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
------- |