  Environmental
     Protection Agency's (1971a)basic bottle algal  assay  procedure.
     Details are provided in the text.

 2.   Alkalinity - Run  colorimetrically  using the methyl orange  brora-
     cresol  green mixed  indicator  (A.P.H.A.,  1971)

 3.   Chemical  Oxygen Demand  - Determined  using  the  dichromate-sulfuric
     acid  technique described in Standard Methods  (A.P.H.A.,  1971).

*"   Chlorophyll  a - Samples  were Millipore  filtered  (0.45 u), extracted
     in 90 percent acetone, and analyzed  spectrophotometrically as
     described by Strickland  and Parsons  (1968).

5.   Conductivity - Determined in the laboratory using a Beckman Model
     RC 16BZ conductivity bridge.

6.  Dissolved Oxygen - For most field and simulation measurements, a
    YSI oxygen electrode was used.  For the diurnal productivity mea-
    surements, the Winkler-Azide modification technique (A.P.H.A.,
    1971) was used.

7.  Nutrients -Inorganic nitrogen and phosphorus forms were determined
    witn a Technicon Auto Analyzer on samples preserved with one ml
    saturated mecuric chloride per liter of sample.  Ammonia was
    determined usin3 modifications (Yorton,  1971) cf the  phenol-hypo-
    chlorite method (Technicon Corporation,  1969).   Nitrate  was deter-
    mined by the automated hydrazine reduction technique  (Henriken,
    1965), and nitrite  by adaption  of the  Standard Methods  (A.P.H.A.)
    diazotization procedure  to the Auto Analyzer.   Orthophosphate was
    measured by the Murphy and Riley (1972)  single  reagent method.
    Total  organic nitrogen was determined by the macro-Kjeldahl method
    according to Standard Methods  (A.P.H.A.,  1971).   Recovered  ammonia
    was measured by titration or by the Auto Analyzer.  Partlculate
    organic  nitrogen was  measured by micro or macro-Kjeldahl  on solids
    scraped  from an MgC03 mat placed on a 0.45  u Millipore filter.
    Total phosphate analyses  were performed  by  persulfate-sulfurlc acid
    autoclaving  for one hour  at 15  psi  followed by  manual determination
    of orthophosphate by  the  Murphy and Riley  (1962) method.


                                  91

-------
 8.  pH - Field measurements were made with Orion Model 401 Meter.   A
     Corning Model 12 was used in the laboratory.

 9.  Productivity - Field and simulation determinations were made
     utilizing the diurnal free water oxygen method described by Odum
     and lloskin (1958) and the radiolsotopic carbon bottle technique
     as presented by Strickland and Parsuns (1968).

10.  Turbidity - Determined in the laboratory with a llach Model 2100A
     Turbidimeter.

SEDIMENT ANALYSES

1.   Adenoslr.e Triphosphate - A spatula was used to sample sediments
     from test and control simulations.  The sample consisted of about
     10 cm2 of surface sediment (0.5 cm deep).  A sediment-filtered
     lake water suspension was Millipore filtered and the ATP extracted
     in boiling TRIS buffer.  A Dupt.nt luminescence biometer was used
     to detect the amount of light emitted by the sample ATP in the
     presence of luciferin and luciferase (the firefly reaction).

2.   Bacterial Counts - Samples were taken as described above for ATP.
     Samples were suspended in Millipore filtered (0.45 u) lake water
     and serial d'lutions plated on tryptone glucose extract agar.
     Incubation w.-s at 25°C for 72 hours.  Colony counts were related
     to sediment dry weight.

3.   Consolidation - The decrease in sediment height, as measured by a
     meter stick, was expressed as the percentage of the original sedi-
     ment height.

4.   Nutrients - Total organic nitrogen and free ammonia determinations
     were maH° on fresh-sediments diluted with demineralized water into
     a thin slurry.  Ammonia was distilled from the slurry buffered at
     pH 8.3 with sodium bicarbonate and the recovered ammonia measured
     by titration.  Total organic nitrogen was determined similarly by
     micro-Kjeldahl digestion.  Results were converted to a dry weight
     basis b. drying and weighing a known volume of the sediment slurry.
     Total phosphate was run on fresh sediment slurries using persulfate
     oxidation and the Murphy and Riley (1962) manual method for ortho-
     phosphate.

5.   Weights - Wet weight, dry weight and ashed weight determinations
     were made according to Standard Methods  (A.P.H.A., 1971).  From
     those data, appropriate calculations were made for percent water
     and volatile solids content.

STATISTICAL ANALYSIS

     An analysis of variance  (ANOVA) was used to analyze  the experimental
     pool results.  ANOVA  (Sokal and Rohlf, 1969)  is a  test of whether
     two or more  sample means could have been obtained  from populations
     with the same parametric mean with respect  to a given variable.  The
     variables in this experiment were  the physical, chemical  and biolo-

                                   92

-------
gical parameters.  The two populations are a population of draw-
down lakes versus a population of control lakes, and the two
sample means upon which we based our popularion conclusions were
the test and the control pools.  ANOVA determines whether the
sample means from the test and control pools differ from each
other to such an extent that we must assume (with 95% confidence)
that they were sampled from different populations.  Two ANOVA
designs were considered:  a single classification ANOVA and a
block design ANOVA.  The single classification ANCVA is used :o
determine if there is a significant mean variable difference
between the control and test pools during the time-week period.
The block design is similar except it removes the weekly (block)
variation, making it a more exact test of treatment (test versus
control) effects.  Weekly variation in the pools could be caused
by climatic factors and it is assumed that these factors affect
the test and control pools in a similar fashion.

References cited in this appendix are listed in Section VIII.
                             93

-------
                                   TECHNICAL REPORT DATA
                            tfleate read /aunclions on lite ret tne before completing/
  REPORT MO
   EPA-600/3-77-005
                                                           3 RECIPIENT'S ACCESSION-NO
4. TITLE A«0 SUBTITLE

   Lake Drawdown as  a  Method of Improving Water Quality
                                  6. REPORT DATE
                                      January 1977
                                                           B. "ERFORMINC ORGANIZATION CODE
7. AUTHORlSl
   Jackson L. Fox. Patrick L.  Brezorik and Michael A.
   Keirn
                                                           B. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS
   University of Florida
   Gainesville, Florida    32611
                                                           10. PROGRAM ELEMENT NO.
                                    1BA031	
                                  Tr"CONTRACT75RAN~r f.6~"

                                    EPA  R-800305
 12. SPONSORING AGENCY H»M& AND AOORCSS
   U.S. Environmental  Protection Agency
   Corvallis Environmental  Research Laboratory
   200 S.W. 35th Street
   Corvallis, Oregca
                                   13. TYPE OF REPORT AND PERIOD COVERED
                                   Final
                                   14. SPONSORING AGENCY CODE
                                     EPA/600/02
19. SUPPLEMENTARY NOTES
16. ABSTRACT
   Investigations vere wade  to  determine the feasibility of  radical drawdown as a
   restoration techaime  for Lake Apopka, Florida, a 12,545  hectare lake in central
   Florida.  Field steadies showed the lake to be hypereutrophic with continual algal
   blooos, mats of fixating  water hyacinths, and a flocculent  organic nuck bottom
   lich in interstitial water nutrients.  Sediments were dredged from the lake bottom
   and placed in aquaria, columns, tanks and pools.  Following dewatering and varying
   drying periods, the containers of sediment were refilled.   A large number of
   physical, cheaical and biological parameters were monitored before, during and
   following sediaeat drying.   Results indicate that drawdown  improves subsequent
   refill water quality.  In muck sediments, drying causes significant water loss
   and shrinkage.  Loss of organic material is minimal.  During and following refill,
   sediment is colooized by  two macroscopic aquatic plants,  Typha (cattail) and the
   alga, Chara.  Dryiag results in only minor chemical changes in muck sediment.
   Refill water in rbs pool  test simulations has the same or lower nutrient content,
   lower turbidity, higher dissolved oxygen, lower temperature, fewer algae and a
   more diverse benchie Invertebrate population.  Based on these laboratory scale
   Investigations, drawdown  appears to be an effective restoration technique for
   Lake Apopka.
17.
                               KEY WORDS AND DOCUMENT ANALYSIS
                        'ORS
                                              O.IOENTIFIERS.OPEN ENOEO TERMS
                                                                        c  COSATI F-:ld/Croup
   Sediments*
   renovating
   lakes
   limnology
   phosphorus

   *HaJor descriptors
nitrogen
aquatic weeds*
aquatic biology
benthos
bioassay
  08M
  07B
  06F
18. DISTRIBUTION STATEMENT

   Release unlimited
                      19. SECURITY CLASS ITha Report)
                         unclassified
21. NO OF PAGES

     104
                                              20 SECURITY CLASS (rins paft)
                                                 unclassified
                                                                        22 PRICE
EPA Form JMO-1 (9-7J)
                           • us covf MMCNT POINTING OFFICE nn-nt.iiti 31 QCSION 10

-------