Citation
Phosphorus dynamics in an acidic, soft-water Florida lake

Material Information

Title:
Phosphorus dynamics in an acidic, soft-water Florida lake
Creator:
Ogburn, Reuben Walter
Copyright Date:
1984
Language:
English

Subjects

Subjects / Keywords:
Adsorption ( jstor )
Bodies of water ( jstor )
Incubation ( jstor )
Lakes ( jstor )
Macrophytes ( jstor )
Nutrients ( jstor )
pH ( jstor )
Phosphorus ( jstor )
Phytoplankton ( jstor )
Sediments ( jstor )
City of Gainesville ( local )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
ACN9006 ( ltuf )
00473797 ( ALEPH )
11665687 ( OCLC )

Downloads

This item has the following downloads:


Full Text










PHOSPHORUS DYNAMICS IN AN ACIDIC, SOFT-WATER FLORIDA LAKE





by

REUBEN WALTER OGBURN. III


A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY
OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


1984


i

















ACKNOWLEDGMENTS


I would like to thank the many individuals who have contributed to

this effort through their technical advice and assistance as well as

those who provided moral and financial support. My research was funded

by a grant from the U.S. Environmental Protection Agency-NCSU Acid Pre-

cipitation Program to Drs. Patrick Brezonik and Tom Crisman, and its

second year of funding was administered by Dr. Bob Volk in the Soil

Science Department. Dr. Joseph Delfino and Dr. Brezonik have been my

principal advisors; their encouragement and direction have been greatly

appreciated.

Work at McCloud Lake was a cooperative effort between the chem-

istry and biology groups of the Environmental Engineering Sciences

Department. Drs. Jeff Foran and Tom Crisman coordinated the biological

data collection in the lake during 1980-1981 and in the littoral meso-

cosms. Robert Garren and Chan Clarkson assisted in surveying the

aquatic macrophytes in McCloud Lake. Larry Baker contributed to the

fieldwork and chemical analysis during the same time period. Ray

Bienert provided the biological data and interpretation from the lake

and the mid-lake enclosures, as well as assistance and encouragement in

the field. Dr. Michael Binford, a postdoctoral associate at the Flor-

ida State Museum, contributed data, field assistance, and valuable dis-

cussions related to sediment processes in McCloud Lake. Eric Edgerton










collected the hydrological data from McCloud Lake. These contributions

were essential to the completion of my research.

My parents have encouraged and supported me throughout my graduate

studies, and their early guidance and direction were instrumental in my

decision to pursue a career in science. My wife Marlyn and our child-

ren deserve equal credit for this effort. Marlyn supported my decision

to leave my job, move to Gainesville, and return to graduate school

with two small children. She made tremendous sacrifices of time and

energy to be a working mother and housewife at a time when our contemp-

oraries were pursing traditional American family lifestyles. Our sons,

Walt and Doug, have missed a great deal in the way of father-son rela-

tionships. Without the patience, understanding, and endurance of my

family, this effort would not have been possible.
















TABLE OF CONTENTS



ACKNOWLEDGMENTS .... ................................................ i

ABSTRACT ............................................................ vi

CHAPTER 1- INTRODUCTION ...............................................1

Background.... .................................................. 1
Objectives ....................................... ........ ...... 4
Site Description................................................5

CHAPTER 2--LITERATURE REVIEW................................ ....... 8

Sediment-Water Exchanges........................................8
Decomposition. ................................................ 13
Planktonic Phosphorus Cycling...................................15
Lake Phosphorus Models.......................................... 19
Effects of pH on Phosphorus Cycling ............................ 21

CHAPTER 3-PHOSPHORUS DYNAMICS IN MCCLOUD LAKE....................... 24

Materials and Methods.......................................... 24
Routine Sampling ........................................... 24
Phosphorus Budget ........................................ 25
Water budget .........................................25
Atmospheric phosphorus loading .......................27
Sedimentation........................................... 27
McCloud Lake Phosphorus Compartments ......................30
Macrophyte survey.....................................30
Sediment phosphorus.................................... 31
Phosphorus uptake................................... 31
Results and Discussion............................................ 32
Limnology and Historical Nutrient Trends ..................32
McCloud Lake Hydrology.................................... 40
McCloud Lake Phosphorus Budget.............................42
Precipitation. ....................................... 42
Mass balance..........................................46
Sedimentation................................................ 46
Phosphorus Compartments...................................... 50
Macrophytes .........................................50
Sediments............................................. .54
Phosphorus Uptake. ........................................ 54
Summary....................................................... 54

CHAPTER 4-EFFECT OF PH ON PLANKTONIC PHOSPHORUS DYNAMICS............58









Materials and Methods.......................................... 59
Mesocosm (Limno-Enclosure) Experiments ....................59
Littoral mesocosms.................................... 59
Mid-lake mesocosms.................................... 59
Laboratory Microcosms.....................................64
Results and Discussion............................................ 65
Littoral Mesocosms........................................... 65
Mid-Lake Mesocosms........................................... 70
Nutrient trends ......................................70
Biological trends................................... 75
Radiophosphorus experiments............................76
Community Metabolism......................................... 84
Laboratory Microcosms.....................................89
Acid phosphatase activity ............................91

CHAPTER 5--EFFECT OF PH ON PHOSPHORUS RELEASE DURING DECOMPOSITION..93

Experimental Methods...........................................93
Preliminary Experiments...................................... 94
Final Experimental Design .................................. 97
Results and Discussion .........................................98
Preliminary Experiments.................................. .. 98
Final Experiment............................................ 105

CHAPTER 6--SEDIMENT-WATER INTERACTIONS.............................113

Introduction................................................. 113
Background.................................................... 113
Methods. ..................................................... 115
Sediment Characterization.................................115
Batch Adsorption/Desorption Experiments ..................116
Undisturbed Core Experiments.............................117
Results and Discussion ........................................119
McCloud Sediment Characteristics.........................119
Batch Adsorption/Desorption Experiments ..................122
Sediment Core Experiments ............................... 134

CHAPTER 7- SUMMARY AND CONCLUSIONS .................................141

LITERATURE CITED. .................................................. 144

BIOGRAPHICAL SKETCH.................................................... 152
















Abstract of Dissertation Presented to the Graduate Council of
the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



PHOSPHORUS DYNAMICS IN AN ACIDIC, SOFT-WATER FLORIDA LAKE

by

REUBEN WALTER OGBURN, III

April 1984


Chairman: Joseph Delfino
Cochairman: Patrick L. Brezonik
Major Department: Environmental Engineering Sciences


Laboratory and in situ experiments as well as historical data were

used to characterize phosphorus dynamics in acidic, soft-water McCloud

Lake, Florida, and to evaluate the effect of acidification on phos-

phorus cycling processes. McCloud presently exhibits nutrient and

chlorophyll-a concentrations typical of oligotrophic Florida lakes. A

15-year pH decline (4.85 to 4.55) has not been accompanied by signifi-

cant changes in TP, chlorophyll-a, or nitrogen to phosphorus ratios,

which indicate phosphorus-limited primary production.

Total phosphorus shows maxima during late spring and summer, and

variations appeared to be related to rainfall patterns and lake levels

during 1980-1982. Atmospheric phosphorus deposition is near the

loading rate necessary to maintain mesotrophic conditions, which

suggests that low pH may contribute to the low TP in McCloud Lake.

Rooted submergent macrophytes represent an in-lake storage of









phosphorus that is approximately 2.5 times the average water column

phosphorus storage, although the macrophytes do not appear to compete

with phytoplankton for SRP.

In situ littoral and open-water mesocosm data indicated that acid-

ification (from 4.6 to 3.7) does lead to reduced water column TP

levels, although the trends were not consistent in the open-water

enclosures, which were not connected to the sediments. No relation was

seen between pH and rates of phosphorus uptake by planktonic communi-

ties, and pH did not affect the activity of extracellular acid phos-

phatase enzymes in laboratory microcosms.

The amount of SRP released from decomposing submersed macrophytes

was independent of pH (over the range 3.7 to 5.5) after 227 days of

aerobic dark incubation, although initial rates of release were some-

what faster at the lowest pH. These experiments showed that acidifica-

tion does not inhibit phosphorus release during decomposition of

aquatic plants.

Effects of pH on surface charge characteristics and SRP specia-

tion cause sediment adsorption of SRP to vary with pH. Maximum SRP

adsorption occurred near pH 4.7, although there was little variation

between pH 5.0 and 3.5. However, significant decrease in SRP adsorp-

tion at pH > 5 indicates that this mechanism may contribute to the

observed trend of low TP in acidic lakes. This effect would be great-

est over the pH range 7.0 to 5.0, and further acidification of lakes

near the pH of McCloud Lake would have little effect on SRP adsorption.
















CHAPTER 1
INTRODUCTION



Background


Acidic precipitation is considered to have pH < 5.6, which is the

pH of pure water in equilibrium with atmospheric CO2 (Likens et al.

1979). Sulfur and nitrogen oxides from anthropogenic emissions (and

from natural sources to a smaller extent) react with water vapor in the

atmosphere to form sulfuric acid and nitric acid. The return of these

acids to the earth with rainfall is known as acid precipitation, or

acid rain. However, dryfall of particulates and gaseous deposition can

also contribute significant amounts of acid to the earth's surface.

Therefore, acid precipitation generally refers to rainfall acidity,

while acid deposition includes wetfall, gaseous, and dryfall acidity.

Acid precipitation was described in England as early as 1852 and

was linked to changes in water chemistry by Gorham in the 1950's,

although it was not recognized as a widespread and serious threat to

aquatic and terrestrial ecosystems until the 1960's and 1970's (Cowling

1982). The chemical and biological changes associated with acidifica-

tion of Scandinavian lakes and streams generated intense political and

scientific interest in determining the sources, extent and effects of

atmospheric acidity. In North America, acid precipitation has been

documented in the northeastern United States and southeastern Canada,

in the southeastern U.S. (including Florida), and in the Rocky






2


Mountains. Effects of acid deposition on aquatic and terrestrial

ecosystems are difficult to demonstrate conclusively, and the problem

of differentiating between long-term trends in natural processes and

short-term changes caused by relatively recent increases in atmospheric

acidity remains a controversial issue.

Acid deposition has been implicated in accelerated erosion of

buildings, human health problems including cancer, forest decline,

decreased agricultural yields, and acidification of poorly buffered

surface waters. Perhaps the most dramatic effect of aquatic acidifica-

tion has been the elimination of trout populations from some temperate

lakes and streams. Other aquatic effects have been inferred from sur-

veys of lakes over a range of pH values.

Regional studies have shown similarities in the chemistry and

biology of acidic lakes from different geographic areas. Phytoplankton

and zooplankton assemblages tend to become more simplified with

decreasing lake pH, and similar groups of species are found in acidic

lakes of Scandinavia and temperate North America (Sprules 1975; NRCC

1981; Confer et al. 1983). Grahn et al. (1974) found that acidic lakes

in Scandinavia showed greater transparency and lower chlorophyll-a and

macrophyte abundance than non-acidic lakes. They hypothesized that

acidification causes an "oligotrophication" process in which reduced

rates of organic matter decomposition and nutrient recycling lead to

lower rates of primary production. Surveys in Canada and the north-

eastern U.S. have shown similar trends in transparency, chlorophyll-a,

and macrophyte abundance (Dillon et al. 1978; Hendrey et al. 1976).

However, other studies have shown trends in phytoplankton production

and nutrient concentrations which were not consistent with the









oligotrophication theory (Dillon et al. 1979; Hendry and Brezonik

1984). Although acid deposition has been studied less intensively in

the southeastern U.S., some trends in acid deposition and its aquatic

effects have been demonstrated for Florida. Brezonik et al. (1980)

found that the northern two-thirds of Florida receives a mean annual

rainfall pH of 4.7 or less and excess sulfate deposition (non-marine

origin, based on S04:C1 ratios) around 20 kg/ha'yr. Furthermore,

Florida has approximately 2500 lakes that are sensitive to acidifica-

tion based on the criterion of alkalinity <100 peq/L (Hendry and Brez-

onik 1984). In a study of 20 soft-water Florida lakes over the pH

range 4.6-6.7, Brezonik et al. (1984) found strong correlations of

phytoplankton, chlorophyll-a, and total phosphorus with pH. Data from

165 Florida lakes were analyzed by Canfield (1981), who concluded that

the relation between pH and chlorophyll-a was due to a strong correla-

tion of TP with pH rather than to other factors related to pH. How-

ever, his data base included many hard-water and eutrophic lakes, which

would tend to mask a pH-TP or pH-chlorophyll-a correlation in soft-

water lakes.

These survey results have left it unclear whether the low TP con-

centrations (and thus low chlorophyll-a and phytoplankton levels) in

acidic lakes are a consequence of low pH (as suggested by Grahn et al.

1974) or whether they reflect the conditions that originally make the

lakes susceptible to acidification. Lakes with small watersheds

receive a large proportion of their water and phosphorus inputs from

rainfall directly to the lake surface. There is thus little opportun-

ity for watershed buffering, and phosphorus loading rates to such lakes

are low.









Phosphorus is a major nutrient requirement of primary producers in

aquatic and terrestrial habitats. In lakes, phosphorus availability is

the factor which most often limits phytoplankton production. Increased

cultural input of phosphorus was recognized as the primary cause for

the eutrophication of many lakes during the 1960's and 1970's. The key

role of phosphorus in the eutrophication process led to much research

on ways to control or reduce phosphorus levels in lakes. Although this

research necessarily included the processes involved in phosphorus cyc-

ling, the primary emphasis was on productive lakes. Some phosphorus

cycling studies have considered temperate oligotrophic lakes, but few

have included unproductive subtropical lakes.

Lake acidification seems to have the opposite effect of eutrophi-

cation, but the role of pH in determining lake productivity remains a

controversial issue. As mentioned earlier, lake surveys from different

geographic areas have found decreasing TP concentrations as [H+]

increases. On the other hand, although acidic lakes generally tend to

be unproductive, some evidence indicates that acid lakes are no less

productive than similar, oligotrophic lakes with circumneutral pH

values (Dillon et al. 1979).



Objectives


This dissertation addresses the hypothesis that acidification of

Florida lakes can directly or indirectly affect their total phosphorus

concentrations. The approach taken to answer this question involved a

combination of laboratory and field studies designed to accomplish the

following major objectives:









1. to characterize phosphorus cycling in an acidic, subtropical

lake, and

2. to assess the effect of [H+] (acidification) on the major

processes involved in phosphorus cycling. These include plank-

tonic uptake and turnover of phosphorus, exchange reactions

between lake water and sediments, and the release of phosphorus

from decomposing organic matter.



Site Description


McCloud Lake is a small, soft-water lake located in the Trail

Ridge area of north-central Florida, about 40 Km east of Gainesville in

Putnam County (Figure 1-1). Past studies of the lake include its use

as a control in a whole-lake nutrient enrichment experiment on nearby

Anderson-Cue Lake in 1966-1969 (Brezonik et al. 1969) and quarterly

sampling during a 1978-1979 survey of the chemistry and biology of 20

Florida lakes (Hendry and Brezonik 1984). The region is characterized

by sparse vegetation and numerous small lake basins perched among sandy

hills. Longleaf pine/turkey oak assemblages provide a broken canopy,

while lichens and wiregrasses dominate the understory and open areas.

The McCloud Lake watershed (1 Km2) is uninhabitated and is part of a

controlled-access area, the University of Florida Katharine Ordway

Ecological Preserve.

Surface soils consist of non-spodic marine sands (Candler typic

quartzipsamment, cation exchange capacity %2.5 meq/100 g) that allow

very little overland runoff to the lake. Other components of the

unconfined (non-artesian) surface aquifer include gravels and sandy







LITTORAL MESOCOSMS


JANUARY 1982


I I


25 50


Figure 1-1.


Bathymetric map of McCloud Lake, September 1982 (contour
interval = 2 ft).


t
N


100m










clays of the Citronelle Formation. The sandy clays, clays, and phos-

phatic sands of the Hawthorne Formation constitute a relatively imperm-

eable confining layer (24-30 m thick) which separates the limestone of

the artesian Floridan Aquifer from the perched shallow water table.

Since there are no surface inflows or outflows to McCloud Lake, the

only sources of water are rainfall directly to the lake and subsurface

seepage. The hydraulic residence time of the lake is about 9.6 years

(Baker 1984).

McCloud Lake occupies a sub-rectangular solution basin (Figure

1-1) which had a surface area of about 5 ha and a maximum depth of 5 m

during 1980-1982. Water level and surface area vary widely in response

to long-term rainfall patterns. In 1966-1967, the maximum depth was

nearly 6.5 m and the surface area was about 9 ha, while in 1968 the

surface area was reduced to 6.8 ha and maximum depth was only 5.5 m

(Brezonik et al. 1969).
















CHAPTER 2
LITERATURE REVIEW



Phosphorus cycling in aquatic environments is the result of many

processes which involve different phosphorus forms and numerous storage

compartments, as generalized in Figure 2-1. While inputs and losses

determine the total phosphorus concentration in a lake, within the

lake soluble inorganic phosphorus is incorporated into organic com-

pounds by primary producers, cycled through dissolved and particulate

compartments, and returned to inorganic form. The duration of individ-

ual processes can vary from seconds to days or months, and the relative

importance of any particular compartment or transformation varies from

one lake to another. Because of the importance of phosphorus as a

major plant nutrient and its role in lake eutrophication, many re-

searchers have studied the aquatic phosphorus cycle. Their approaches

have ranged in scope from focusing on one process or compartment, to

complex mathematical models designed to simulate the major processes

that control lake phosphorus concentration. The following review

considers research in the major areas included in this study.



Sediment-Water Exchanges


Lake sediments act as a net sink for phosphorus through accumula-

tion of particulate organic matter, but under certain conditions they























WATER




















SEDIMENT


Figure 2-1. Generalized pathways of phosphorus cycling in lakes (after
Syers et al. 1973).










can constitute a significant source to lake water. Early studies of

the effectiveness and fate of phosphate fertilizers have contributed to

our understanding of the behavior of phosphorus in sediments. The

realization that a large fraction of phosphorus in fertilizer was fixed

or retained in soil led to investigations of the mechanisms involved.

Coleman (1944a, 1944b) found that the presence of iron and aluminum

oxides was more important for phosphorus fixation than the type of clay

in coarse or fine soil fractions, and he suggested that retention

involved an exchange of OH" for H2PO4- at the oxide surface.

Other soils researchers corroborated the importance of iron and alum-

inum oxides in phosphate fixation (Swenson et al. 1949; Haseman et al.

1950) and demonstrated that some organic acids can decrease phosphate

retention over specific pH ranges by forming complexes with the Fe and

Al (Struthers and Sieling 1950; Bradley and Sieling 1953).

It was unclear whether phosphate retention involved adsorption or

precipitation until Fried and Dean (1955) concluded that because a

large portion of fixed phosphate was exchangeable with carrier-free

inorganic 32P, adsorption must be the principal mechanism. Hings-

ton et al. (1967) demonstrated that retention of phosphate on oxide

surfaces is a specific adsorption mechanism which is independent of the

properties of the diffuse double layer or the outer Helmholtz layer.

Adsorption is accomplished by exchange of the anion for water and

hydroxyl ions at the oxide surface, and the reaction always results in

a decrease in surface charge. They further pointed out that an undis-

sociated free acid and its most highly charged anion are not adsorbed

if present alone because of the requirement for a proton donor and

acceptor for specific adsorption to occur. Maximum adsorption of










phosphate increases as pH decreases, with a discontinuity near each pK

value, and H2P04- is the form most readily adsorbed.

Mortimer (1941, 1942) included phosphorus in his studies of mud-

water exchanges of dissolved substances. He found that anoxic condi-

tions at the sediment-water interface cause an increased release of

phosphate to the water. Numerous other studies have reaffirmed the

relation between anoxic bottom water and enhanced release of phosphate

from sediments (Porcella et al. 1970; Li et al. 1972; Syers et al.

1973; Kamp-Nielsen 1974; Fillos and Swanson 1975; Armstrong 1979).

Reduced forms of iron and manganese are soluble, and mobilization of

these elements from lake sediments into anoxic bottom water also solu-

bilizes adsorbed inorganic phosphorus (Mortimer 1971; Syers et al.

1973; Armstrong 1979). Vertical mixing processes can recycle the phos-

phorus to the euphotic zone, where it would be available for biological

uptake. Kamp-Nielson (1974) reported a linear relationship between the

release of phosphate and its concentration gradient across an anaerobic

mud-water interface, but he found that sorption reactions dominated

phosphate exchange under oxygenated conditions. Other workers (Hynes

and Grieb 1970; Fillos and Swanson 1975) have reported sediment phos-

phate release under aerobic conditions, but at a much slower rate.

The amount of sediment inorganic phosphorus available for release

to overlying lake water depends on the size of this phosphorus pool and

on sediment characteristics. Li et al. (1972) estimated exchangeable

inorganic sediment phosphorus by following the rate of disappearance of

inorganic carrier-free 32P from solution in well-mixed sediment-

water systems. The exchangeable fraction of four Wisconsin lake sedi-

ments ranged from 19% to 43% of total inorganic phosphorus for both










aerobic and anoxic conditions, although a significant release of sedi-

ment inorganic phosphorus occurred under anoxic conditions. Porcella

et al. (1970) set up microcosms with sediments as the only source of

phosphorus for algal growth. They observed a repeatable series of

events in which phosphorus released to an anaerobic layer above the

sediment surface led to a benthic mat of the blue-green alga Oscilla-

toria sp., followed by a bloom of the same species in the overlying

water. The authors concluded that the Oscillatoria mat enhanced sedi-

ment phosphorus release by disrupting the sediment-water interface when

bubbles occasionally lifted portions of the mat and attached sediment.

Biological reworking of sediments is another mechanism that can

accelerate sediment-water phosphorus exchange. Davis et al. (1975)

investigated the effect of burrowing tubificid worms on phosphorus

dynamics in intact mud-water columns. The worms caused an increased

removal of 32P from the water (this became bound to Fe and Al

oxides), but did not affect release of 32P back into the water. In

subsequent bioturbation studies (Gallepp et al. 1978; Gallepp 1979),

burrowing larvae of chironomid midges increased the phosphorus concen-

tration in overlying water, but the increase was attributed to excre-

tion rather than an accelerated sediment release.

Lake sediments can adsorb large amounts of added phosphorus in

addition to serving as an internal source of inorganic phosphorus.

This ability of sediments to buffer aquatic phosphate concentrations

has been pointed out by numerous workers (Carritt and Goodgal 1954;

Hayes and Phillips 1958; Phillips 1964; Pomeroy et al. 1965; Harter

1968). Carritt and Goodgal (1954) demonstrated that retention of










inorganic phosphorus by estuarine sediments involves a rapid initial

adsorption process followed by a slower diffusion reaction.

The phosphate adsorbed by sediments from a eutrophic Connecticut

lake (Harter 1968) was associated with two sediment fractions: a

loosely bound iron fraction and a more tightly bound aluminum fraction.

The work of Shukla et al. (1971) with sediments from nine soft-water

and five hard-water Wisconsin lakes showed that noncalcareous sediments

adsorbed more phosphorus than did calcareous sediments. Furthermore,

phosphorus adsorption by both sediment types was corelated more closely

with oxalate-extractable Fe than with any other parameter. The authors

postulated that adsorption occurred on a large complex consisting pri-

marily of hydrated Fe oxide, with smaller amounts of organic matter,

A1203 and Si(OH)4.



Decomposition


The amount of inorganic phosphorus present in sediment interstit-

ial water is affected by sediment characteristics and by decomposition

of organic matter. The effect of various environmental parameters on

phosphorus release from decomposing algae and aquatic macrophytes has

been studied in field and laboratory situations. In a summary of pre-

vious studies, Foree et al. (1970) listed three general stages in the

nutrient regeneration process: (1) A rapid ("24 h) initial step in

which nutrients are released, absorbed, or released and then re-

absorbed; (2) a stationary phase of several days with no net change in

nutrient concentration; and (3) active net release of nutrients to

solution over several hundred days.










Foree and McCarty (1968) followed phosphorus release during the

anoxic decomposition of cultured algae. They found that after 200 days

of incubation about 40% of the initial particulate phosphorus remained

in refractory solids. In a related study, the same group (Foree et al.

1970) developed a mathematical model to describe phosphorus regenera-

tion under anoxic and aerobic conditions as a function of measurable

quantities. However, the applicability of their model is limited by

the fact that some of the terms can only be obtained after lengthy

laboratory decomposition studies. After one year of decomposition a

larger fraction of initial particulate phosphorus remained in the

aerobic (,50%) than in the anoxic (040%) experiments.

Nichols and Keeney (1973) followed the release of phosphorus from

herbicide-killed aquatic macrophytes (Myriophyllum exalbescens) in

water-only and water-plus-sediment systems. They found a rapid initial

release of soluble organic phosphorus after the plants were killed,

followed by an increase in inorganic phosphorus. Levels of inorganic

phosphorus were lower in the systems that contained sediments. The

authors concluded that phosphorus released from plants decomposing in a

lake was available for incorporation into biomass or adsorption onto

sediments.

Acid and alkaline phosphatase enzymes produced by bacteria and al-

gae are important in the remineralization of organic phosphorus com-

pounds. Reichardt (1975) found sharp increases in bacterial biomass

and alkaline phosphatase activity in the first 1 cm of lake sediments

and noted that below the upper aerobic layer, bacterial densities de-

creased while enzyme activity remained nearly constant. He concluded










that the phosphatases at this sediment depth were longer-lived than the

bacteria that produced them.

Landers (1982) conducted field decomposition studies in the lit-

toral zone of a soft-water Indiana reservoir. He isolated areas with

and without naturally senescing Myriophyllum spicatum in open-ended

plastic enclosures and observed changes in nutrients and chlorophyll-a

over 119 days. Phosphorus released from the macrophytes (extrapolated

to a whole-lake basis) equalled about 2-18% of the total annual phos-

phorus loading to the lake, and concurrent increases in chlorophyll-a

indicated a significant phytoplankton response to the release.



Planktonic Phosphorus Cycling


Observations that nearly undetectable levels of dissolved inor-

ganic phosphorus are often adequate to support phytoplankton blooms led

to the hypothesis that phosphorus cycling within the water column is a

rapid process. The use of radioactive 32P has facilitated accurate

estimates of planktonic phosphorus uptake rates and turnover times with

addition of as little as 0.002% of ambient inorganic phosphorus levels.

Hayes and Phillips (1958) and Phillips (1964) used 32P to study

phosphorus equilibrium in systems containing mud, water, plants and

bacteria; they summarized their findings and earlier work to provide an

integrated concept of phosphous cycling among the components of a

whole-lake system. Their estimated turnover times included 1 week for

the water of a whole lake; 0.3 days for bacterial or phytoplankton

cells (but 5 min for initial equilibration); 3-4 days for rooted

aquatic macrophytes; and 1 day for zooplankton (which can utilize only










organic phosphorus). The authors emphasized the influence of bacteria

in retaining phosphorus in the water column through incorporation into

organic forms (and thus preventing adsorption by sediments) or by

accelerating the return of phosphorus from the sediments.

Rigler (1964) examined water column phosphorus fractions in dif-

ferent types of temperate North American lakes and found that soluble

organic phosphorus (SOP) represented about 18% of total phosphorus (TP)

in all trophic types. He attributed wide variations in SOP reported in

the literature to variations in filter pore size and methods used to

remove seston from the water. He found that turnover of inorganic

phosphorus by seston (using carrier-free 32p) was less than 10 min

in all eight lakes during the summer, and increased in winter. Rigler

(1966, 1968, 1973) later contended that his 32P uptake data demon-

strated that colorimetric analyses overestimate inorganic phosphorus

concentrations. However, as pointed out by Lean and White (1983), the

inconsistency in Rigler's data was due to his failure to consider that

plankton could excrete unlabelled inorganic phosphorus, rather than

overestimation.

Understanding of planktonic phosphorus cycling was advanced by

Lean (1973a, 1973b), who used Sephadex gel to separate soluble 32p

fractions on the basis of molecular size. From his results with this

technique in a eutrophic Canadian lake, Lean proposed a generalized

description of phosphorus movement between biologically important

forms. He demonstrated the rapid formation of a dissolved algal or

bacterial organic phosphorus compound (molecular weight about 250) that

becomes associated with a high-molecular weight colloid, releasing

orthophosphate in the process. The colloidal phosphorus form comprises










a large proportion (about 77%) of nonparticulate phosphorus, but this

fraction is not available for algal uptake.

Both chemical and radioisotope methods can be used to study phos-

phorus uptake by lake plankton. The application of these methods and

the significance of their results were recently reviewed by Lean and

White (1983), who used both techniques to estimate phosphorus uptake

rate constants for the same lake samples. They concluded that small

cells dominate uptake when low amounts of phosphorus are added, while

at high added phosphorus concentrations, uptake is primarily by large

cells. The comparison of phosphorus uptake rates by seston from dif-

ferent lakes is therefore practically impossible because of differences

in size distribution of plankton and variations in amounts of phos-

phorus added by different researchers.

Zooplankton phosphorus excretion has long been recognized as a

recycling mechanism in the water column, but there has been little

agreement about its relative importance or whether organic or inorganic

forms predominate. Johannes (1965) investigated interactions between

marine protozoa and bacteria and their effect on phosphorus cycling.

He found that protozoan phosphorus excretion (per unit weight) is 1-2

orders of magnitude faster than that of marine microcrustaceans, and

several orders of magnitude faster than marine macrofauna. The proto-

zoan-bacterial interaction involves consumption of organic detritus by

bacteria, which in turn are grazed by protozoans. Bacterial popula-

tions are maintained in a state of "physiological youth" by protozoan

grazing, thus increasing regeneration of inorganic phosphorus. Buech-

ler and Dillon (1974) found that phosphorus uptake by freshwater cili-

ated protozoans (Paramecium spp.) was effected by ingestion of bacter-

ial biomass, and that phosphorus turnover rates were extremely fast.






18


Hargrave and Geen (1968) measured rates of excretion of unlabelled

soluble phosphorus for several species of marine crustaceans and one

rotifer. Although soluble organic phosphorus constituted up to 75% of

the amount regenerated, they calculated that zooplankton released

enough inorganic phosphorus to the photic zone to supply one-fifth to

two times the daily phytoplankton requirement. In all cases the mea-

sured excretion rate was decreased by increased bacterial activity and

experimental duration. Use of Sephadex to fractionate labelled phos-

phorus compounds has provided an explanation for these observations

(Peters and Lean 1973; Peters and Rigler 1973). This technique showed

that about 90% of soluble phosphorus released by Daphnia rosea and

Diaptomus minutus was inorganic phosphorus. However, bacteria quickly

assimilated most of the released inorganic phosphorus, which accounts

for the relation Hargrave and Geen (1968) found between phosphorus

excretion and bacterial activity or length of incubation. Bacterial

uptake also explains the high proportion of SOP found by earlier work-

ers. Peters and Rigler (1973) further estimated that overall phos-

phorus cycling efficiency of zooplankton [(P regenerated)/(P regener-

ated + P sedimented) x 1000] is as high as 88-93%, which again empha-

sizes the important role of zooplankton excretion in maintaining phos-

phorus in the water column.

As pointed out by Johannes (1965) for marine zooplankton, there is

an inverse relationship between body size or body weight and rate of

phosphorus regeneration, so that small species are potentially more

important in regenerating soluble phosphorus than larger forms. This

relationship has been noted by numerous others (e.g., Hargrave and Geen

1968; Peters and Rigler 1973), and it implies that a lake's trophic










state could be affected by processes which change the size distribution

of its zooplankton.

Fish constitute another influence on water column phosphorus cyc-

ling. Kitchell et al. (1975) employed a mass-balance approach to eval-

uate the importance of phosphorus flux through fishes. They calculated

that production of fish biomass fixes 60-70% of the annual phosphorus

input to Lake Wingra, Wisconsin. The fraction that is incorporated

into bones and scales ('50%) will not be remineralized through decompo-

sition, and thus is effectively lost to the system. However, the auth-

ors suggest that the seasonal pattern of fish mortality in temperate

lakes (high mortality after spring spawning) results in a significant

supply of phosphorus from decomposing fish biomass in late spring and

early summer.



Lake Phosphorus Models


Efforts to model phosphorus dynamics in lakes have shown varying

degrees of success, depending in part on the complexity and objectives

of the modeling efforts. Lake management applications began with a

simple mass balance approach to lake phosphorus concentration. This

involves estimating the change in lake phosphorus storage that results

from a balance of loading terms and loss terms. Non-point sources are

usually estimated from land use data, while sedimentation rates fre-

quently are derived from the literature and adjusted to calibrate the

model.

By examining the relationship between phosphorus loading and

trophic state indicators (e.g., total phosphorus, chlorophyll-a, Secchi

depth, hypolimnetic oxygen deficiency), Vollenweider (1975), Dillon and

Rigler (1974) and others have developed critical loading limits, below










which eutrophication could be avoided. Furthermore, these relation-

ships could be used to predict the effect of changes in loading rates

on lake phosphorus concentration, including the time required to reach

a new equilibrium after a change in input. The application of some of

these models has resulted in the need to modify them to fit observed

conditions. Yeasted and Morel (1978) used a combination of phosphorus

budget modeling and stepwise discriminant analysis to evaluate the

ability of water residence time, mean depth, and lake surface area to

describe the non-conservative behavior of phosphorus in 128 phosphorus-

limited lakes (71 eutrophic, 42 mesotrophic, and 15 oligotrophic).

They found that only hydraulic residence time gave consistent statisti-

cal significance. Shannon and Brezonik (1972) and Baker et al. (1981)

have developed nutrient loading-trophic state relationships specific to

Florida lakes.

In spite of the problems inherent in the application of mass bal-

ance models, their very simplicity makes them an attractive management

tool. They can be used with a reasonable degree of accuracy to simu-

late and evaluate the effect of various options, provided data are

collected carefully and the assumptions are not violated.

In contrast to the simple mass balance approach are more complex

approximations of the non-conservative behavior of phosphorus within a

lake. These models use differential equations to represent changes in

various processes (e.g., production of phytoplankton biomass) as a

function of time. Models involving phosphorus range from those that

consider only phytoplankton as a biotic component (Fleming 1975) to

more complex, multi-component ecosystem models (Chen 1969). The

simpler models do not yield realistic results, but it is difficult to

obtain all the coefficients needed in more complicated models.






21


Nevertheless, ecosystem modeling provides the only feasible alternative

for integrating so many processes and parameters.



Effects of pH on Phosphorus Cycling


Acid deposition and the acidification of surface waters are recent

enough phenomena that most research has been focused on identifying-

effects and documenting the extent of affected areas, instead of ident-

ifying the mechanisms involved. As mentioned earlier, while many stud-

ies have shown decreasing TP levels with decreasing lake pH, there is

little evidence to link the observed TP decrease to acidification.

Most available information concerning pH effects on phosphorus dynamics

relates to sediment-water interactions and decomposition.

MacPherson et al. (1958) examined the effect of pH on the parti-

tioning of inorganic phosphorus between water and the mud of unproduc-

tive, moderately productive, productive, and acid bog lakes. They

found similar trends in all lake types, with minimum phosphorus release

from the mud in the pH range 5.5-6.5. More inorganic phosphorus was

released at higher and lower pH values. The acid bog and productive

lake muds did not adsorb appreciable amounts of added phosphorus; the

unproductive lake mud removed most of the added phosphorus in the acid

pH range but not at pH 7. Increased phosphorus adsorption as pH

decreases has been shown for soils (Lopez-Hernandez and Burnham 1974).

Andersson et al. (1978) and Gahnstrom et al. (1980) varied the pH of

water overlying sediment cores from acidic and alkaline Swedish lakes.

They found more inorganic phosphorus was released from the sediments at

high pH than at low pH.









Consideration of the effect of pH on solubility of metal phos-

phates shows the controlling phases under equilibrium conditions (Stumm

and Morgan 1981). At low pH strengite (FePO4) and variscite

(AlP04) are the solid phases that may control phosphate solubility,

while above pH 6 calcium phosphates (notably apatite) are the predomin-

ant solids. However, due to rapid biological transformations, equilib-

rium is rarely attained in the pH range of natural waters.

Singer et al. (1983) added 32P to the water over intact sedi-

ment cores from an acidic Adirondack lake. Some had a mat of Sphagnum

sp., and two concentrations of aluminum were used (0 and 300 yg/L).

They concluded that the algal mat was much more efficient than bare

sediment at removing water column phosphorus, and that precipitation

reactions (with Al concentrations up to 300 pg/L) were unimportant.

Grahn et al. (1974) theorized that acidification inhibits decompo-

sition because of the accumulation of organic matter which they

observed in the sediments of many acid lakes. Studies on the effect of

pH on decomposition have been inconclusive. Some measures of decompo-

sition, such as leaf litter weight loss and numbers of total bacteria,

decrease at low experimental pH (Leivestad et al. 1976; Traaen 1980).

Others (sediment oxygen demand, glucose turnover) show no effect of

reduced experimental pH (Andersson et al. 1978; Gahnstrom et al. 1980),

although sediment oxygen demand and glucose turnover both increased in

lakes after lime treatment. Finally, the experimental acidification of

a Canadian lake (Schindler 1980) resulted in no significant change in

TP and no evidence of decreased decomposition over 3 years. However,










it should be pointed out that the pH change during this period was only

from 6.6 to 5.6.

Another potential effect of lake acidification relates to the

importance of pH in controlling phosphate uptake by algae (Wetzel

1983). Different species have distinct pH ranges in which they show

optimum growth and phosphate uptake. This is due in part to the pH

specificity of extracellular or membrane-associated enzymes, but pH can

also alter the permeability of the cell membrane and change the ionic

form of inorganic phosphate in the growth medium.
















CHAPTER 3
PHOSPHORUS DYNAMICS IN MCCLOUD LAKE



This chapter includes a discussion of the general limnology of-

McCloud Lake as well as the contributions of important processes and

storage compartments to the dynamics of phosphorus cycling within the

lake. The historical data base for McCloud Lake is examined, as well

as the results of laboratory and in situ experiments.



Materials and Methods


Routine Sampling

Limnological data were collected monthly at a mid-lake station

from October 1980 through September 1982. Field data included Secchi

disk transparency, and temperature and dissolved oxygen profiles, which

were measured with a YSI model 54A DO meter. In the laboratory

specific conductance was measured with a YSI model 31 conductivity

bridge; pH was measured with a Fisher Accumet model 230A pH/ion meter

equipped with an Orion internal reference calomel electrode. Chemical

samples were collected at 1.0-m intervals and stored in separate poly-

ethylene bottles for major ions (conc. HNO3 to pH < 2) and nutrients

(1 mL saturated HgC12 per L of sample). Chlorophyll-a and phyto-

plankton samples were collected as a water column composite (1-m inter-

vals), while zooplankton were collected by vertical tows of a #20

Wisconsin plankton net (80 pm mesh). Phytoplankton and zooplankton










were preserved with 1% Lugol's iodine and 5% buffered formalin,

respectively.

Standard procedures were followed for all chemical analyses (APHA

1980; U.S. EPA 1979). Major cations were analyzed (flame mode) on a

Perkin-Elmer Model 5000 atomic absorption spectrophotometer. Chloride,

sulfate, silica, and nutrient forms were analyzed by automated colori-

metric procedures (Table 3-1). Semi-micro digestion procedures were

used for total phosphorus (autoclaved persulfate digestion) and total

Kjeldahl nitrogen (block digestion).

Chlorophyll-a was measured by the trichromatic method (APHA 1980).

Phytoplankton aliquots (10-30 mL) were concentrated in Utermohl cham-

bers and counted using methods described by Lund et al. (1958). Zoo-

plankton were identified and counted in 1-mL Sedgwick-Rafter cells

under a light microscope.


Phosphorus Budget

Water budget. Monthly data for precipitation, seepage, evapora-

tion, and change in McCloud Lake level (stage) were compiled by Baker

(1984), who also discussed the equipment and methods used to collect

the data. The resultant water budget is a necessary prerequisite for

the construction of a phosphorus budget for the lake.

Stage-area and stage-volume relationships were determined from a

bathymetric map of McCloud Lake (Figure 1-1). I constructed the map

from 14 fathometer transects and aerial photographs taken in January

1982, which corresponded to the lowest lake stage during the 1980-1982

period. Nine north-south and five east-west transects were run using a

Lowrance model 1510B Truline recording fathometer in a skiff powered by










Table 3-1. Automated colorimetric procedures used in McCloud Lake
study.



Parameter EPA Method


Chloride 325.1

Sulfate 375.2

Silica 370.1

Ammonium 350.1

Nitrate + Nitrite 353.1

Phosphorus 365.2










a small outboard motor at constant speed. The ends of the tranects

were marked with black plastic sheeting and white plastic milk bottles

to ensure their visibility in the subsequent aerial photography. In

addition, the distances from markers to the beginning or end of each

fathometer transect were recorded.

An outline of the lake (including the transect markers) was drawn

from the best aerial photograph, and the scale was determined from

measured distances between markers. Changes in depth along each

transect were plotted on this outline map, and contours were drawn at

2-foot depth intervals. Lake stage was 2 feet higher on September 22,

1982, than when the bathymetric survey and aerial photography were

conducted. This new shoreline was added to the bathymetric map from

measurements of distances between the transect markers (which were

still marked by wooden stakes) and the new lake shore.

The volume and area of the lake were calculated for each date with

a Hewlett Packard 9810A calculator equipped with a digitizer surface.

These data were used to establish the stage-area and stage-volume

relationships needed for the water budget calculations.

Atmospheric phosphorus loading. Rainfall samples collected at the

lake from September 1981 to August 1982 were analyzed for TP. Total

phosphorus deposition rates (wetfall and dryfall) were estimated from

the relation between wet and total phosphorus deposition measured at

several sites in Florida (Brezonik et al. 1983).

Sedimentation. Rates of sedimentation of particulate matter and

phosphorus were measured in McCloud Lake with cylindrical sediment

traps (Figure 3-1). Design of the traps followed the general recommen-

dations of Blomqvist and Hakanson (1981), who published an extensive



















































SEDIMENT
S- SURFACE


Figure 3-1.


Sediment trap design.


FLOTATION

./ \


- -










review of the design and performance of sediment traps in aquatic sys-

tems. They concluded that a simple cylinder gives better results than

other shapes when a proper height-to-diameter ratio (H/D) is used to

limit resuspension losses. In general they recommended a vessel with

diameter > 20 mm and H/D > 3 or 4.

The base of the traps consisted of a plexiglass rectangle (20 x 27

cm) with six holes for the cylinders and plastic foam for flotation.

Rubber bands around the Pyrex test tubes (22 x 150 mm, H/D = 6.7) pre-

vented them from falling through the holes. This buoyant trap appar-

atus was moored "0.4 m above the lake bottom in the center of the lake

(1'4.5 m total depth). Removal of the large flotation bucket allowed

the trap platform to float up to the surface to facilitate recovery of

the test tubes with minimal disturbance of the sediments.

Three of the six tubes were placed upright to trap sedimenting

particulate matter, while the other three were installed upside-down to

estimate the biomass of colonizing invertebrates and attached algae.

At the beginning of the first incubation period the platform was raised

to the surface and six tubes were installed. The platform was then

carefully lowered from the surface using the mooring line, and the flo-

tation bucket was attached approximately 0.5 m below the lake surface

to maintain a constant tension.

At the end of each incubation period (30 or 60 days), I carefully

swam down to the trap array and inserted rubber stoppers in all six

tubes. The bucket was removed from the mooring line so the platform

could be raised to the the surface for recovery of the tubes. After

new tubes were installed the platform was again carefully lowered to

initiate a new incubation.










The test tubes were tared in the laboratory prior to incubation.

After an incubation, the outsides of the stoppered tubes were carefully

cleaned to remove any attached particulates. The stoppers were removed

and the tubes were placed in a drying oven ('60C) to evaporate all

water. When the contents were dry, each tube was cooled and reweighed

to allow calculation of sedimentation rates on a dry-weight basis.

Next, 20 mL of distilled deionized water was added to each tube and a

wet persulfate digestion was performed in the autoclave. Total phos-

phorus was measured as SRP in the filtered digestate.


McCloud Lake Phosphorus Compartments

In addition to the water column phosphorus analyses previously

described, two other in-lake reservoirs of phosphorus were evaluated.

Macrophyte survey. At the beginning of this study in December

1980, McCloud Lake sediments were relatively barren and free of algae

or higher plants except in the very shallow littoral areas, where some

emergent species were found. However, during the study two submersed

macrophytes, Websteria ap. and Eleocharis sp., became established in

significant proportions in both littoral and deeper areas of the lake.

In September 1982, a survey was conducted to determine the areal extent

and density of these macrophytes and to estimate the amount of phos-

phorus bound in their biomass.

I sampled 14 transects spaced around the lake by swimming (with

SCUBA equipment) from the shore out toward the lake center. Each

transect started at a marker used in the bathymetric survey in order to

facilitate mapping the macrophytes. Data recorded for each transect

included the distances from shore and depths at which the macrophytes










beds began and ended, as well as the dominant species. In addition, a

composite macrophyte sample was obtained for each transect by manually

collecting all plants within a quadrat (0.016 m2) which was randomly

placed at four points more or less evenly spaced along the transect.

Additional samples of Eleocharis and Websteria were collected for

digestion and TP analysis.

The composite samples were carefully washed in the laboratory and

dried in tared envelopes at 60"C to get dry weight per unit area. Sub-

samples (0.5 g) of dried tissue were ashed at 5000C and ash-free dry

weight was calculated. The samples collected for phosphorus analysis

were also washed and dried at 600C. Weighed subsamples (%0.1 g) were

placed in large test tubes; distilled deionized water (20 mL) was added

and a wet persulfate digestion was performed in the autoclave. Total

phosphorus was measured by the SRP procedure on an aliquot of the

filtered digestate.

Sediment phosphorus. A composite surficial sediment sample was

collected from the center of McCloud Lake by pooling 6 grabs of a

petite Ponar dredge. Subsamples of this sediment were analyzed for TP

using a method which involved ashing at 550C followed by HCI diges-

tion, as described by Andersen (1976).

Phosphorus uptake. Rates of phosphorus uptake and turnover were

measured for the submergent macrophytes and for mid-lake seston using

radiolabeled (32P) orthophosphate. Intact Eleocharis plants were

collected from the lake, transferred to the lab, and carefully washed

to remove sediment and attached algae. The washed plants were blotted

dry and about 2.0 g of intact plants were placed in each of four PVC

trays in 250 mL of membrane-filtered (0.45 Um) lake water at room










temperature. Two trays were covered with aluminum foil to exclude

light, and the other two plus a control (250 mL filtered lake water

without plants) were incubated under fluorescent light.

An aliquot (1 mL) of a stock solution of 32P enriched ortho-

phosphate solution was added to each tray, and its disappearance from

the medium was followed by periodically withdrawing 1-mL aliquots.

These samples were placed in scintillation vials and counted with a

liquid scintillation counter.

Mid-lake seston samples (1 L) were incubated under fluorescent

light and slowly mixed with magnetic stirrers. Aliquots of a stock

K2HP04 carrier for 32p04 were added to the seston samples,

and uptake of 32P was followed by periodically withdrawing and

filtering 5-mL subsamples through 0.45 pm membrane filters. Methods

used to analyze 32P samples and to plot and analyze the data are

discussed in Chapter 4.



Results and Discussion


Limnology and Historical Nutrient Trends

McCloud Lake has uncolored, soft water that usually is clear.

Conductivity is low (n40 Vmho/cm), and divalent cation (Ca+2 +

Mg+2) concentrations are only about 150 peq/L. A Secchi disk is

visible on the lake bottom during winter months, but Secchi transpar-

ency is as low as 1.75 m during summer peaks of phytoplankton. Water

clarity is occasionally reduced when littoral sediments are suspended

by wave action, but the surrounding hills and small fetch minimize wind

influence on the lake.










The water column does not stratify, although bottom temperatures

and dissolved oxygen concentrations are generally somewhat lower than

surface values (Figure 3-2). The average difference between surface

and bottom temperature is 1IC; maximum differences up to 3C occur

during warm months. Percent oxygen saturation shows no difference

between surface and bottom waters in winter months, but oxygen satura-

tion is consistently lower near the bottom during warm months, reflect-

ing increased biological activity. Overall, oxygen saturation ranges

from 73% to just over 100%, and surface and bottom averages are 93% and

89%, respectively. The generally undersaturated conditions reflect the

oligotrophic status of the lake.

The pH of McCloud Lake decreased from 4.85 in 1967-1968 (Brezonik

et al. 1969) to 4.71 in 1978-1979, and generally was less than 4.60

during 1980-1982 (Table 3-1 and Figure 3-2). This represents nearly a

doubling of H+ concentration in 15 years. Present pH values

correspond closely to rainfall pH. The increase in conductivity from

1967-1968 (32 pmho/cm) to 1980-1982 (42 tmho/cm) indicates that concen-

tration of ions by evaporation may account for some of the pH decline.

This hypothesis is supported by the decrease in pH from September 1981

through January 1982 (Figure 3-2) which corresponded to the lowest lake

level in 2 years. Lake level began to rise as normal rains resumed in

February 1982, and the subsequent pH increase reflected this dilution.

Table 3-1 summarizes nutrient conditions in the lake during 1967-

1968 (Brezonik et al. 1969), 1978-1979 (unpublished data), and for the

2 years of this study. There has been remarkably little difference in

average nutrient concentrations over this 15-year period. Mean values

of several nutrient parameters (TON, TP, SiO2, TN/TP) for 1978-1979





















+ Sfc T

* Sfc 02
0 ot 02
0 Sot o


50'- Iri,
0 N DJ
80 i


I I I I I II I I I I I i I I
F M A M J J A S N DJ F M A M J J AS
81 i 82


4.O L II I fI t I I I II I I I I I I
0 N D J F M A M J J A S 0 N D iJ F M A M J J A S
80 81 82





Figure 3-2. Dissolved oxygen and pH trends in McCloud Lake, October
1980 through September 1982.


l00r


32


30
a-

I-
LU
22 O
LL

C,
18
CO

14


5.0r


SC4.5









Table 3-2.


Annual means and standard deviations (n = number of samp-
ling dates) of nutrient and limnological parameters for
McCloud Lake.


1967-68 1978-79 1980-81 1981-82

Parameter (n = 12) (n = 4) (n = 12) (n = 12)


pH 4.85 4.71 4.56 4.50 -

Conductivity* 32 44.8 42 + 2.6

TONT 420 243 290 + 120 423 + 220

NH4-Nt 105 137 62 + 50 56 + 40

N03-Nt 41 47 49 + 20 68 + 40

N02-Nt 1 2 1 + 0.4 1 + 1

TPt 12 16 9 + 3 12 + 7

SRPt 6 4 5 + 0.2 3 + 3

SiO2t 100 265 213 + 110 118 + 70

Chlorophyll-at 1.94 0.88 5.7 + 2.9 4.7 + 3.9

TN/TPS 47.3 26.8 44.7 45.7

SCa + Mg** 77 98 147

S04-2** 104 142 140

SCations** 205 234 313

SAnions** 271 287 310


*imho/cm.
tpg/L.
wt/wt.
**Ieq/L.









are not consistent with mean data from the other years, but this may

reflect the limited number of sampling dates in 1978-1979 rather than

changes in lake chemistry. TON, TP, and NH4-N were lower for

1980-1981 than for 1967-1968. Ammonium in 1980-1982 was about 50% of

1967-1968 levels, but both were low; TON and TP means were identical

for the two periods. TN/TP ratios also show little change over the 15

years since 1967-1968, with the exception of the 1978-1979 data.

According to criteria proposed for Florida lakes (Huber et al. 1982),

the TN/TP value of about 45 (weight basis) indicates that phosphorus is

the limiting nutrient in McCloud Lake.

The average chlorophyll-a (Table 3-1) during 1980-1982 (5.2 pg/L)

was more than two times as high as the mean value reported for 1967-

1968 (1.9 ig/L), and was nearly six times the 1978-1979 mean. Never-

theless, these chlorophyll-a levels all are indicative of oligotrophic

conditions and the differences appear insignificant. A relationship

established for TP and chlorophyll-a in phosphorus-limited Florida

lakes (Huber et al. 1982) predicts that McCloud Lake (TP 11 pg/L)

should have a mean chlorophyll-a concentration of 3.1 pg/L. This value

agrees well with measured chlorophyll-a and indicates that conditions

in McCloud are typical of those in phosphorus-limited oligotrophic

Florida lakes.

Figures 3-3 and 3-4 show monthly variations in McCloud nutrient

and biological parameters from October 1980 to September 1982. TP and

TON exhibited maximum values in spring and summer when chlorophyll-a

and total zooplankton abundances were highest. No trends can be dis-

cerned in variations of SRP. Both TP and SRP showed summer increases

during 1967-1969. Total phosphorus and TON were generally higher


__=Mimi



















-j

Z

0
z

0
z

z








-J


a.
z














z

0.
I
CL


150



100-


F M A M J J A S 0 N D J F M A M J J AS
81 82


30r


* TP o SRP


J F M A M J J AS
82


400r


S300


200
0
rO 100


O I I I I I I I I I I I I I I I I 1
O N DJ F M A M J A S 0 N D J F M A M J J A S
80 81 82


Figure 3-3. Variations in dissolved silica and nitrogen and phosphorus
forms in McCloud Lake, October 1980 through September 1982.


I


















800


3 600


S400

0
200-


0L I L I
0ONDJ
80
I


l l 1 1 1 1 1 1 1 1 1 1 1 1 1


F M A M J J A S 0 N 0 J F M A M J J A S
81 t 82

,956,600


0 Chi a o PHYTOPLANKTON


1 82
I


40 i
E

30-
z
20 '
Z
z
10 J
.,
0
I-
0 >-
0.


O L I I I I I I I I I I I I 1 I I I I I I
0 N D;J F M A M J J A S 0 N Dj F M A M J J A S
80 I 81 82
I I


Figure 3-4. Variations in TON and biological parameters in McCloud
Lake, October 1980 through September 1982.


600



400


2001










throughout the 1981-1982 period than for 1980-1981. The lower TP and

TON of 1980-1981 correspond to a drought and falling lake levels, while

the higher TP and TON of 1981-1982 reflect increased nutrient loadings

due to increased rainfall and rising lake levels.

Nitrate maxima occurred during winter months when biological

activity was lowest, and minimum nitrate concentrations corresponded to

peaks of phytoplankton and zooplankton abundance. Ammonium showed

peaks during warm months and low levels in winter. The sum of ammonium

and nitrate was almost never <50 pg/L; maxima of both species were

always <150 pg/L, and usually were <100 pg/L. The increase in nitrate

and concurrent decrease in ammonium which occurred during winter 1981-

1982 suggest that nitrification was occurring. Silica concentrations

also were low in winter and peaked in mid- to late summer, but never

reached levels (>0.5 ppm) considered optimal for diatom production

(Fogg 1975; Wetzel 1983).

Chlorophyll-a (Figure 3-4) showed peaks of algal production during

spring and fall in 1981 and early summer in 1982, although these trends

did not correspond closely to changes in phytoplankton abundance

(Figure 3-4). This was probably due to wide fluctuation in densities

of microflagellates and small green coccoid and spindle-shape phyto-

plankters. As a group ultraplankton (1-10 pm) represented an important

fraction of total phytoplankton numbers in the lake during both years

of the study. A large pulse in ultraplankton occurred during June

through September 1982, when total phytoplankton densities exceeded

50,000 cells/mL. With the exception of occasional pulses of Dinobryon

cylindricum, D. divergens, and to a lesser extent Asterionella sp., net

plankton (>50 pm) rarely constituted a major component of the phyto-










plankton. Oocystis gloeocystiformis consistently formed a large frac-

tion of total phytoplankton and often comprised greater than 50% of

total abundance. Other less abundant but common genera included Perid-

inium, Kirchneriella, Staurastrum, Mallamonas, Cosmarium, and several

unidentified penate diatoms. McCloud Lake exhibits an impoverished

phytoplankton community with few consistent seasonal trends in species

succession. Large masses of filamentous algae were observed in certain

areas of the littoral zone in early spring.

Peaks in total zooplankton abundance generally followed peaks in

chlorophyll-a. Maximum densities occurred during late summer and fall

1981 and mid-summer 1982 (Figure 3-4). Diaptomus mississippiensis,

Eubosmina tubicen, Diaphanosoma sp., and Keratella gracilenta usually

comprised 75-100% of total zooplankton numbers. No consistent pattern

of species succession was demonstrated, although D. mississippiensis

generally increased in importance during summer, while K. gracilenta

frequently dominated during winter months. Eubosmina constituted

40-50% of zooplankton totals during September and October 1981 and was

a principal sub-dominant in nearly every lake sample. The 34 zooplank-

ton species observed during 1980-1982 included four copepod, five clad-

oceran, and 25 rotifer species. Except for a reduced standing stock,

the zooplankton community of McCloud Lake closely resembles those found

in more productive Florida lakes of higher pH.


McCloud Lake Hydrology

Table 3-2 presents monthly precipitation amounts measured at

McCloud Lake from August 1981 through July 1982, as well as estimates










Table 3-3. McCloud Lake hydrology data, August 1981 through July 1982.



Precipitation, Lake Volume, Lake Area,
Date cm 103 m3 103 m2


1981

Aug

Sep

Oct

Nov

Dec

1982

Jan

Feb

Mar

Apr

May

Jun

Jul

TOTAL

MEAN


14.30

7.70

1.58

6.86

4.12


17.60

9.16

12.50

23.11

8.47

34.34

13.36

153.10


134.81

131.09

124.59

123.19

120.56



121.49

121.33

117.46

130.16

134.03

134.50

143.17


51.57

50.84

49.55

49.27

48.75



48.94

48.91

48.14

50.65

51.42

51.51

53.23


128.03


50.23










of lake volume and surface area calculated from lake stage measure-

ments. Sixty percent of the total annual rainfall occurred between

March and July 1982. The dry conditions between August 1981 and Febru-

ary 1982 caused lake volume to decrease by nearly 13% while surface

area decreased about 6.5%.

Baker (1984) constructed a water budget for McCloud Lake for the

time of this study. He found that precipitation accounted for 90% of

the total annual water input, and seepage into the lake contributed the

remaining 10%. The sandy soils in the watershed preclude significant

surface runoff and allow most rainfall to percolate directly to the

shallow water table rather than to the lake. Evaporation was the most

important mechanism for loss of water from McCloud Lake during the

study, although outseepage rates can exceed evaporation rates when the

shallow water table is low.


McCloud Lake Phosphorus Budget

Precipitation. Table 3-3 presents monthly phosphorus deposition

to McCloud Lake from August 1981 through July 1982. Wet deposition

values are based on TP concentrations in precipitation samples and

rainfall amounts. Total deposition rates were estimated from a previ-

ous study because dryfall samples were not collected at McCloud Lake.

Brezonik et al. (1983) monitored precipitation chemistry with a network

of 26 stations in Florida that included four wet-dry collectors (1

urban, 1 coastal, and 2 agricultural stations). Wet TP deposition

averaged 20% of total TP (wet plus dry) deposition at the four sites

for May 1978 through April 1979. However, dry deposition of phosphorus

was more important at the agricultural sites because of fertilizer use










Table 3-4.


McCloud Lake phosphorus storage and atmospheric loadings,
August 1981 to July 1982.


Phosphorus

TP Storage, Wet Wet, Total 1,a Total 2,a
Month Kg mg/m2 g g g


1981

Aug 1.35 0.810 41.77 126.6 167.1

Sep 1.57 0.405 20.59 62.4 82.4

Oct 1.25 0.126 6.24 18.9 25.0

Nov 1.36 0.755 37.20 112.7 148.8

Dec 0.60 0.376 18.33 55.5 73.3

1982

Jan 0.85 1.507 73.75 223.5 295.0

Feb 0.36 0.980 47.93 145.2 191.7

Mar 1.88 0.800 38.51 116.7 154.0

Apr 1.04 3.175 160.81 487.3 643.2

May 2.41 0.677 34.81 105.5 139.2

Jun 2.56 3.323 171.17 518.7 684.7

Jul 3.29 1.440 76.64 232.2 306.6

TOTAL 14.374 727.76 2205.2 2911.0

MEAN 1.54


aAssumes Wet = 0.33 total.
bAssumes Wet = 0.25 total.










and increased dust associated with agricultural practices. Since

McCloud Lake is not located in an area of intense agriculture, two

estimates of total TP deposition were calculated, based on assumptions

that wet deposition represented 25% and 33% of total TP deposition.

The estimated total TP loading to McCloud Lake ranged from 2.21 to

2.91 Kg/yr, or 43.9 to 58.0 mg/m2.yr. This range compares favor-

ably with the mean statewide total TP deposition of 51.0 Kg/ha*yr

reported by Brezonik et al. (1983), but it is more than double the

value they found at rural non-agricultural sites (27.0 mg/m2.yr).

Local land use patterns and annual climatic variations can strongly

affect nutrient deposition rates, and it is possible that McCloud Lake

is atypical of the rural non-agricultural sites monitored by Brezonik

et al. (1983).

Baker (1984) compared atmospheric nutrient deposition rates at

McCloud Lake to nutrient loading criteria established by Vollenweider

(1968, 1975) and Shannon and Brezonik (1972). He concluded that atmos-

pheric nitrogen loading exceeded the minimum input required to sustain

mesotrophic conditions, and that atmospheric phosphorus loading was

less than half the minimum mesotrophic loading rate. However, his

estimate of phosphorus deposition was based on the average rate at

rural, non-agricultural sites from Brezonik et al. (1983). When atmos-

pheric phosphorus loading is estimated from precipitation samples

collected at McCloud Lake, two of the three minimum mesotrophic loading

criteria are exceeded (Table 3-4). Given the oligotrophic status of

McCloud Lake, this suggests that lake acidification may in fact

contribute to the low TP and production which are typical of acidic










Table 3-5. Atmospheric deposition of phosphorus at McCloud Lake and
loading criteria.



Minimum* McCloudt
Mesotrophic Total
Reference Units Loading Deposition


Vollenweider (1968) mg/m2"yr 44.0 43.9-58.0

Shannon and Brezonik (1972) mg/m3 yr 22.0 17.3-22.8

Vollenweider (1975) mg/m2.yr 100.0-110.0 43.9-58.0


*Calculated by Baker (1984).
tThis study.










lakes. However, more detailed studies using data for many lakes would

be necessary to provide a satisfactory answer to this question.

Mass balance. The seepage contribution to total phosphorus load-

ing to McCloud Lake was not evaluated in this study because water in

the seepage meters became anoxic, thereby promoting solubilization of

TP from the sediments. However, since seepage accounted for only 10%

of the annual water input, it was assumed that the relative importance

of phosphorus input by seepage was minor. This is supported by the

fact that SRP tends to be low in groundwater because it adsorbs to

clays and hydrous oxide surfaces. The budget summarized in Table 3-5

indicates that phosphorus has a very short residence time in McCloud

Lake (0.5-0.7 yr), which is in keeping with rapid SRP uptake rates and

internal phosphorus cycling mechanisms.


Sedimentation

Sediment traps were deployed in McCloud Lake for one 2-month and

three 1-month incubations. Gross monthly sedimentation rates were cal-

culated for total dry sediment and for total phosphorus (Table 3-6).

Particulate and phosphorus sedimentation rates were higher during sum-

mer months when lake productivity was at a maximum. Extrapolation from

the 158 days of measured sedimentation to annual figures yields a dry

sedimentation rate of 429 g/m2-yr and a phosphorus sedimentation

rate of 370 mg/m2.yr. These results are probably overestimates for

the lake as a whole since more sediment accumlates in the center of the

lake than in littoral areas, but several lines of evidence indicate

that the trap estimates are reasonable for the pelagic zone. First,

210Pb dating of one profundal McCloud sediment core indicates that











Table 3-6. McCloud Lake phosphorus budget, August 1981 through July
1982.


Precipitation

Wet (Kg P)

Dry (Kg P)

Total (Kg P)

Mean Storage (Kg P)

AS (Kg P)

Residence Time (Yr)


0.73

1.48-2.19

2.21-2.92

1.54

1.94

0.5-0.7










Table 3-7. Sediment trap results from McCloud Lake (mean + standard
deviation).



Dry Phosphorus
Sedimentation, TP, Sedimentation,
Dates g/m2.mo mg P'g dry wt mg/m2.mo


10/23-12/21/82 25.9 + 0.78 1.024 + 0.064 26.5 + 0.78

04/28-06/03/83 11.8 + 4.98 0.844 + 0.084 10.0 + 4.20

06/03-07/05/83 54.6 + 1.07 0.802 + 0.095 43.8 + 0.86

07/05-08/04/83 66.5 + 0.54 0.724 + 0.131 48.2 + 0.39


MEAN 36.9 + 22.7 0.849 + 0.127 31.0 + 15.3

ANNUAL RATE 429* 378.8t


*g/m2.yr.
tmg/m2.yr.










the recent annual sediment accumulation rate is about 300 g dry

wt/m2-yr (personal communication, Michael Binford, Florida State

Museum). The sediment trap estimate is 43% greater than the 210Pb

estimate, but this difference is probably within the range of hori-

zontal variation within the lake and annual variations due to changing

lake levels. The sediment trap measurements were obtained during a

lake level rise of "'2 ft, when plant litter from formerly terrestrial

areas would be incorporated into the expanding littoral zone of the

lake. Thus pelagic sedimentation rates would logically be higher dur-

ing rising water levels than when lake level remains fairly constant.

Second, although the annual atmospheric phosphorus loading rate

(44-58 mg P/m2*yr) is much lower than the measured phosphorus sedi-

mentation rate (370 mg P/m2.yr), the trap results provide a gross

annual rate that is applicable only to an undefined part of the pelagic

zone. A correction can be applied to account for phosphorus release

from the sedimented particulate matter, based on the difference between

TP in surface sediments and TP in particulates retained in the traps.

Surface sediments contain about 0.034% phosphorus discussedd later in

this chapter), while the sedimenting material contained approximately

0.085% phosphorus. Thus 60% of the phosphorus in sedimenting material

appears to be returned to the water column. Net phosphorus sedimenta-

tion therefore would be "150 mg/m2.yr, which is 2.6 to 3.4 times

the atmospheric phosphorus input.

The third observation that indicates sedimentation rates are

higher in the deeper areas of the lake relates to patterns of sediment

accumulation in the McCloud Lake basin. Extensive emergent and sub-

mergent macrophyte communities are found along the shore and in the










shallow littoral areas, although very little organic sediment accumula-

tion is noted there. On the other hand, pelagic sediments are as much

as several meters thick (personal communication, Michael Binford) and

highly organic (see Chapter 6). Therefore sedimentation rates measured

in the middle of McCloud Lake should not be applied to the lake as a

whole. At a minimum, these data apply to the area of lake bottom where

water depth > water depth at which the traps were located (%4.5 m). At

a maximum they represent the area of thick organic sediment accumula-

tion (water depth > %2-2.5 m).

Phosphorus Compartments

Macrophytes. Figure 3-5 shows the extent of coverage by macro-

phyte beds in McCloud Lake on September 30, 1982. This figure deline-

ates the areas of relatively uniform and continuous coverage by

Websteria and Eleocharis. Although emergent species (e.g., Leersia,

Fuirena, and Xyris) predominated shoreward of the submergent beds,

sparse Websteria stands commonly were observed among these emergents.

Likewise, Eleocharis beds did not end abruptly at the deeper border

indicated in Figure 3-5. The density of the Eleocharis bed decreased

with depth until coverage was no longer complete, but scattered

"clumps" of various sizes were observed even in the deepest areas of

the lake. This figure and subsequent discussions therefore represent

conservative estimates of the importance of Websteria and Eleocharis in

McCloud Lake.

These two species covered about 14,000 m2, or 26% of the total

lake bottom. Eleocharis comprised the majority of macrophyte coverage

(93%), and as Table 3-7 shows, this species also contained a much

larger proportion of phosphorus than Websteria. Although the density

















































DISTANCE IN METERS


WEBSTERIA 945 n

ELEOCHARIS 13180 m2


Figure 3-5. Extent of submerged macrophyte coverage in McCloud Lake,
September 1982.










Table 3-8. Physical characteristics and phosphorus content (mean and
standard deviation) of submergent macrophytes in McCloud
Lake, September 30, 1982.



Eleocharis Websteria


Dry Weight (g/m2) 39.8 + 28.5 ND*

Volatile Solids (g/m2) 25.6 + 17.6 ND*

Phosphorus (mg/g dry wt) 7.27 + 1.01 1.45 + 0.033


*Not determined.










of Websteria beds was not measured separately, these plants tended to

be much shorter and to provide a sparser cover than the Eleocharis.

Eleocharis therefore constituted a much larger in-lake phosphorus stor-

age than did Websteria. Using average values from Table 3-7, the total

Eleocharis biomass contained about 3.81 Kg of phosphorus, which was 2.5

times the amount of dissolved and particulate phosphorus present in the

water column of the lake at that time.

The significance of macrophyte phosphorus storage to water column

processes depends on the source of phosphorus which these plants

utilize. If they obtain most of their phosphorus from the water, then

Eleocharis and Websteria compete with planktonic primary producers,

but if the sediments supply their phosphorus, the macrophytes represent

a significant mechanism of sediment phosphorus mobilization. Carignan

and Kalff (1980) used 32P-labeled sediments to determine whether

nine rooted aquatic macrophyte species obtained their phosphorus from

sediments or water during in situ incubations in Canadian lakes. They

found that sediments were the only significant source of phosphorus to

the macrophytes in oligotrophic and mildly eutrophic lakes that had

relatively high interstitial phosphorus concentrations. Barko and

Smart (1980) obtained similar results with laboratory incubations of

three species of submersed macrophytes with minor root systems. They

further concluded that release of phosphorus from these species to the

water column was a result of tissue decay instead of excretory proces-

ses. It thus appears that Eleocharis and Websteria do not compete with

phytoplankton for phosphorus in McCloud Lake. However, based on the

phosphorus stored in these macrophytes, they represent a potentially










important mechanism for returning sediment phosphorus to the lake when

they senesce and die.

Sediments. Surface sediment in the center of McCloud Lake is

highly organic with a large proportion of water (Table 3-8). The phos-

phorus content of this surficial sediment is about 0.34 mg/g dry

weight. If these sediment characteristics are typical where lake depth

is 8 feet or greater (the approximate extent of macrophyte beds), the

upper 1 cm represents a storage of approximately 7.74 Kg of phosphorus.

This is about 5 times the mean phosphorus storage in the lake water

(1.54 Kg P), and 2.7 to 3.5 times the annual total atmospheric phos-

phorus loading to the lake surface.


Phosphorus Uptake

Results of the phosphorus uptake experiments using the two major

groups of primary producers in McCloud Lake are summarized in Table

3-9. Even though mid-lake seston showed a first-order uptake rate con-

stant (k) nearly twice the magnitude of the k value for Websteria, the

variability in seston uptake rendered the two mean k values statistic-

ally indistinguishable (t-test, a < 0.05). The fact that rooted sub-

mergent macrophytes appear to obtain most of their phosphorus from the

sediments indicates that there is no competition for phosphorus between

the macrophytes and planktonic algae.


Summary

The pH of McCloud Lake has decreased from 4.85 to about 4.55 over

the past 15 years (nearly a doubling of H+ concentration), although

this has not been accompanied by a reduction of TP, chlorophyll-a, or

other nutrient species. The lake exhibits TP, chlorophyll-a, and











Table 3-9. Physical characteristics and phosphorus content of surfic-
ial mid-lake sediment from McCloud Lake.



Parameter Mean Standard Deviation


Water, % 93.03 0.18

Volatile Solids, % 77.7 1.00

Phosphorus, mg P/g dry wt 0.343 0.027

Phosphorus, % 0.034 0.003

Interstitial TP, mg/L 0.045 0.004










Table 3-10.


Phosphorus uptake rate constants (mean + standard devia-
tion) for submergent macrophytes and mid-lake seston from
McCloud Lake.


Uptake Rate
Component Constant (k), hr-I n


Mid-Lake Seston 0.486 + 0.364 4

Websteria sp. 0.257 + 0.064 3


t = 1.05; t.05,5 = 2.57.










phytoplankton and zooplankton communities that are typical of oligo-

trophic Florida lakes. Nitrogen-to-phosphorus ratios indicate that

primary production is limited by phosphorus. Total phosphorus shows

increased concentrations during late spring and summer, but SRP trends

are not evident.

Nutrient levels during 1980-1982 appear to be related to rainfall

patterns and variations in lake stage. This trend is consistent with

the finding (Baker 1984) that rainfall to the lake surface accounts for

90% of the total annual water input. Furthermore, atmospheric phos-

phorus deposition to McCloud Lake appears to approximate the minimum

phosphorus loading rate required to sustain mesotrophic conditions,

although the lake is oligotrophic. This suggests that McCloud Lake's

low pH inhibits water column productivity, or that phosphorus removal

mechanisms are faster than in the lakes used to develop nutrient load-

ing criteria.

Rooted submersed macrophytes constitute a significant in-lake

storage of phosphorus that is approximately 5.0 times the mean water

column phosphorus storage. It thus appears that the littoral zone

contributes much of the primary production in McCloud Lake, although

the macrophytes do not compete with phytoplankton for water column

phosphorus. Dense periphytic growths, which were common on the macro-

phytes, may minimize the role these macrophytes play in recycling sedi-

ment phosphorus to the water column. The following three chapters

discuss the results of experiments designed to test the effect of lake

pH on processes that contribute to removal or recycling of water column

phosphorus.


















CHAPTER 4
EFFECT OF PH ON PLANKTONIC PHOSPHORUS DYNAMICS



There is ample evidence to suggest that variations in aquatic pH

are accompanied by changes in phytoplankton and zooplankton communities

and by changes in phosphorus dynamics in the pelagic environment. How-

ever, it is difficult to know how these processes relate to each other.

Plankton community changes related to acidification could theoretically

affect phosphorus cycling, but conversely pH-related changes in phos-

phorus availability could also affect planktonic community structure.

Rates of phosphorus uptake by algae and bacteria generally increase as

cell size decreases (Lean and White 1983), and phosphorus regeneration

is faster for small zooplankters than for large species (Johannes 1965;

Hargrave and Geen 1968; Peters and Rigler 1973). Thus pH-related

trends in body size of phytoplankton and zooplankton communities should

alter phosphorus dynamics.

On the other hand, the activity of extracellular enzymes used by

algae in assimilating phosphorus is influenced strongly by pH. Thus

decreased pH could favor phytoplankton species which produce phosphat-

ase enzymes that are active at the new pH value. This scenario would

have the greatest potential impact on phytoplankton community structure

in phosphorus-limited lakes.


-IV










A series of experiments was designed to test the effect of pH man-

ipulation on phytoplankton and zooplankton community structure and on

sestonic rates of phosphorus uptake and turnover.



Materials and Methods


Mesocosm (Limno-Enclosure) Experiments

Littoral mesocosms. Three polyethylene enclosures were con-

structed according to Landers (1979) and installed in the littoral zone

(1 m depth) of McCloud Lake in February 1981. Each mesocosm enclosed a

12-m2 water column, and the polyethylene material was inserted into

the sediment and secured with wooden stakes and rope to ensure a good

seal. Before pH adjustment began, the mesocosms were allowed to

equilibrate for 4 weeks. The pH of enclosure A was decreased to 3.6

over a 4-week period with 1.0-L additions of 0.7 N H2S04, while the

pH of enclosure B was raised in the same manner to >5.1 with 1.0-L

additions of 0.1 to 0.4 N NaOH. Further acid and base additions were

made as necessary to maintain the desired pH ranges, although enclosure

B never reached the intended pH of 5.6 because of buffering by the

sediments (Baker 1984). Enclosure C was left at the ambient pH (4.6 +

0.1) throughout the experiment. These mesocosms were sampled on a

weekly basis to follow changes in the littoral chemistry and biology

resulting from pH alteration. In addition, a littoral lake station

adjacent to the enclosures was sampled on the same schedule. Chemical

and biological analyses were performed as described in Chapter 3.

Mid-lake mesocosms. Six enclosures were placed in the middle of

the lake in July 1982 to evaluate the effects of acidification and










nutrient addition on phosphorus dynamics and phytoplankton and zoo-

plankton communities. Two groups of three enclosures were installed 1

week apart. These mesocosms were 0.92 m in diameter, 2.2 m deep, and

each contained 1.2 m3 of depth-composited lake water (added with a

gasoline-powered pump). The pH treatment in these enclosures consisted

of duplicates at each of three values: M1 and M4 were left at the lake

pH of 4.7; M2 and M5 were lowered to 4.1 with 0.1 N H2S04; and M3

and M6 were first lowered to 4.1 and then further acidified to 3.7 1

week later. No additional pH adjustment was required since the enclo-

sures were isolated from the sediments. During the tenth week after pH

adjustment, NH4-NO3 and KH2PO4 were added to enclosures 1-3 to

increase TN and TP each by a factor of about 10. These enclosures and

a mid-lake station were sampled twice each month from the end of July

through November 1982 using the previously described methods for sample

collection and analysis (Chapter 3).

Radiolabeled orthophosphorus (32p) was used to measure ses-

tonic uptake and turnover of phosphorus in the mid-lake enclosures.

One-liter samples of seston (unfiltered water) from each enclosure were

transported to the laboratory and incubated at room temperature (22"C +

2C) in clear glass bottles. Fluorescent lighting was provided from

above, and magnetic stirrers slowly mixed the contents. Aliquots of a

stock solution containing K2HP04 as a carrier for 32p04 (obtained from

New England Nuclear) were added to the seston samples, and uptake of

added 32P was followed by periodically filtering 5-mL subsamples

(0.45 pm membrane filters). The volume of stock added in each experi-

ment was adjusted according to its specific activity, but the range was

25 UL to 1000 1L stock/L sample. The stock solution contained 3.1 pg











P/mL, and the activity introduced to each sample yielded approximately

12,000-60,000 counts per minute (CPM) in the 5-mL subsample. The

filter and filtrate were stored in separate scintillation vials until

they were counted on a Packard Tri-Carb model 4530 liquid scintillation

counter. Counts were corrected for background activity and decay so

that results corresponded to the time at which samples were collected.

Total SRP was calculated as SRP originally in each enclosure plus the

amount added in the laboratory.

A similar design was used to measure directly the release of phos-

phorus by seston from the mid-lake enclosures. A 2-L sample was col-

lected from each enclosure and transported immediately to the labora-

tory. The seston was concentrated by a factor of 10 by vacuum filter-

ing all but a small volume (120 mL) through 0.45 pm membrane filters.

This concentrated volume was transferred to an Erlenmeyer flask, and

seston retained on the filter was resuspended by vortexing three times

in 10 mL of filtered enclosure water. The resuspended seston was added

to the Erlenmeyer and additional filtered water was used to obtain a

final volume of 200 mL. Stock radiophosphorus was added and the flasks

were incubated under fluorescent lighting for 24 h to allow uniform

labeling of the seston. After the incubation period, the seston was

again concentrated by a combination of centrifugation and vacuum fil-

tration of the supernatant. The concentrated, labelled seston ("10 mL)

was added to unlabelled filtered enclosure water to obtain a volume of

500 mL. Phosphorus release was followed by periodically filtering 5-mL

aliquots (0.45 um membrane filters), and storing filter and filtrate in

separate scintillation vials for later analysis with the liquid scin-

tillation counter.










Several methods have been used for quantifying phosphorus uptake

and turnover. Many 3P users have adopted a method described by

Zilversmit et al. (1943) for calculating uptake rates and turnover

times in radioisotope experiments. The method necessitates three

assumptions:

1) Steady state conditions in which the rate of appearance of the

isotope equals its rate of disappearance;

2) Constant rate of appearance and disappearance; and

3) Random appearance and disappearance of the element and its iso-

tope.

They further define:

p = rate of disappearance of B from fluid;

x = amount of radioactive B in the fluid at any time;

r = total amount of B present in fluid (assumed to be constant);

and

tt = turnover time, the time required for the tissue to com-

pletely remove and replace r.

The change in x with time is given as


dx/dt = -p(x/r). (EQ 1)


Integrating EQ 1 yields


x/r = ce(-p/rt) (EQ 2)


and taking natural logarithms, EQ 2 becomes


In x/r = In c p/r't


(EQ 3)











Equation 3 describes a straight line with slope = -p/r which in turn

equals -1/tt. Therefore, from a plot of In x/r versus time and

knowledge of r, both uptake rate and turnover time can be calculated.

Zilversmit et al. (1943) cautioned that their method of uptake and

turnover calculations was valid only during the time interval in which

none of the radioisotope is returned from the tissue to the fluid.

Fast initial uptake of SRP (and thus quick turnover) coupled with the

difficulty of measuring accurately the initial SRP concentration led

Lean and White (1983) to recommend that first-order uptake rate con-

stants (k) should be used instead. The value of k (time-1) can be

obtained as the slope of a plot of percent isotope in the filtrate (or

the filter) versus time.

Diurnal productivity estimates were made in conjunction with both

mesocosm studies by following diel DO changes in the water columns of

the enclosures with a YSI model 57A dissolved oxygen meter. Oxygen

measurements in the littoral enclosures were made at 2- to 5-h inter-

vals over 24-h periods, with shorter intervals used in early morning

and early evening, when DO changes occurred most rapidly. Dissolved

oxygen was measured every 3 h over 24-h periods in the mid-lake enclo-

sures. Oxygen changes in the open water of both sets of enclosures

were corrected for diffusion across the air-water interface with a dif-

fusion coefficient experimentally determined for the lake using the

dome method of Copeland and Duffer (1964). Gross primary production

(P) and respiration (R) were determined by planimetry from plots of

corrected rate of DO change over time as described by Odum (1956) and

numerous others (Odum and Hoskin 1958; Hall and Moll 1975) for measure-

ment of community metabolism.











Laboratory Microcosms

An experiment similar to the mesocosms involved microcosms set up

in the laboratory in March 1983. Three 12-L glass carboys were filled

simultaneously by siphoning from a continuously stirred 40-L container

of depth-composited lake water. The microcosms were aerated slowly to

provide mixing, and overhead fluorescent lighting was timed to mimic

the natural day length. One microcosm was left at ambient pH (4.60),

and 1 N H2SO4 was used to achieve pH values of 4.0 and 3.6 in the

other two. SRP, TP, and total dissolved phosphorus (TDP) were measured

periodically using previously described methods. Because of an

accidental shift in TN/TP ratios, this experiment was terminated 4

weeks after pH adjustment.

Three new microcosms were set up in the same manner in May 1983.

Two weeks after pH adjustment, NH4-N03 and NaH2PO4 solutions

were added to increase TN and TP to 1.0 mg/L and 0.1 mg/L, respec-

tively. TP, TDP, and SRP were measured periodically before and after

nutrient addition. The activity of extracellular acid phosphatase

enzymes was measured in the microcosm experiments by a fluorometric

method developed by Swedish limnologists (Petterson and Jansson 1978;

Jansson et al. 1981; Jansson 1981). The procedure involves addition of

a buffered fluorogenic substrate (4-methylumbelliferyl phosphate, MUP)

to the water sample. Phosphomonoesterase activity is calculated from

the rate of liberation of fluorescent hydrolysis product 4-methylumbel-

liferone (MU). Fluorescence of MU was determined with an American

Instrument Co. SPF-125 spectrofluorometer using an excitation wave-

length of 320 nm and an emission wavelength of 450 nm. A stock solu-

tion of 10-2 M MUP was prepared in autoclaved distilled water and









frozen in 1-mL crimp-seal vials until needed. Working MUP solutions

(10-4 M) were obtained by dilution of the stock in 0.1 M acetate

buffer at each microcosm pH (4.6, 4.0, and 3.6). For the assay, 0.5 mL

of working MUP solution was added to 4 mL of the test water. Thus the

test could be run at the pH of each sample, or all microcosms could be

tested at pH 4.6. The amount of fluorescent MU released was quantified

by comparison to standard MU solutions prepared in acetate buffer.



Results and Discussion


Littoral Mesocosms

Figure 4-1 shows variations in soluble reactive phosphorus in the

littoral enclosures over 15 weeks (including 4 weeks of pH adjustment

but not the initial 4-week equilibration after installation). Figure

4-2 shows total phosphorus variations over the same period. Peaks in

SRP occurred in all three enclosures during initial pH adjustment,

although the increase was smallest in the control enclosure (C) left at

ambient pH. Thereafter, enclosure B (pH = 5.1) often showed the high-

est SRP concentration, although no consistent trend was evident. Total

phosphorus concentrations also were high initially, and then varied in

the range 1-10 pg/L after the first 4 weeks. With few exceptions, TP

was consistently higher in the high pH enclosure (B) than in the acidi-

fied one (A), after initial pH adjustment.

Table 4-1 summarizes mean nutrient concentrations in the littoral

enclosures and at the littoral lake station from late March to July

1981. TP shows a trend apparently related to pH, with the highest mean

in the base-treated enclosure (13 pg/L), the lowest in the acid treat-


I .


















25r


320




0)

' 15
0
a
0






w
05


oL '


26 MAR 81


A A (3.7)

* B (5.1)

O C (4.6)


I I I I t I I


3 5 7 9 II 13 15

WEEKS 1 JUL 81


Figure 4-1. Soluble reactive phosphorus variations in littoral meso-
cosms.























60
a A (3.7)

B (5.1)

50 0C (4.6)



-J

%5 40
0O
0
0 0
I
O

0. 3o


,-J
c 20


0
I-


20






0
I 3 5 7 9 II 13 j 15

28 MAR 81 24 JUNE 81
WEEKS


Figure 4-2. Total phosphorus variations in littoral mesocosms.










Table 4-1. Nutrient means (pg/L except TN/TP) in the littoral enclo-
sures.



Enclosure
Littoral
Parameter A B C Lake


SRP 4 4 8 3

TP 7 13 9 12

TN/TP (wt/wt) 54.6 31.0 36.0 45.5

TON 281 357 258 442

NH--N 49 27 29 50

NOj-N 52 27 37 54

NO-N 1 1 1 1









ment (7 pg/L), and 9 pg/L in the ambient pH enclosure. The means are

not significantly different (ANOVA, a > 0.05); however, analysis of the

average TP differences (calculated for each sampling date) between

pairs of enclosures (paired-difference t-test) shows that TP was

significantly higher in enclosures B and C than in the acidified enclo-

sure (TPB TPA, a = 0.0396; TPC TPA, a = 0.0138). The mean

difference between B and C is not significant (a > 0.05).

Mean TN/TP ratios in the littoral enclosures ranged from 31.0 in

the base treatment to 54.6 in the acid treatment, while in the littoral

lake the ratio was 45.5 (Table 4-1). These values all indicate phos-

phorus limitation (TN/TP > 30 according to Huber et al. 1982), and they

are similar to mean annual ratios for the mid-lake station (Table 3-1).

The trend of increasing TN/TP values as pH decreases suggests that

phosphorus becomes more limiting as pH is lowered. This reflects the

fact that TN variations in the littoral enclosures were insignificant,

while TP decreased with decreasing pH.

Although TN means did not vary significantly in these enclosures,

means of some of the nitrogen forms did suggest a pH effect. TON

levels did not appear to be related to pH, and the differences among

enclosure means were not significant (ANOVA, a > 0.05). However,

ammonium and nitrate means were slightly higher in the littoral lake

than in the enclosures. Both ions tended to increase as enclosure pH

decreased, which could indicate increased mineralization or decreased

utilization rates, although the differences between treatments were not

significant.

Among biological parameters measured in the littoral enclosures,

only zooplankton abundance varied significantly. Differences in the










mean values of log-transformed chlorophyll-a and total phytoplankton

abundance were not significant (ANOVA, a > 0.05), although total phyto-

plankton means did decrease with decreasing pH. Means of total zoo-

plankton and copepod abundances (log-transformed) also decreased with

pH. In both cases the differences between enclosures B and C were not

significant, while the values in enclosure A were significantly lower

(Duncan's Multiple Range Comparison Test, a = 0.05). The variations in

total zooplankton numbers were due to the reduction in copepods as pH

decreased, as evidenced by the fact that cladoceran and rotifer abund-

ances did not show a treatment effect.


Mid-Lake Mesocosms

Nutrient trends. Figure 4-3 shows TP variations in the unfertil-

ized mid-lake enclosures. TP stayed relatively constant in M1 (con-

trol) while it increased in acidified enclosures M2 and M3. In the

other set of enclosures however, TP was consistently lower in M6 (pH =

3.7) than in the higher pH enclosures M4 and M5. TP trends were not

similar in the control enclosures (Ml and M4). Table 4-2 summarizes

the mean nutrient concentrations in the mid-lake enclosures by pH

treatment. Total phosphorus was lowest at pH 3.7 (6 ug/L), and was

nearly the same at 4.1 (10 pg/L) and 4.7 (9 pg/L). According to

Duncan's Multiple Range Test (a = 0.05) the two higher TP means did not

differ significantly, while both were statistically higher than TP at

the low pH of 3.7. SRP concentrations were consistently low in these

enclosures and mean SRP was 2 yg/L for all three pH treatments.

TN/TP ratios in the mid-lake enclosures (Table 4-2) were lower

than in the littoral enclosures or the lake, and they were indicative

































M1 (4.6)
0 M2 (4.1)
A M3 (3.7)
3


OCT NOV


M4 (4.6)
0 M5 (4.1)
A Me (3.7)
0-


AUG



Total phosphorus in m
of nutrient addition.


SEP


OCT


NOV


lid-lake mesocosms, excluding period


AUG


SEP


Figure 4-3.











Table 4-2.


Nutrient means (pg/L except TN/TP) for mid-lake enclosure
pH groups (excluding data from M1 to M3 after nutrient
addition.


pH Group

Parameter 4.6 4.1 3.7


SRP 2 2 2

TP 9 10 6

TN/TP (wt/wt) 15.4 11.5 17.3

NH -N 11 11 13

NO-N + NOj-N 12 7 8

TN 139 115 104









of nutrient-balanced conditions according to criteria proposed for

Florida lakes (10 < TN/TP < 30; Huber et al. 1982). While the lowest

pH enclosures did show the highest TN/TP ratio (17.3), the inverse

relationship found between TN/TP and pH in the littoral enclosures was

not seen in the mid-lake mesocosms.

Total nitrogen in the unfertilized mid-lake enclosures was highest

at the ambient pH (139 pg/L) and decreased with enclosure pH. ANOVA

showed a significant treatment (pH) effect (a < 0.05), and Duncan's

Multiple Range Test yielded the following relationship among mean TN

values:


pH 4.7 pH 4.1 pH 3.7



(TN means at underlined pH values are not significantly different, =

0.05). Means of ammonium and nitrate/nitrite were all less than 15

pg/L, and no pH-related trends were evident for these nitrogen forms.

Figure 4-4 illustrates the effect of nutrient addition on the

phosphorus fractions in mid-lake enclosures Ml, M2, and M3. SRP

removal rates were similar in all three pH treatments, and SRP concen-

trations decreased to pre-fertilization levels within 13 days after

nutrient addition. Particulate organic phosphorus (POP) also responded

in a similar manner at the three pH values. POP reached maximum levels

(- 30 ug/L) about 2 weeks after nutrient addition, and had returned to

pre-fertilization concentrations within 33 days. The response of solu-

ble organic phosphorus (SOP) did indicate a pH effect, however. SOP in

M1 (pH 4.6) and M2 (pH 4.1) stayed below 10 pg/L after nutrient addi-

tion, but at the lowest pH (3.7), SOP increased to nearly 30 ig/L.







74








50

40
-j
S30-









0 10 20 30 40


50-

40-

5
-0

30-


C 20




ioo

o go 0'o 30 40


100
M1 (4.7)

80 0 M2 (4.1)
6 M3 (3.6)
-cj
o 60

0 40-
Co
20-

O I I I I
0 10 20 30 40

DAYS

Figure 4-4. Changes in phosphorus forms after nutrient addition to Ml,
M2, and M3.










These results suggest a reduced ability of the biota to utilize SOP at

a pH of 3.7.

Biological trends. Total phytoplankton densities initially

decreased in all six mid-lake enclosures, but in the intermediate pH

(4.1) enclosures, phytoplankton then began to increase relative to the

other enclosures. After nutrient addition to Ml, M2, and M3, phyto-

plankton in both pH 4.1 enclosures (M2 and M5) declined to levels com-

parable to those in the other enclosures. The addition of nutrients

was not followed by an increase in phytoplankton numbers in Ml, M2, and

M3, although chlorophyll levels did reflect the increased nutrient

concentrations. This phenomenon was apparently the result of intensi-

fied zooplankton grazing in the fertilized mesocosms, which kept phyto-

plankton numbers low while production was high. Chlorophyll-a concen-

trations did increase because the analysis included material extracted

from phytoplankters and zooplankters. Total zooplankton populations

also increased markedly during the same period in the fertilized

enclosures.

Species composition of phytoplankton and zooplankton communities

responded to increased acidity in the mid-lake enclosures. Ultraplank-

ton (1-10 pm) comprised 80-100% of total phytoplankton numbers at the

lake pH (4.6), but decreased in importance at lower pH values. The

green alga Oocystis gloeocystiformis increased in abundance at reduced

pH in the unfertilized enclosures, while after fertilization, Crypto-

monas marsonii dominated at all three pH values.

Zooplankton response to reduced pH was similar to that seen in the

littoral enclosures. Copepod numbers (predominantly Diaptomus missis-

sippiensis) decreased markedly as pH was lowered, so that this group









accounted for less than 1% of total zooplankton in the pH 3.7 mesocosms

shortly after initial pH adjustment. The acid-tolerant cladoceran

genera Eubosmina and Diaphanosoma became increasingly important at more

acidic pH values. Rotifer percent composition in the low pH enclosures

increased shortly after pH reduction, but after about 6 more weeks,

their importance decreased to levels comparable to the higher pH

enclosures. In contrast to phytoplankton composition, zooplankton

species composition was not affected by nutrient addition.

Radiophosphorus experiments. Phosphorus uptake by seston in the

mid-lake enclosures was measured on six dates using radiolabelled phos-

phorus in the laboratory. In Figures 4-5 through 4-10, uptake of

32P is represented as the increase in radioactivity on filters

versus time during the incubations. (Note variations in abscissas and

ordinates on different dates.) The first two dates (Figures 4-5 and

4-6) were before nutrient addition to Ml-M3, while the last four were

after the addition. Missing data for Ml, M4, and M5 in Figures 4-9 and

4-10 (the last two dates) were because bird droppings had caused

increases in pH and TP in those enclosures. Mesocosm M1 (pH 4.6)

showed the fastest uptake of 32P before nutrient addition, but M6

(pH 3.7) consistently exhibited the fastest uptake after the addition.

Table 4-3 gives uptake rates and turnover times calculated accord-

ing to Zilversmit et al. (1943) from the data presented in Figures 4-5

through 4-10. Uptake rates ranged from 0.1 to 12 ig/L'h, while the

range of turnover times was 0.3 h to 12.2 h. It is interesting to note

that the unfertilized mesocosms with the lowest TP values (Ml before

nutrient addition, and M6) showed the fastest uptake rates.



























20





o" 15-


I_ 5









5- 4



3
o I 3
0-








0 120 240

TIME (min)


32P uptake by mid-lake enclosure seston, 8/24/82.


Figure 4-5.


















60



6

3
50






40 3


0 5


a. 30
0

I-
c-
I5
20




44


20 -
4



10- 4






0 120 240

TIME (min)


32p uptake by mid-lake enclosure seston, 9/14/82.


Figure 4-6.































6






6 5
6




6



2

5
3

2 21



3 4
1 4
06. | -4
0 60 120

TIME (min)


Figure 4-7. 32p uptake by mid-lake enclosure seston, 11/9/82.





























30F


z6


20-


f-44 4
4


TIME (min)


32p uptake by mid-lake enclosure seston, 11/11/82.


Figure 4-8.







81













20-

6 6




15 /6
S6

x

I0
o 10

w /
I-
-J
5

5
33



5 2
0 2 2-- 2 ----- 2
0 60 120

TIME (min)


32P uptake by mid-lake enclosure seston, 11/18/82.


Figure 4-9.





























































Figure 4-10.


8




6
6





4- 2




3

2






0 60 12

TIME (min)
32p uptake by mid-lake enclosure seston, 11/23/82.











Table 4-3.


Phosphorus uptake rates and turnover times by seston from
mid-lake enclosures (underlined values indicate enclosures
with nutrient addition).


Date Var. M1 M2 M3 M4 M5 M6


08/24 p 12 1.8 0.6 2.4 4.2 2.4
tt 0.4 3.9 10.0 3.3 1.2 1.7

09/14 p 6 3.0 3.6 1.2 2.4 4.2
tt 0.5 1.0 0.8 1.7 1.2 0.5

11/09 p 0.6 4.8 1.8 0.2 2.4 4.2
tt 6.2 1.2 2.0 9.4 0.7 0.6

11/11 p 0.4 0.2 1.2 0.1 1.2 1.2
tt 9.7 9.2 5.8 28.0 3.3 2.7

11/18 p -- 0.2 0.6 0.6 1.8 11.0
tt -- 12.2 11.6 2.8 2.2 0.3

11/23 p -- 0.5 0.4 -- 6
tt -- 7.4 5.6 -- 0.7


p = Uptake rate (Ug/L'h).
tt = Turnover time (h).











The same procedure was used to calculate phosphorus release rates

from the seston concentration experiments, which were carried out on

two dates before nutrient addition to enclosures Ml, M2, and M3. These

release rates, corrected to reflect the original seston concentrations,

are given in Table 4-4. No effect of enclosure pH is evident in these

data, although the range of values is comparable to the uptake rates in

Table 4-3.

Table 4-5 lists the means and ranges of uptake rate constants cal-

culated by the method of Lean and White (1983) from the same experi-

mental data presented in figures and tables above. Results using this

method were similar to those of the previous method. In the group con-

sisting of M4, M5, and M6 there was a clear trend of increasing rate

constants as pH decreased. However, this was not repeated in the pre-

fertilization data for Ml, M2, and M3, where the high pH enclosure (Ml)

showed the largest rate constants. ANOVA revealed no significant

effect of pH on k values (a > 0.05). It thus appears that phosphorus

availability has more effect on the planktonic uptake and turnover of

phosphorus than does any direct effect of hydrogen ion concentration.

Furthermore, phosphorus availability (as indicated by SRP concentra-

tions) did not appear to be related to pH in these mesocosms.


Community Metabolism

The diel oxygen technique was used to obtain two estimates of com-

munity metabolism in the littoral enclosures, while the mid-lake enclo-

sures were monitored on three dates. Table 4-5 presents means of these

productivity (P) and respiration (R) measurements for the littoral and











Table 4-4. Phosphorus release rates (pg/L'h) by seston from mid-lake
enclosures.



Enclosure

Date M1 M2 M3 M4 M5 M6


08/27 2.0 10.2 0.5 0.7 0.6 1.7

09/16 2.1 0.7 3.2 -- 2.1 0.5











Table 4-5. Phosphorus uptake rate constants (h-1) for the mid-lake
enclosures.



Enclosure pH Mean k Range


Ml (Pre) 4.6 0.74 0.52-0.96
M1 (Post) 4.6 0.11 0.09-0.13

M2 (Pre) 4.1 0.25 0.14-0.36-
M2 (Post) 4.1 0.25 0.13-0.50

M3 (Pre) 3.7 0.32 0.02-0.61
M3 (Post) 3.7 0.28 0.19-0.37

M4 4.6 0.14 0.02-0.32

M5 4.1 0.40 0.01-1.30

M6 3.6 1.12 0.17-1.99










mid-lake mesocosms, and a mid-lake station adjacent to the open-water

enclosures. The data are expressed on a volumetric basis (g 02/m3.day)

to allow comparison of metabolism in communities of different depth

(Lind and Campbell 1970).

P and R means were similar in any given enclosure. Mean produc-

tivity in the mid-lake mesocosms ranged from 0.70 to 0.84 g 02/m3.day,

while mean respiration varied from 0.75 to 1.01 g 02/m3.day.

Average P/R ratios were close to unity in these communities and at the

mid-lake station. Two-way ANOVA and Duncan's Multiple Range Comparison

showed no significant pH effect on P, R, or P/R ratios for the

enclosures, and no significant differences in these parameters between

the open lake and any enclosure.

Total community P and R were much higher in the littoral enclo-

sures than in the open lake or the mid-lake enclosures. Mean P ranged

from 3.8 to 6.1 g 02/m3.day and the range in R was 3.8 to 7.1 g 02/m3.day.

The difference between littoral and open lake communities probably

reflects the contribution of submersed macrophytes and periphytic algae

to littoral productivity. It is interesting to note that the higher

littoral primary productivity was balanced by a comparable community

respiration, so that P/R ratios were near 1.0 for littoral and plank-

tonic communities. A P/R value near unity is considered indicative of

a balanced ecosystem (Lind and Campbell 1970). Two-way ANOVA of the

littoral metabolism data showed no significant pH effect ( a> 0.10 in

all cases) on productivity, respiration, or P/R values.

Although areal or volumetric rates of primary productivity are

higher in the littoral zone of McCloud Lake than in its pelagic waters,

the area represented by both habitats would have to be determined to










Table 4-6. Volumetric productivity
in littoral and pelagic


(P), respiration (R), and P/R means
mesocosms.


Community pH n P* R* P/R


Open-water M1 4.6 3 0.70 0.75 1.05
Open-water M4 4.6 3 0.82 0.92 1.01
Mid-lake 4.6 3 0.83 1.01 0..87
Littoral C 4.6 2 6.07 7.12 0.86

Littoral B 5.6 2 4.98 5.36 0.93

Open-water M2 4.1 3 0.84 0.88 1.12
Open-water M5 4.1 3 0.76 0.94 0.83

Open-water M3 3.7 3 0.83 0.88 1.12
Open-water M6 3.7 3 0.77 1.01 0.77
Littoral A 3.7 2 3.82 3.83 1.02


*g 02/m3.day.











evaluate the overall importance of each zone to total lake productiv-

ity. Variations in primary productivity are matched by similar changes

in respiration, so that P/R ratios remain close to unity. In addition,

variation of pH between 5.6 and 3.6 did not significantly affect the

metabolism of littoral or planktonic communities originally adapted to

a pH of 4.6.


Laboratory Microcosms

Total phosphorus concentrations in the unfertilized microcosms

were too low (generally 2-4 pg/L) to permit determination of a pH

effect. Initial TP levels were higher ('6 pg/L), but filamentous algae

growing on the glass microcosm walls soon reduced the phosphorus avail-

able to planktonic biomass. SRP uptake after nutrient addition to the

second set of microcosms is shown in Figure 4-11. Removal of SRP from

the water columns appeared to follow first order kinetics for 5-6 days,

but it was essentially linear from that point through day 15. This

could be accounted for by a 5-6-day lag in the increase of zooplankton

numbers, after which zooplankton grazing would encourage a constant

rate of algal productivity.

There was no indication that microcosm pH influenced the rate of

SRP uptake, which was essentially identical at the ambient pH (4.6) and

the low pH (3.7), and was only slightly slower at the intermediate pH.

In addition, the shapes of SRP loss curves were the same at all three

pH values.

After fertilization, these microcosms did not exhibit the SOP

increase at pH 3.7 that was seen in the mid-lake mesocosms. The large

surface-to-volume ratio in the microcosms increased the importance of


















120-





100-





80-


* 4.6
A 4.0
0 3.6


DAYS


Figure 4-11. SRP in microcosms after nutrient addition.











attached algae, so that the response of water column TP to fertiliza-

tion was much less at all pH values than in the mid-lake enclosures.

Despite the inherent differences between the microcosm and enclosure

experiments, microcosm fertilization did indicate that the SOP

increase seen in the pH 3.7 enclosure was not a universal effect of low

pH. Both experiments also showed little or no effect of pH on SRP

uptake.

Acid phosphatase activity. Table 4-7 presents typical phosphatase

assay results from the first set of microcosms. Activity decreased as

assay pH decreased, regardless of initial pH of the sample, and the

potential activity (all assayed at pH 4.7) decreased as microcosm pH

decreased. Activity in the two low pH microcosms was 80-150% higher

when assayed at pH 4.7 than at the ambient microcosm pH. This indi-

cates that the enzymes were adapted to the ambient pH of the lake, and

also suggests that production of the extracellular enzymes decreased at

the lower pH values. While pH definitely affected the potential activ-

ity of the enzymes, many factors can influence production of phosphat-

ase enzymes.

Potential phosphatase activity in the second set of microcosms

showed the same pH effect (Table 4-7). Samples from the acidified

microcosms showed higher enzyme activity when assayed at pH 4.7 than at

ambient pH. However, enzyme production did not follow the same pat-

tern. When samples from all three microcosms were assayed at pH 4.7

the activity of the pH 3.7 microcosm was frequently equal to or higher

than that of the pH 4.7 microcosm, and activity of the intermediate pH

microcosm was higher than in either of the other two. These results

indicate that some factor (or factors) other than pH affects the pro-

duction of acid phosphomonoesterase enzymes.










Table 4-7.


Effect of pH on phosphatase enzyme activity (n mole/L'min)
in microcosm experiments.


FIRST EXPERIMENT SECOND EXPERIMENT

Microcosm pH Microcosm pH
Assay
pH 4.7 4.0 3.7 4.7 4.0 3.7


4.7 30.2 16.1 9.1 14.5 24.1 15.1

4.0 6.1 9.3

3.7 5.1 7.8

















CHAPTER 5
EFFECT OF PH ON PHOSPHORUS RELEASE DURING PLANT DECOMPOSITION



Grahn et al. (1974) hypothesized that acidification of fresh

waters causes reduced rates of organic matter decomposition and thus

slower rates of nutrient remineralization. Experiments to test this

hypothesis have taken several forms. Most researchers have examined

the effect of pH on loss of particulate or soluble organic substrate,

and little attention has been focused on the release of nutrients dur-

ing decomposition of naturally occurring particulate organic matter.

This chapter presents experiments designed to follow release of soluble

reactive phosphorus from plant matter decomposing at different experi-

mental pH values. Because the submergent macrophytes Websteria sp. and

Eleocharis sp. represent a large reservoir of phosphorus within the

lake, they were used as the organic substrate in these experiments.



Experimental Methods


Live, freshly collected Eleocharis or Websteria plants were used

in all experiments in order to simulate as closely as possible the con-

ditions of decomposition in McCloud Lake. Intact plants were collected

from the littoral zone, brought to the laboratory, and gently washed to

remove attached algae from the leaves and organic sediment from the

roots. The plants were blotted dry with paper towels to obtain fresh




Full Text

PAGE 1

PHOSPHORUS DYNAMICS IN AN ACIDIC, SOFT-WATER FLORIDA LAKE by REUBEN WALTER OGBURN. III A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1984

PAGE 2

ACKNOWLEDGMENTS I would like to thank the many individuals who have contributed to this effort through their technical advice and assistance as well as those who provided moral and financial support. My research was funded by a grant from the U.S. Environmental Protection Agency-NCSU Acid Pre cipitation Program to Drs. Patrick Brezonik and Tom Crisman, and its second year of funding was administered by Dr. Bob Volk in the Soil Science Department. Dr. Joseph Delfino and Dr. Brezonik have been my principal advisors; their encouragement and direction have been greatly appreciated. Work at McCloud Lake was a cooperative effort between the chem istry and biology groups of the Environmental Engineering Sciences Department. Drs. Jeff Foran and Tom Crisman coordinated the biological data collection in the lake during 198~1981 and in the littoral meso cosms. Robert Garren and Chan Clarkson assisted in surveying the aquatic macrophytes in McCloud Lake. Larry Baker contributed to the fieldwork and chemical analysis during the same time period. Ray Bienert provided the biological data and interpretation rem the lake and the mid-lake enclosures, as well as assistance and encouragement in the field. Dr. Michael Binford, a postdoctoral associate at the Flor ida State Museum, contributed data, field assistance, and valuable dis cussions related to sediment processes in McCloud Lake. Eric Edgerton ii

PAGE 3

collected the hydrological data from McCloud Lake. These contributions were essential to the completion of my research. My parents have encouraged and supported me throughout my graduate studies, and their early guidance and direction were instrumental in my decision to pursue a career in science. My wife Marlyn and our child ren deserve equal credit for this effort. Marlyn supported my de~ision to leave my job, move to Gainesville, and return to graduate school with two small children She made tremendous sacrifices of time and energy to be a working mother and housewife at a time when our contemp oraries were pursing traditional American family lifestyles. Our sons, Walt and Doug, have missed a great deal in the way of father-son rela tionships. Without the patience, understanding, and endurance of my family, this effort would not have been possible. iii

PAGE 4

TABLE OF CONTENTS ACKNOWLEDGMENTS .. .. .. . ....... ...... ......... ................... ii ABSTRACT ..... .. .. ....... .. .... .................... ... ....... vi CHAPTER 1-INTRODUCTION ................ .... .. ........... .... ...... 1 Bae kg round ............. ........... ... ..................... . 1 Objectives ...................... .. .......... ................. 4 Si t e Des c rip t ion . . . . . . . . . . . . . . . . . . . . . . ... 5 CHAPTER 2--LITERATURE REVIEW ................. .................... .. 8 Sediment-Water Exchanges ... .... .............. .............. . 8 Decomposition .................. ..... ...... ............ ..... 13 Planktonic Phosphorus Cycling ... . ...... ... ... ............. 15 Lake Phosphorus Models . ... ..... . ... ..... .... .............. 19 Effects of pH on Phosphorus Cy c lin g ... .... ............ .. .. .. 21 CHAPTER 3---FHOSPHORUS DYNAMICS IN MCCLOUD LAKE ...................... 24 Materials and Methods ............... ... ........ ..... ....... 24 Routine Sampling ....... ................. .............. 24 Phosphorus Budget ............. .... ......... .... ... .. ... 25 Water budget .......... ....... ......... ............ 25 Atmospheric phosphorus loading .. .. .. ....... ........ 27 Sedimentation ........ ........... .... ........... .. 27 McCloud Lake Phosphorus Compartments . ...... .... .. ..... 30 Mac rophyte survey ....... ........ .. ... ........... .. 30 Sediment phosphorus ............................... .. 31 Phosphorus uptake . ...... .. .......... .... ... ...... 31 Results and Discussion .... .. ........ ......... . ... ..... .. 32 Limnology and Historical Nutrient Trends . .......... ..... 32 McCloud Lake Hydrology ...... ............. .. . ..... . ... 40 McCloud Lake Phosphorus Budget ..... ... .. ... .. .......... 42 Pre e ipi tat ion ........ ................ ... .. ... ..... 42 Mass balance .............. .. ........... ............. 46 Sedimentation ... ........... ..... ....................... 46 Phosphorus Compartments .......................... ........ 50 Ma c rophyt e s ....... . .. .......................... .... 50 Sediments .......................... .. ...... .... .. 54 Phosphorus Uptake .......... ......... .. .. .... .......... 54 Summar y ....................... .......................... 54 CHAPTER 4--EFFECT OF PH ON PLANKTONIC PHOSPHORUS DYN A MICS .. .. .... .. 58

PAGE 5

Materials and Methods .......................................... 59 Mesocosm (Limno-Enclosure) Experiments .................... 59 Littoral mesocosms ................................... 59 Mid-lake mesocosms ................................... 59 Laboratory Microcosms ..................................... 64 Results and Discussion ......................................... 65 Littoral Mesocosms ........................................ 65 Mid-Lake Mesocosms ........................................ 70 Nutrient trends ...................................... 70 Biological trends .................................... 75 Radiophosphorus experiments .......................... 76 CoTmllunity Metabolism ...................................... 84 Laboratory Microcosms ... ................................. 89 Acid phosphatase activity ............................ 91 CHAPTER 5--EFFECT OF PH ON PHOSPHORUS RELEASE DURING DECOMPOSITION .. 93 Experimental Methods .................................... .. ..... 93 Preliminary Experiments ................................... 94 Final Experimental Design ................................. 97 Results and Discussion ......................................... 98 Preliminary Experiments ................................... 98 Final Experiment ......................................... 105 CHAPTER 6---SEDIMENT-WATER INTERACTIONS . ........................... 113 Int rod uc t ion ...................... ........................... 113 Background .................................................... 113 Met hods ...................... .. ............................... 115 Sediment Characterization ................................ 115 Batch Adsorption/Desorption Experiments ........ ......... 116 Undisturbed Core Experiments ............................. 117 Results and Discussion ........................................ 119 McCloud Sediment Characteristics ......................... 119 Batch Adsorption/Desorption Experiments ... .. ............. 122 Sediment Core Experiments ................................ 134 CHAPTER 7-SUMMARY AND CONCLUSIONS .... ............................ 141 LITERATURE CITED ................. ................... .. ............ 144 BIOGRAPHICAL SKETCH ................................................ 152 V

PAGE 6

Abstract of Dissertation Presented to the Graduate Council of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy PHOSPHORUS DYNAMICS IN AN ACIDIC, SOFT-WATER FLORIDA LAKE by REUBEN WALTER OGBURN, III April 1984 Chairman: Joseph Delfino Cochairman: Patrick L. Brezonik Major Department: Environmental Engineering Sciences Laboratory and in situ experiments as well as historical data were used to characterize phosphorus dynamics in acidic, soft-water McCloud Lake, Florida, and to evaluate the effect of acidification on phos phorus cycling processes. McCloud presently exhibits nutrient and chlorophyll-~ concentrations typical of oligotrophic Florida lakes. A IS-year pH decline (4.85 to 4.55) has not been accompanied by significant changes in TP, chlorophyll-~, or nitrogen to phosphorus ratios, which indicate phosphorus-limited primary production. Total phosphorus shows maxima during late spring and summer, and variations appeared to be related to rainfall patterns and lake levels during 1980--1982. Atmospheric phosphorus deposition is near the loading rate necessary to maintain mesotrophic conditions, which suggests that low pH may contribute to the low TP in McCloud Lake. Rooted submergent macrophytes represent an in-lake storage of

PAGE 7

phosphorus that is approximately 2.5 times the average water column phosphorus storage, although the macrophytes do not appear to compete with phytoplankton for SRP .!.!!_ situ littoral and open-water mesocosm data indicated that acid ification (from 4.6 to 3.7) does lead to reduced water column TP levels, although the trends were not consistent in the open-water enclosures, which were not connected to the sediments. No relation was seen between pH and rates of phosphorus uptake by planktonic communi ties, and pH did not affect the activity of extracellular acid phos phatase enzymes 1n laboratory microcosms. The amount of SRP released from decomposing submersed macrophytes was independent of pH (over the range 3. 7 to 5.5) after 227 days of aerobic dark incubation, although initial rates of release were some what faster at the lowest pH. These experiments showed that acidifica tion does not inhibit phosphorus release during decomposition of aquatic plants. Effects of pH on surface charge characteristics and SRP specia tion cause sediment adsorption of SRP to vary with pH. Maximum SRP adsorption occurred near pH 4. 7, although there was little variation between pH 5.0 and 3.5 However, significant decrease in SRP adsorp tion at pH> 5 indicates that this mechanism may contribute to the observed trend of low TP in acidic lakes. This effect would be great est over the pH range 7.0 to 5.0, and further acidification of lakes near the pH of Mccloud Lake would have little effect on SRP adsorption. vii

PAGE 8

---------------------CHAPTER 1 INTRODUCTION Background Acidic precipitation 1s considered to have pH< 5.6, which 1s the pH of pure water in equilibrium with atmospheric CO 2 (Likens et al. 1979). Sulfur and nitrogen oxides from anthropogenic emissions ( and from natural sources to a smaller extent) react with water vapor in the atmosphere to form sulfuric acid and nitric acid. The return of these acids to the earth with rainfall is known as acid precipitation, or acid rain. However, dryfall of particulates and gaseous deposition can also contribute significant amounts of acid to the earth's surface. Therefore, acid precipitation generally refers to rainfall acidity, while acid deposition includes wetfall, gaseous, and dryfall acidity. Acid precipitation was described in England as early as 1852 and was linked to changes in water chemistry by Gorham in the 19SO's, although it was not recognized as a widespread and serious threat to aquatic and terrestrial ecosystems until the 1960's and 1970's (Cowling 1982). The chemical and biological changes associated with acidifica tion of Scandinavian lakes and streams generated intense political and scientific interest in determining the sources, extent and effects of atmospheric acidity. In North America, acid precipitation has been documented in the northeastern United States and southeastern Canada, in the southeastern U.S. (including Florida), and in the Rocky 1

PAGE 9

2 Mountains. Effects of acid deposition on aquatic and terrestrial ecosystems are difficult to demonstrate conclusively, and the problem of differentiating between long-term trends in natural processes and short-term changes caused by relatively recent increases in atmospheric acidity remains a controversial issue. Acid deposition has been implicated in accelerated erosion of buildings, human health problems including cancer, forest decline, decreased agricultural yields, and acidification of poorly buffered surface waters. Perhaps the most dramatic effect of aquatic acidifica tion has been the elimination of trout populations fr001 some temperate lakes and streams. Other aquatic effects have been inferred from sur veys of lakes over a range of pH values. Regional studies have shown similarities in the chemistry and biology of acidic lakes from different geographic areas. Phytoplankton and zooplankton assemblages tend to become more simplified with decreasing lake pH, and similar groups of species are found in acidic lakes of Scandinavia and temperate North America (Sprules 1975; NRCC 1981; Confer et al. 1983). Grahn et al. (1974) found that acidic lakes in Scandinavia showed greater transparency and lower chlorophyll-!! and macrophyte abundance than non-acidic lakes. They hypothesized that acidification causes an "ol igotrophication" process in which reduced rates of organic matter decomposition and nutrient recycling lead to lower rates of primary production. Surveys in Canada and the north eastern U.S have shown similar trends in transparency, chlorophyll-~, and macrophyte abundance (Dillon et al. 1978; Hendrey et al. 1976). However, other studies have shown trends in phytoplankton production and nutrient concentrations which were not consistent with the

PAGE 10

---oligotrophication theory (Dillon et al. 1979; Hendry and Brezonik 1984). Although acid deposition has been studied less intensively 1n the southeastern U.S., some trends in acid deposition and its aquatic effects have been demonstrated for Florida. Brezonik et al. (1980) found that the northern two-thirds of Florida receives a mean annual rainfall pH of 4.7 or less and excess sulfate deposition (non-marine origin, based on S04:Cl ratios) around 20 kg/hayr. Furthermore, Florida has approximately 2500 lakes that are sensitive to acidifica tion based on the criterion of alkalinity < 100 eq/L (Hendry and Brez onik 1984). In a study of 20 soft-water Florida lakes over the pH range 4 6-6. 7, Brezonik et al. (1984) found strong correlations of phytoplankton, chlorophyll-a, and total phosphorus with pH. Data from 165 Florida lakes were analyzed by Canfield (1981), who concluded that the relation between pH and chlorophyll-.!!_ was due to a strong correla tion of TP with pH rather than to other factors re l ated to pH. How ever, his data base included many hard-water and eutrophic lakes, which would tend to mask a pH-TP or pH-chlorophyll-.!!_ correlation in soft water lakes. These survey results have left it unclear whether the low TP con centrations (and thus low chlorophyll-a and phytoplankton levels) in acidic lakes are a consequence of low pH (as suggested by Grahn et al. 1974) or whether they reflect the conditions that originally make the lakes susceptible to acidification. Lakes with small watersheds receive a large proportion of their water and phosphorus inputs from rainfall directly to the lake surface. There is thus little opportun ity for watershed buffering, and phosphorus loading rates to such lakes are low. 3

PAGE 11

Phosphorus is a maJor nutrient requirement of primary producers in aquatic and terrestrial habitats. In lakes phosphorus availability is the factor which most often limits phytoplankton production. Increased cultural input of phosphorus was recognized as the primary cause for the eutrophication of many lakes during the 1960's and 1970's. The key role of phosphorus in the eutrophication process led to much research on ways to control or reduce phosphorus levels in lakes. Although this research necessarily included the processes involved in phosphorus cyc ling, the primary emphasis was on productive lakes. Some phosphorus cycling studies have considered temperate oligotrophic lakes, but few have included unproductive subtropical lakes. Lake acidification seems to have the opposite effect of eutrophi cation, but the role of pH in determining lake productivity remains a controversial issue. As mentioned earlier, lake surveys frcxn different geographic areas have found decreasing TP concentrations as [H+] increases. On the other hand, although acidic lakes generally tend to be unproductive, some evidence indicates that acid lakes are no less productive than similar, oligotrophic lakes with circumneutral pH values (Dillon et al. 1979). Objectives This dissertation addresses the hypothesis that acidification of Florida lakes can directly or indirectly affect their total phosphorus concentrations. The approach taken to answer this question involved a combination of laboratory and field studies designed to accomplish the following major objectives: 4

PAGE 12

1. to characterize phosphorus cycling 1n an acidic, subtropical lake, and 2. to assess the effect of [H+] (acidification) on the major processes involved in phosphorus cycling. These include plank tonic uptake and turnover of phosphorus, exchange reactions between lake water and sediments, and the release of phosphorus from decomposing organic matter. Site Description Mccloud Lake is a small, soft-water lake located 1n the Trail Ridge area of north-central Florida, about 40 Km east of Gainesville 1n Putnam County (Figure 1-1). Past studies of the lake include its use as a control 1n a whole-lake nutrient enrichment experiment on nearby Anderson-Cue Lake in 1966-1969 (Brezonik et al. 1969) and quarterly sampling during a 1978-1979 survey of the chemistry and biology of 20 Florida lakes (Hendry and Brezonik 1984) The region is characterized by sparse vegetation and numerous small lake basins perched among sandy hills. Longleaf pine/turkey oak assemblages provide a broken canopy, while lichens and wiregrasses dominate the understory and open areas. The McCloud Lake watershed (1 Km 2 ) is uninhabitated and is part of a controlled-access area, the University of Florida Katharine Ordway Ecological Preserve. Surface soils consist of non-spodic marine sands (Candler typic quartzipsannnent, cation exchange capacity ~2.S meq/100 g) that allow very little overland runoff to the lake. Other components of the unconfined (non-artesian) surface aquifer include gravels and sandy 5

PAGE 13

-------------------LITTORAL t 0 25 50 100m N Figure 1-1. Bathymetric map of Mccloud Lake, September 1982 (contour interval= 2 ft) 6

PAGE 14

clays of the Citronelle Formation. The sandy clays, clays, and phos phatic sands of the Hawthorne Formation constitute a relatively imperm eable confining layer (24-30 m thick) which separates the limestone of the artesian Floridan Aquifer from the perched shallow water table. Since there are no surface inflows or outflows to McCloud Lake, the only sources of water are rainfall directly to the lake and subsurface seepage. The hydraulic residence time of the lake is about 9.6 years (Baker 1984). McCloud Lake occupies a sub-rectangular solution basin (Figure 1-1) which had a surface area of about 5 ha and a maximum depth of 5 m during 1980-1982. Water level and surface area vary widely in response to long-term rainfall patterns. In 1966-1967, the maximum depth was nearly 6.5 m and the surface area was about 9 ha, while in 1968 the surface area was reduced to 6.8 ha and maximt.nn depth was only 5.5 m (Brezonik et al. 1969). 7

PAGE 15

CHAPTER 2 LITERATURE REVIEW Phosphorus cycling in aquatic environments is the result of many processes which involve different phosphorus fonns and numerous storage compartments, as generalized in Figure 2-1. While inputs and losses determine the total phosphorus concentration in a lake, within the lake soluble inorganic phosphorus is incorporated into organic com pounds by primary producers, cycled through dissolved and particulate compartments, and returned to inorganic form. The duration of individ ual processes can vary frcm seconds to days or months, and the relative importance of any particular compartment or transformation varies from one lake to another. Because of the importance of phosphorus as a major plant nutrient and its role in lake eutrophication, many researchers have studied the aquatic phosphorus cycle. Their approaches have ranged in scope from focusing on one process or compartment, to complex mathematical models designed to simulate the maJor processes that control lake phosphorus concentration. The following review considers research in the major areas included in this study. Sediment-Water Exchanges Lake sediments act as a net sink for phosphorus through accumula tion of particulate organic matter, but under certain conditions they 8

PAGE 16

WATER -SEDIMENT EXTERNAL P SOLUBLE PARTICULATE INORGANIC 1---------'~0RGANIC P (ALGAE P ZOOPLANK TON, SOLUBLE ORGANIC p -SOLUBLE ORGANIC p -BACTERIA, DETRITUS) --SOLUBLE ORGANIC p INORGANIC p --ORGANIC Figure 2-1. Generalized pathways of phosphorus cycling 1n lakes (after Syers et al. 1973). 9

PAGE 17

can constitute a significant source to lake water. Early studies of the effectiveness and fate of phosphate fertilizers have contributed to our understanding of the behavior of phosphorus 1n sediments. The realization that a large fraction of phosphorus 10 fertilizer was fixed or retained in soil led to investigations of the mechanisms involved. Coleman (1944a, 1944b) found that the presence of iron and aluminum oxides was more important for phosphorus fixation than the type of clay in coarse or fine soil fractions, and he suggested that retention involved an exchange of OHfor H2P04at the oxide surface. Other soils researchers corroborated the importance of iron and alum inum oxides in phosphate fixation (Swenson et al. 1949; Haseman et al. 1950) and demonstrated that some organic acids can decrease phosphate retention over specific pH ranges by forming complexes with the Fe and Al (Struthers and Sieling 1950; Bradley and Sieling 1953). It was unclear whether phosphate retention involved adsorption or precipitation until Fried and Dean (1955) concluded that because a large portion of fixed phosphate was exchangeable with carrier-free inorganic 32 P, adsorption must be the principal mechanism. Hingston et al. (1967) demonstrated that retention of phosphate on oxide surfaces is a specific adsorption mechanism which is independent of the properties of the diffuse double layer or the outer Helmholtz layer. Adsorption 1s accomplished by exchange of the anion for water and hydroxyl ions at the oxide surface, and the reaction always results 1n a decrease in surface charge. They further pointed out that an undis sociated free acid and its most highly charged anion are not adsorbed if present alone because of the requirement for a proton donor and acceptor for specific adsorption to occur. Maximum adsorption of 10

PAGE 18

phosphate increases as pH decreases, with a discontinuity near each pK value, and H2PO4is the form most readily adsorbed. Mortimer (1941, 1942) included phosphorus in his studies of mud water exchanges of dissolved substances. He found that anoxic condi tions at the sediment-water interface cause an increased release of phosphate to the water. Numerous other studies have reaffirmed the relation between anoxic bottom water and enhanced release of phosphate from sediments (Porcella et al 1970; Li et al. 1972; Syers et al. 1973; Kamp-Nielsen 1974; Fillos and Swanson 1975; Armstrong 1979) Reduced forms of iron and manganese are soluble, and mobilization of these elements from lake sediments into anoxic bottcm water also solu bilizes adsorbed inorganic phosphorus (Mortimer 1971; Syers et al. 1973; Armstrong 1979). Vertical mixing processes can recycle the phos phorus to the euphotic zone, where it would be available for biological uptake. Kamp-Nielson (1974) reported a linear relationship between the release of phosphate and its concentration gradient across an anaerobic mud-water interface, but he found that sorption reactions dominated phosphate exchange under oxygenated conditions. Other workers (Hynes and Grieb 1970; Fillos and Swanson 1975) have reported sediment phos phate release under aerobic conditions, but at a much slower rate. The amount of sediment inorganic phosphorus available for release to overlying lake water depends on the size of this phosphorus pool and on sediment characteristics. Li et al. (1972) estimated exchangeable inorganic sediment phosphorus by following the rate of disappearance of inorganic carrier-free 32 P from solution in well-mixed sedimentwater systems. The exchangeable fraction of four Wisconsin lake sedi ments ranged from 19% to 43% of total inorganic phosphorus for both 11

PAGE 19

------aerobic and anox1c conditions, although a significant release of sedi ment inorganic phosphorus occurred under anoxic conditions. Porcella et al. (1970) set up microcosms with sediments as the only source of phosphorus for algal growth. They observed a repeatable series of events in which phosphorus released to an anaerobic layer above the sediment surface led to a benthic mat of the blue-green alga Oscilla toria sp., followed by a bloom of the same species in the overlying water The authors concluded that the Oscillatoria mat enhanced sedi ment phosphorus release by disrupting the sediment-water interface when bubbles occasionally lifted portions of the mat and attached sediment. Biological reworking of sediments is another mechanism that can accelerate sediment-water phosphorus exchange. Davis et al. (1975) investigated the effect of burrowing tubificid worms on phosphorus dynamics in intact mud-water columns. The worms caused an increased removal of 32 P from the water (this became bound to Fe and Al oxides), but did not affect release of 32 P back into the water. In subsequent bioturbation studies (Gallepp et al. 1978; Gallepp 1979), burrowing larvae of chironomid midges increased the phosphorus concen tration in overlying water, but the increase was attributed to excre tion rather than an accelerated sediment release. Lake sediments can adsorb large amounts of added phosphorus in addition to serving as an internal source of inorganic phosphorus. This ability of sediments to buffer aquatic phosphate concentrations has been pointed out by numerous workers (Carritt and Goodgal 1954; Hayes and Phillips 1958; Phillips 1964; Pomeroy et al. 1965; Harter 1968). Carritt and Goodgal (1954) demonstrated that retention of 12

PAGE 20

inorganic phosphorus by estuarine sediments involves a rapid initial adsorption process followed by a slower diffusion reaction. The phosphate adsorbed by sediments from a eutrophic Connecticut lake (Harter 1968) was associated with two sediment fractions: a loosely bound iron fraction and a more tightly bound aluminum fraction. The work of Shukla et al. (1971) with sediments fran nine soft-water and five hard-water Wisconsin lakes showed that noncalcareous sediments adsorbed more phosphorus than did calcareous sediments. Furthermore, phosphorus adsorption by both sediment types was corelated more closely with oxalate-extractable Fe than with any other parameter. The authors postulated that adsorption occurred on a large complex consisting pri marily of hydrated Fe oxide, with smaller amounts of organic matter, Al 2 0 3 and Si(OH)4. Decomposition The amount of inorganic phosphorus present in sediment interstit ial water is affected by sediment characteristics and by decomposition of organic matter. The effect of various enviromnental parameters on phosphorus release from decomposing algae and aquatic macrophytes has been studied in field and laboratory situations. In a sunnnary of pre vious studies, Foree et al. (1970) listed three general stages in the nutrient regeneration process: (1) A rapid (~24 h) initial step in which nutrients are released, absorbed, or released and then re absorbed; (2) a stationary phase of several days with no net change in nutrient concentration; and (3) active net release of nutrients to solution over several hundred days. 13

PAGE 21

Foree and McCarty (1968) followed phosphorus release during the anoxic decomposition of cultured algae. They found that after 200 days of incubation about 40% of the initial particulate phosphorus remained 1n refractory solids. In a related study, the same group (Foree et al. 1970) developed a mathematical model to describe phosphorus regenera tion under anoxic and aerobic conditions as a function of measurable quantities. However, the applicability of their model 1s limited by the fact that some of the terms can only be obtained after lengthy laboratory decomposition studies. After one year of decomposition a larger fraction of initial particulate phosphorus remained in the aerobic (~50%) than in the anoxic (~40%) experiments. Nichols and Keeney (1973) followed the release of phosphorus from herbicide-killed aquatic macrophytes (Myriophyllum exalbescens) 1n water-only and water-plus-sediment systems. They found a rapid initial release of soluble organic phosphorus after the plants were killed, followed by an increase in inorganic phosphorus. Levels of inorganic phosphorus were lower in the systems that contained sediments. The authors concluded that phosphorus released from plants decomposing in a lake was available for incorporation into biomass or adsorption onto sediments. Acid and alkaline phosphatase enzymes produced by bacteria and al gae are important in the remineralization of organic phosphorus com pounds. Reichardt (1975) found sharp increases in bacterial biomass and alkaline phosphatase activity in the first 1 cm of lake sediments and noted that below the upper aerobic layer, bacterial densities de creased while enzyme activity remained nearly constant. He concluded 14

PAGE 22

that the phosphatases at this sediment depth were longer-lived than the bacteria that produced them. Landers (1982) conducted field decomposition studies in the lit toral zone of a soft-water Indiana reservoir. He isolated areas with and without naturally senescing Myriophyllum spicatum in open-ended plastic enclosures and observed changes in nutrients and chlorophyl~-~ over 119 days. Phosphorus released from the macrophytes (extrapolated to a whole-lake basis) equalled about 2-18% of the total annual phos phorus loading to the lake, and concurrent increases in chlorophyll-~ indicated a significant phytoplankton response to the release. Planktonic Phosphorus Cycling Observations that nearly undetectable levels of dissolved inor ganic phosphorus are often adequate to support phytoplankton blooms led to the hypothesis that phosphorus cycling within the water column is a rapid process. The use of radioactive 3 2p has facilitated accurate estimates of planktonic phosphorus uptake rates and turnover times with addition of as little as 0.002% of ambient inorganic phosphorus levels. Hayes and Phillips (1958) and Phillips (1964) used 3 2p to study phosphorus equilibrium in systems containing mud, water, plants and bacteria; they summarized their findings and earlier work to provide an integrated concept of phosphous cycling among the components of a whole-lake system. Their estimated turnover times included 1 week for the water of a whole lake; 0.3 days for bacterial or phytoplankton cells (but 5 min for initial equilibration); 3---4 days for rooted aquatic macrophytes; and 1 day for zooplankton (which can utilize only 15

PAGE 23

organic phosphorus). The authors emphasized the influence of bacteria in retaining phosphorus in the water column through incorporation into organic forms (and thus preventing adsorption by sediments) or by accelerating the return of phosphorus from the sediments. Rigler (1964) examined water column phosphorus fractions in dif ferent types of temperate North American lakes and found that soluble organic phosphorus (SOP) represented about 18% of total phosphorus (TP) in all trophic types. He attributed wide variations in SOP reported in the literature to variations 1n filter pore size and methods used to remove seston from the water. He found that turnover of inorganic phosphorus by seston (using carrier-free 32 P) was less than 10 min in all eight lakes during the summer, and increased in winter. Rigler (1966, 1968, 1973) later contended that his 32 P uptake data demon strated that colorimetric analyses overestimate inorganic phosphorus concentrations. However, as pointed out by Lean and White (1983), the inconsistency in Rigler's data was due to his failure to consider that plankton could excrete unlabelled inorganic phosphorus, rather than overestimation. Understanding of planktonic phosphorus cycling was advanced by Lean (1973a, 1973b), who used Sephadex gel to separate soluble 32 P fractions on the basis of molecular size. From his results with this technique 1n a eutrophic Canadian lake, Lean proposed a generalized description of phosphorus movement between biologically important forms. He demonstrated the rapid formation of a dissolved algal or bacterial organic phosphorus compound (molecular weight about 250) that becomes associated with a high-molecular weight colloid, releasing orthophosphate in the process. The colloidal phosphorus fonn comprises 16

PAGE 24

a large proportion (about 77%) of nonparticulate phosphorus, but this fr act ion is not available for algal uptake. Both chemical and radioisotope methods can be used to study phos phorus uptake by lake plankton. The application of these methods and the significance of their results were recently reviewed by Lean and White (1983), who used both techniques to estimate phosphorus uptake rate constants for the same lake samples. They concluded that small cells dominate uptake when low amounts of phosphorus are added, while at high added phosphorus concentrations, uptake is primarily by large cells. The comparison of phosphorus uptake rates by seston fran dif ferent lakes is therefore practically impossible because of differences in size distribution of plankton and variations in amounts of phos phorus added by different researchers. Zooplankton phosphorus excretion has long been recognized as a recycling mechanism in the water column, but there has been little agreement about its relative importance or whether organic or inorganic forms predominate. Johannes (1965) investigated interactions between marine protozoa and bacteria and their effect on phosphorus cycling. Re found that protozoan phosphorus excretion (per unit weight) is 1-2 orders of magnitude faster than that of marine microcrustaceans, and several orders of magnitude faster than marine macrofauna. The proto zoan-bacterial interaction involves consumption of organic detritus by bacteria, which in turn are grazed by protozoans. Bacterial popula tions are maintained in a state of "physiological youth" by protozoan grazing, thus increasing regeneration of inorganic phosphorus. Buech ler and Dillon (1974) found that phosphorus uptake by freshwater cili ated protozoans (Paramecium spp.) was effected by ingestion of bacter ial biomass, and that phosphorus turnover rates were extremely fast. 17

PAGE 25

Hargrave and Geen (1968) measured rates of excretion of unlabelled soluble phosphorus for several species of marine crustaceans and one rotifer. Although soluble organic phosphorus constituted up to 75% of the amount regenerated, they calculated that zooplankton released enough inorganic phosphorus to the photic zone to supply one-fifth to two times the daily phytoplankton requirement. In all cases the mea sured excretion rate was decreased by increased bacterial activity and experimental duration. Use of Sephadex to fractionate labelled phos phorus compounds has provided an explanation for these observations (Peters and Lean 1973; Peters and Rigler 1973). This technique showed that about 90% of soluble phosphorus released by Daphnia rosea and Diaptomus minutus was inorganic phosphorus. However, bacteria quickly assimilated most of the released inorganic phosphorus, which accounts for the relation Hargrave and Geen (1968) found between phosphorus excretion and bacterial activity or length of incubation. Bacterial uptake also explains the high proportion of SOP found by earlier work ers. Peters and Rigler (1973) further estimated that overall phos phorus cycling efficiency of zooplankton [(P regenerated)/(P regener ated+ P sedimented) x 1000] is as high as 88--93%, which again empha sizes the important role of zooplankton excretion in maintaining phos phorus in the water column. As pointed out by Johannes (1965) for marine zooplankton, there is an inverse relationship between body size or body weight and rate of phosphorus regeneration, so that small species are potentially more important in regenerating soluble phosphorus than larger forms. This relationship has been noted by nlUllerous others (e.g., Hargrave and Geen 1968; Peters and Rigler 1973), and it implies that a lake's trophic 18

PAGE 26

state could be affected by processes which change the size distribution of its zooplankton. Fish constitute another influence on water column phosphorus cyc ling. Kitchell et al. (1975) employed a mass-balance approach to eval uate the importance of phosphorus flux through fishes. They calculated that production of fish biomass fixes 60-70% of the annual phosphorus input to Lake Wingra, Wisconsin. The fraction that is incorporated into bones and scales ( "v 50%) will not be remineralized through decompo sition, and thus is effectively lost to the system However, the auth ors suggest that the seasonal pattern of fish mortality in temperate lakes (high mortality after spring spawning) results in a significant supply of phosphorus from decomposing fish biomass in late spring and early summer. Lake Phosphorus Models Efforts to model phosphorus dynamics in lakes have shown varying degrees of success, depending in part on the complexity and objectives of the modeling efforts. Lake management applications began with a simple mass balance approach to lake phosphorus concentration. This involves estimating the change in lake phosphorus storage that results from a balance of loading terms and loss terms. Non-point sources are usually estimated from land use data, while sedimentation rates fre quently are derived from the literature and adjusted to calibrate the model. By examining the relationship between phosphorus loading and trophic state indicators (e.g., total phosphorus, chlorophyll-_!, Secchi depth, hypolimnetic oxygen deficiency), Vollenweider (1975), Dillon and Rigler (1974) and others have developed critical loading limits, below 19

PAGE 27

which eutrophication could be avoided. Furthermore, these relation ships could be used to predict the effect of changes in loading rates on lake phosphorus concentration, including the time required to reach a new equilibrium after a change in input. The application of some of these models has resulted in the need to modify them to fit observed conditions. Yeasted and Morel (1978) used a combination of phosphorus budget modeling and stepwise discriminant analysis to evaluate the ability of water residence time, mean depth, and lake surface area to describe the non-conservative behavior of phosphorus in 128 phosphorus limited lakes (71 eutrophic, 42 mesotrophic, and 15 oligotrophic). They found that only hydraulic residence time gave consistent statisti cal significance. Shannon and Brezonik (1972) and Baker et al. (1981) have developed nutrient loading-trophic state relationships specific to Florida lakes. In spite of the problems inherent in the application of mass bal ance models, their very simplicity makes them an attractive management tool. They can be used with a reasonable degree of accuracy to simu late and evaluate the effect of various options, provided data are collected carefully and the assumptions are not violated. In contrast to the simple mass balance approach are more complex approximations of the non-conservative behavior of phosphorus within a lake. These models use differential equations to represent changes in various processes (e.g., production of phytoplankton biomass) as a function of time. Models involving phosphorus range from those that consider only phytoplankton as a biotic component (Fleming 1975) to more complex, multi-component ecosystem models (Chen 1969). The simpler models do not yield realistic results, but it is difficult to obtain all the coefficients needed in more complicated models. 20

PAGE 28

Nevertheless, ecosystem modeling provides the only feasible alternative for integrating so many processes and parameters. Effects of pH on Phosphorus Cycling Acid deposition and the acidification of surface waters are recent enough phenomena that most research has been focused on identifying effects and documenting the extent of affected areas, instead of ident ifying the mechanisms involved. As mentioned earlier, while many stud ies have shown decreasing TP levels with decreasing lake pH, there is little evidence to link the observed TP decrease to acidification. Most available information concerning pH effects on phosphorus dynamics relates to sediment-water interactions and decomposition. MacPherson et al. (1958) examined the effect of pH on the parti tioning of inorganic phosphorus between water and the mud of unproduc tive, moderately productive, productive, and acid bog lakes. They found similar trends in all lake types, with minimum phosphorus release from the mud in the pH range 5.5-6.5. More inorganic phosphorus was released at higher and lower pH values. The acid bog and productive lake muds did not adsorb appreciable amounts of added phosphorus; the unproductive lake mud removed most of the added phosphorus in the acid pH range but not at pH 7. Increased phosphorus adsorption as pH decreases has been shown for soils (Lopez-Hernandez and Burnham 1974). Andersson et al. (1978) and Gahnstrom et al. (1980) varied the pH of water overlying sediment cores from acidic and alkaline Swedish lakes. They found more inorganic phosphorus was released from the sediments at high pH than at low pH. 21

PAGE 29

Consideration of the effect of pH on solubility of metal phos phates shows the controlling phases under equilibrium conditions (Stumm and Morgan 1981). At low pH strengite (FeP04) and variscite (AlP04) are the solid phases that may control phosphate solubility, while above pH 6 calcium phosphates (notably apatite) are the predomin ant solids. However, due to rapid biological transformations, equilib rium is rarely attained in the pH range of natural waters. Singer et al. (1983) added 32 P to the water over intact sedi ment cores from an acidic Adirondack lake. Some had a mat of Sphagnum sp., and two concentrations of aluminum ~re used (0 and 300 g/L). They concluded that the algal mat was much more efficient than bare sediment at removing water column phosphorus, and that precipitation reactions (with Al concentrations up to 300 g/L) were unimportant. Grahn et al. (1974) theorized that acidification inhibits decompo sition because of the accumulation of organic matter which they observed in the sediments of many acid lakes Studies on the effect of pH on decomposition have been inconclusive. Some measures of decompo sition, such as leaf litter weight loss and numbers of total bacteria, decrease at low experimental pH (Leivestad et al. 1976; Traaen 1980). Others (sediment oxygen demand, glucose turnover) show no effect of reduced experimental pH (Andersson et al. 1978; Gahnstrcm et al. 1980), although sediment oxygen demand and glucose turnover both increased in lakes after lime treatment. Finally, the experimental acidification of a Canadian lake (Schindler 1980) resulted in no significant change in TP and no evidence of decreased decomposition over 3 years. However, 22

PAGE 30

it should be pointed out that the pH change during this period was only from 6.6 to 5.6. Another potential effect of lake acidification relates to the importance of pH in controlling phosphate uptake by algae (Wetzel 1983). Different species have distinct pH ranges in which they show optimum growth and phosphate uptake. This is due in part to the pH specificity of extracellular or membrane-associated enzymes, but pH can also alter the permeability of the cell membrane and change the ionic form of inorganic phosphate in the growth medium. 23

PAGE 31

CHAFtER 3 PHOSPHORUS DYNAMICS IN MCCLOUD LAKE This chapter includes a discussion of the general limnology of McCloud Lake as well as the contributions of important processes and storage compartments to the dynamics of phosphorus cycling within the lake. The historical data base for McCloud Lake is examined, as well as the results of laboratory and in situ experiments. Materials and Methods Routine Sampling Limnological data were collected monthly at a mid-lake station from October 1980 through September 1982. Field data included Secchi disk transparency, and temperature and dissolved oxygen profiles, which were measured with a YSI model 54A DO meter. In the laboratory specific conductance was measured with a YSI model 31 conductivity bridge; pH was measured with a Fisher Accumet model 230A pH/ion meter equipped with an Orion internal reference calomel electrode. Chemical samples were collected at 1.0-m intervals and stored in separate polyethylene bottles for major ions (cone. HN03 to pH< 2) and nutrients (1 mL saturated HgCl2 per L of sample). Chlorophyll-~ and phyto plankton samples were collected as a water column composite (1-m inter vals), while zooplankton were collected by vertical tows of a #20 Wisconsin plankton net (80 m mesh). Phytoplankton and zooplankton 24

PAGE 32

were preserved with 1% Lugol's iodine and 5% buffered formalin, respectively. Standard procedures were followed for all chemical analyses (APHA 1980; U.S. EPA 1979). Major cations were analyzed (flame mode) on a Perkin-Elmer Model 5000 atomic absorption spectrophotometer. Chloride, sulfate, silica and nutrient forms were analyzed by automated colori metric procedures (Table 3-1). Semi-micro digestion procedures were used for total phosphorus (autoclaved persulfate digestion) and total Kjeldahl nitrogen (block digestion). Chlorophyll-.!!, was measured by the trichromatic method (APHA 1980). Phytoplankton aliquots (l0-30 mL) were concentrated in Utermohl cham bers and counted using methods described by Lund et al. (1958). Zoo plankton were identified and counted in 1-mL Sedgwick-Rafter cells under a light microscope. Phosphorus Budget Water budget. Monthly data for precipitation, seepage, evapora tion, and change in Mccloud Lake level (stage) were compiled by Baker (1984), who also discussed the equipment and methods used to collect the data. The resultant water budget is a necessary prerequisite for the construction of a phosphorus budget for the lake. Stage-area and stage-volume relationships were determined from a bathymetric map of McCloud Lake (Figure 1-1). I constructed the map from 14 fathometer transects and aerial photographs taken in January 1982, which corresponded to the lowest lake stage during the 1980-1982 period. Nine north-south and five east-west transects were run using a Lowrance model 1510B Truline recording fathometer in a skiff powered by 25

PAGE 33

Table 3-1. Automated colorimetric procedures used in McCloud Lake study. Parameter Chloride Sul fate Silica Ammonium Nitrate+ Nitrite Phosphorus EPA Method 325.1 375. 2 370. 1 350.1 353. 1 365.2 26

PAGE 34

a small outboard motor at constant speed. The ends of the tranects were marked with black plastic sheeting and white plastic milk bottles to ensure their visibility in the subsequent aerial photography. In addition, the distances from markers to the beginning or end of each fathometer transect were recorded. An outline of the lake (including the transect markers) was drawn from the best aerial photograph, and the scale was determined from measured distances between markers. Changes in depth along each transect were plotted on this outline map, and contours were drawn at 2-foot depth intervals. Lake stage was 2 feet higher on September 22, 1982, than when the bathymetric survey and aerial photography were conducted. This new shoreline was added to the bathymetric map from measurements of distances between the transect markers (which were still marked by wooden stakes) and the new lake shore. The volume and area of the lake were calculated for each date with a Hewlett Packard 9810A calculator equipped with a digitizer surface. These data were used to establish the stage-area and stage-volume relationships needed for the water budget calculations. Atmospheric phosphorus loading. Rainfall samples collected at the lake from September 1981 to August 1982 were analyzed for TP. Total phosphorus deposition rates (wetfall and dryfall) were estimated from the relation between wet and total phosphorus deposition measured at several sites in Florida (Brezonik et al. 1983). Sedimentation. Rates of sedimentation of particulate matter and phosphorus were measured in McCloud Lake with cylindrical sediment traps (Figure 3-1). Design of the traps followed the general recommen dations of Blomqvist and Hakanson (1981), who published an extensive 27

PAGE 35

\ \ \ \ \ \ \ \ \ \ \ MARKER BUOY \ FLOTATION I \ __ I \ / '/ ~/ > Figure 3-1. Sediment trap design. 28 SEDIMENT SURFACE

PAGE 36

review of the design and performance of sediment traps in aquatic sys tems. They concluded that a simple cylinder gives better results than other shapes when a proper height-to-diameter ratio (H/D) is used to limit resuspension losses. In general they reconnnended a vessel with diameter> 20 mm and H/D 3 or 4. The base of the traps consisted of a plexiglass rectangle (20 x 27 cm) with six holes for the cylinders and plastic foam for flotation. Rubber bands around the Pyrex test tubes (22 x 150 mm, H/D = 6. 7) pre vented them from falling through the holes. This buoyant trap appar atus was moored ~0.4 m above the lake bottom in the center of the lake ("-4.5 m total depth). Removal of the large flotation bucket allowed the trap platform to float up to the surface to facilitate recovery of the test tubes with minimal disturbance of the sediments. Three of the six tubes were placed upright to trap sedimenting particulate matter, while the other three were installed upside-down to estimate the biomass of colonizing invertebrates and attached algae At the beginning of the first incubation period the platform was raised to the surface and six tubes were installed. The platform was then carefully lowered frcm the surface using the mooring line, and the flo tation bucket was attached approximately 0.5 m below the lake surface to maintain a constant tension. At the end of each incubation period (30 or 60 days), I carefully swam down to the trap array and inserted rubber stoppers in all six tubes. The bucket was removed from the mooring line so the platform could be raised to the the surface for recovery of the tubes. After new tubes were installed the platform was again carefully lowered to initiate a new incubation. 29

PAGE 37

The test tubes were tared in the laboratory prior to incubation. After an incubation, the outsides of the stoppered tubes were carefully cleaned to remove any attached particulates. The stoppers were removed and the tubes were placed in a drying oven ( "'60 c) to evaporate all water. When the contents were dry, each tube was cooled and reweighed to allow calculation of sedimentation rates on a dry-weight basis. Next, 20 mL of distilled deionized water was added to each tube and a wet persulfate digestion was performed in the autoclave. Total phos phorus was measured as SRP in the filtered digestate. McCloud Lake Phosphorus Compartments In addition to the water column phosphorus analyses previously described, two other in-lake reservoirs of phosphorus were evaluated. Macrophyte survey. At the beginning of this study in December 1980, McCloud Lake sediments were relatively barren and free of algae or higher plants except in the very shallow littoral areas, where some emergent species were found. However, during the study two submersed macrophytes, Websteria ap. and Eleocharis sp., became established in significant proportions in both littoral and deeper areas of the lake. In September 1982, a survey was conducted to determine the areal extent and density of these macrophytes and to estimate the amount of phos phorus bound in their biomass. I sampled 14 transects spaced around the lake by swimming (with SCUBA equipment) from the shore out toward the lake center. Each transect started at a marker used in the bathymetric survey in order to facilitate mapping the macrophytes. Data recorded for each transect included the distances from shore and depths at which the macrophytes 30

PAGE 38

beds began and ended, as well as the dominant species In addition, a composite macrophyte sample was obtained for each transect by manually collecting all plants within a quadrat (0.016 m 2 ) which was randomly placed at four points more or less evenly spaced along the transect. Additional samples of Eleocharis and Websteria were collected for digestion and TP analysis. The composite samples were carefully washed in the laboratory and dried in tared envelopes at 60C to get dry weight per unit area. Sub samples (0. 5 g) of dried tissue were ashed at 500C and ash-free dry weight was calculated. The samples collected for phosphorus analysis were also washed and dried at 60C. Weighed subsamples (~0.1 g) were placed in large test tubes; distilled deionized water (20 mL) was added and a wet persulfate digestion was performed 1n the autoclave. Total phosphorus was measured by the SRP procedure on an aliquot of the filtered digestate. Sediment phosphorus. A composite surficial sediment sample was collected from the center of McCloud Lake by pooling 6 grabs of a petite Ponar dredge. Subsamples of this sediment were analyzed for TP us 1ng a method which involved ashing at 550 C followed by HCl diges tion, as described by Andersen (1976). Phosphorus uptake. Rates of phosphorus uptake and turnover were measured for the submergent macrophytes and for mid-lake seston using radiolabeled ( 32 P) orthophosphate. Intact Eleocharis plants were collected from the lake, transferred to the lab, and carefully washed to remove sediment and attached algae. The washed plants were blotted dry and about 2.0 g of intact plants were placed in each of four PVC trays in 250 mL of membrane-filtered (0.45 m) lake water at room 31

PAGE 39

temperature. Two trays were covered with aluminum foil to exclude light, and the other two plus a control (250 mL filtered lake water without plants) were incubated under fluorescent light. An aliquot (1 mL) of a stock solution of 32 P enriched ortho phosphate solution was added to each tray, and its disappearance from the medium was followed by periodically withdrawing 1-mL aliquots. These samples were placed in scintillation vials and counted with a liquid scintillation counter. Mid-lake seston samples (1 L) were incubated under fluorescent light and slowly mixed with magnetic stirrers. Aliquots of a stock K 2 HP0 4 carrier for 3 2po 4 were added to the seston samples, and uptake of 32 P was followed by periodically withdrawing and filtering 5-mL subsamples through 0.45 m membrane filters. Methods used to analyze 32 P samples and to plot and analyze the data are discussed in Chapter 4. Results and Discussion Limnology and Historical Nutrient Trends McCloud Lake has uncolored, soft water that usually is clear. Conductivity is low ("'40 mho/cm), and divalent cation (ca+ 2 + Mg+2) concentrations are only about 150 eq/L. A Secchi disk is visible on the lake bottcm during winter months, but Secchi transpar ency is as low as 1. 75 m during summer peaks of phytoplankton. Water clarity is occasionally reduced when littoral sediments are suspended by wave action, but the surrounding hills and small fetch minimize wind influence on the lake. 32

PAGE 40

The water column does not stratify, although bottom temperatures and dissolved oxygen concentrations are generally somewhat lower than surface values (Figure 3-2). The average difference between surface and bottom temperature is 1 C; maximl.Ull differences up to 3C occur during warm months Percent oxygen saturation shows no difference between surface and bottom waters in winter months, but oxygen satura tion is consistently lower near the bottom during warm months, reflect ing increased biological activity. Overall, oxygen saturation ranges from 73% to just over 100%, and surface and bottom averages are 93% and 89%, respectively. The generally undersaturated conditions reflect the oligotrophic status of the lake. The pH of McCloud Lake decreased from 4.85 in 1967-1968 (Brezonik et al. 1969) to 4.71 in 1978-1979, and generally was less than 4.60 during 1980-1982 (Table 3-1 and Figure 3-2). This represents nearly a doubling of H+ concentration in 15 years. Present pH values correspond closely to rainfall pH. The increase in conductivity from 1967-1968 (32 mho/cm) to 1980-1982 (42 mho/cm) indicates that concen tration of ions by evaporation may account for some of the pH decline. This hypothesis is supported by the decrease in pH from September 1981 through January 1982 (Figure 3-2) which corresponded to the lowest lake level in 2 years. Lake level began to rise as normal rains resumed in February 1982, and the subsequent pH increase reflected this dilution. Table 3-1 summarizes nutrient conditions in the lake during 19671968 (Brezonik et al. 1969), 1978-1979 (unpublished data), and for the 2 years of this study. There has been remarkably little difference in average nutrient concentrations over this 15-year period. Mean values of several nutrient parameters (TON, TP, SiOz, TN/TP) for 1978-1979 33

PAGE 41

100 z 0 j:: < a: => t7!5 < "' N 0 'ii50 5.0 :t: o. 4 5 4 0 + Sfc T Sfc o 2 0 Bot 02 0 No: J 80 I 0 N 80 I I D: J I I F M A M J J A 81 FMAMJJA 81 s 0 S 0 N D' I J I N : I I F M A M J J A s 82 FMAMJJAS 82 32 30 26 22 18 14 Figure 3-2. Dissolved oxygen and pH trends 1.n McClou:i Lake, October 1980 through September 1982. 34 cE Q. w tw u < LL. a: => "'

PAGE 42

Table 3-2. Annual means and standard deviations (n = number of samp ling dates) of nutrient and limnological parameters for Mccloud Lake. 1967-68 1978-79 1980-81 1981-82 Parameter (n = 12) (n = 4) (n = 12) (n = 12) pH 4.85 4. 71 4. 56 4. 50 Conductivity* 32 44.8 42 + 2. 6 TONt 420 243 290 + 120 423 + 220 NH4-Nt 105 137 62 + 50 56 + 40 N03-Nt 41 47 49 + 20 68 + 40 N02-Nt 1 2 1 + 0. 4 1 + 1 TPt 12 16 9 + 3 12 + 7 SRPt 6 4 5 + 0. 2 3 + 3 Si02t 100 265 213 + 110 118 + 70 Chlorophyll-~t 1. 94 0.88 5.7 + 2.9 4.7 + 3.9 TN/TP 47.3 26.8 44. 7 45. 7 l Ca + Mg** 77 98 147 s04-2** 104 142 140 l Cationstt 205 234 313 L Anions** 2 71 287 310 *\Jmho/cm. tg/L. wt/wt. tteq/L. 35

PAGE 43

are not consistent with mean data from the other years, but this may reflect the limited number of sampling dates in 1978-1979 rather than changes in lake chemistry. + TON, TP, and NH4-N were lower for 1980-1981 than for 1967-1968. ArnmoniLilTI in 1980-1982 was about 50% of 1967-1968 levels, but both were low; TON and TP means were identical for the two periods. TN/TP ratios also show little change over the 15 years since 1967-1968, with the exception of the 1978---1979 data. According to criteria proposed for Florida lakes (Huber et al. 1982), the TN/TP value of about 45 (weight basis) indicates that phosphorus is the limiting nutrient in McCloud Lake. The average chlorophyll-~ (Table 3-1) during 198~1982 (5.2 g/L) was more than two times as high as the mean value reported for 19671968 (1.9 g/L), and was nearly six times the 1978-1979 mean. Never theless, these chlorophyll-~ levels all are indicative of oligotrophic conditions and the differences appear insignificant. A relationship established for TP and chlorophyll-~ in phosphorus-limited Florida lakes (Huber et al. 1982) predicts that McCloud Lake (TP 11 g/L) should have a mean chlorophyll-~ concentration of 3.1 g/L. This value agrees well with measured chlorophyll-~ and indicates that conditions in McCloud are typical of those in phosphorus-limited oligotrophic Florida lakes. Figures 3-3 and 3-4 show monthly variations in McCloud nutrient and biological parameters from October 1980 to September 1982. TP and TON exhibited maximum values in spring and summer when chlorophyll-~ and total zooplankton abundances were highest. No trends can be dis cerned in variations of SRP. Both TP and SRP showed summer increases during 1967-1969. Total phosphorus and TON were generally higher 36

PAGE 44

37 150 ..J NH4 0 N0 3 .... 0) :t z 100 I ,., 0 z C z 50 < z !,t 0 z F M A M J J A s 0 N D : J F M A M J J A s 81 82 30 TP o SRP .,J .... 0) :t 20 CL a: u, C 10 z < CL ... 0 0 N o,J F M A M J J A s 0 N D J F M A M J J A s I I 80 81 I 82 400 ..J 300 .... 0) :t 200 N 0 CIJ 100 0 I I I I I I 0 N o:J F M A M J J A s 0 N D l J F M A M J J A s 80 I 81 I 82 I I Figure 3-3 Variations in dissolved silica and nitrogen and phosphorus forms in McCloud Lake, October 1980 through September 1982.

PAGE 45

800 600 ..J Q ::L. 400 z 0 I200 0 I I I 0 N D 'J F M A M J J A s 0 N o!J F M A M J J A s I 80 81 I 82 t56,600 ..J 16 Chi !l O PHYTOPLANKTON 0) ::L. 12 a,j ..J ..J 8 > ::c: a. 0 4 a: 0 ..J ::c: 0 0 M J J A s 82 600 ..J ...... "' z 400 0 Iz < 200 ..J a. 0 0 N 0 I I I I I I I I I I I I I I I I I I I I I I I I 0 N D I J F M A M J J A s 0 N D IJ F M A M J J A s I I 80 I 81 I 82 I I Figure 3-4. Variations in TON and biological parameters in McCloud Lake, October 1980 through September 1982. 33 40 .J E ..... ,., 30 z 0 20 Iz < 10 .J a. 0 I0 > ::c: a.

PAGE 46

throughout the 1981-1982 period than for 198~1981. The lower TP and TON of 1980--1981 correspond to a drought and falling lake levels, while the higher TP and TON of 1981-1982 reflect increased nutrient loadings due to increased rainfall and rising lake levels. Nitrate maxima occurred during winter months when biological activity was lowest, and minimum nitrate concentrations corresponded to peaks of phytoplankton and zooplankton abundance. Ammonium showed peaks during warm months and low levels in winter. The sum of ammonium and nitrate was almost never (50 g/L; maxima of both species were always <150 g/L, and usually were <100 g/L. The increase in nitrate and concurrent decrease in ammonium which occured during winter 19811982 suggest that nitrification was occurring. Silica concentrations also were low in winter and peaked in midto late summer, but never reached levels (>0.5 ppm) considered optimal for diatom production (Fogg 1975; Wetzel 1983). Chlorophyll-.!!_ (Figure 3-4) showed peaks of algal production during spring and fall in 1981 and early summer in 1982, although these trends did not correspond closely to changes in phytoplankton abundance (Figure 3-4). This was probably due to wide fluctuation in densities of microflagellates and small green coccoid and spindle-shape phyto plankters. As a group ultraplankton (1-10 m) represented an important fraction of total phytoplankton numbers in the lake during both years of the study. A large pulse 1n ultraplankton occurred during June through September 1982, when total phytoplankton densities exceeded 50,000 cells/mL. With the exception of occasional pulses of Dinobryon cylindricum, Q. divergens, and to a lesser extent Asterionella sp., net plankton (>50 m) rarely constituted a major component of the phyto39

PAGE 47

plankton. Oocystis gloeocystiformis consistently formed a large frac tion of total phytoplankton and often comprised greater than 50% of total abundance. Other less abundant but common genera included Perid inium, Kirchneriella, Staurastrum, Mallamonas, Cosmarium, and several unidentified penate diatoms. McCloud Lake exhibits an impoverished phytoplankton cotmnunity with few consistent seasonal trends in species succession Large masses of filamentous algae were observed in certain areas of the littoral zone in early spring. Peaks in total zooplankton abundance generally followed peaks in chlorophyll-a. Maximum densities occurred during late summer and fall 1981 and mid-summer 1982 (Figure 3-4). Diaptomus mississippiensis, Eubosmina tubicen, Diaphanosoma sp., and Keratella gracilenta usually comprised 75-100% of total zooplankton numbers. No consistent pattern of species succession was demonstrated, although D. mississippiensis generally increased in importance during summer, while K. gracilenta frequently dominated during winter months. Eubosmina constituted 40-50% of zooplankton totals during September and October 1981 and was a principal sub-dominant in nearly every lake sample. The 34 zooplank ton species observed during 198~1982 included four copepod, five clad oceran, and 25 rotifer species. Except for a reduced standing stock, the zooplankton community of Mccloud Lake closely resembles those found in more productive Florida lakes of higher pH. Mccloud Lake Hydrology Table 3-2 presents monthly precipitation amounts measured at McCloud Lake from August 1981 through July 1982, as well as estimates 40

PAGE 48

41 Table 3-3. Mccloud Lake hydrology data, August 1981 through July 1982. Precipitation, Lake Volume, Lake Area, Date cm 10 3 m 3 10 3 m 2 1981 Aug 14.30 134.81 51. 57 Sep 7.70 131.09 50.84 Oct 1. 58 124.59 49.55 Nov 6.86 123.19 49. 27 Dec 4.12 120.56 48. 75 1982 Jan 17. 60 121. 49 48.94 Feb 9.16 121. 33 48. 91 Mar 12.50 117. 46 48.14 Apr 23.11 130.16 50.65 May 8. 47 134.03 51.42 Jun 34.34 134. 50 51. 51 Jul 13.36 143.17 53.23 TOTAL 15 3. 10 MEAN 128.03 50. 23

PAGE 49

of lake volume and surface area calculated from lake stage measure ments. Sixty percent of the total annual rainfall occurred between March and July 1982. The dry conditions between August 1981 and Febru ary 1982 caused lake volume to decrease by nearly 13% while surface area decreased about 6.5%. Baker (1984) constructed a water budget for Mccloud Lake for the time of this study. He found that precipitation accounted for 90% of the total annual water input, and seepage into the lake contributed the remaining 10%. The sandy soils 1n the watershed preclude significant surface runoff and allow most rainfall to percolate directly to the shallow water table rather than to the lake. Evaporation was the most important mechanism for loss of water from McCloud Lake during the study, although outseepage rates can exceed evaporation rates when the shallow water table is low. McCloud Lake Phosphorus Budget Precipitation. Table 3-3 presents monthly phosphorus deposition to McCloud Lake from August 1981 through July 1982. Wet deposition values are based on TP concentrations in precipitation samples and rainfall amounts. Total deposition rates were estimated from a previ ous study because dryfall samples were not collected at McCloud Lake. Brezonik et al. (1983) monitored precipitation chemistry with a network of 26 stations in Florida that included four wet-dry collectors (1 urban, 1 coastal, and 2 agricultural stations). Wet TP deposition averaged 20% of total TP (wet plus dry) deposition at the four sites for May 1978 through April 1979. However, dry deposition of phosphorus was more important at the agricultural sites because of fertilizer use 42

PAGE 50

Table 3-4. McCloud Lake phosphorus storage and atmospheric loadings, August 1981 to July 1982. Phos:ehorus TP Storage, Wet Wet, Total 1 a Total 2 a mg/m 2 Month Kg g g g 1981 Aug 1. 35 o. 810 41. 77 126.6 167.l Sep 1.57 0.405 20. 59 62.4 82.4 Oct 1. 25 0.126 6.24 18.9 25.0 Nov 1. 36 0.755 3 7. 20 112. 7 148. 8 Dec 0.60 0.376 18.33 55.5 73. 3 1982 Jan 0.85 1. 507 73. 75 223.5 295.0 Feb 0.36 0.980 4 7. 93 145. 2 191. 7 Mar 1.88 0.800 38.51 116. 7 154.0 Apr 1.04 3.175 160. 81 487 .3 643.2 May 2.41 0.677 34. 81 105.5 139.2 Jun 2.56 3.323 171.17 518. 7 684. 7 Jul 3.29 1.440 76.64 232.2 306.6 TOTAL 14.374 72 7. 76 2205.2 2 911. 0 MEAN 1. 54 aAssumes Wet = 0.33 total. bAssumes Wet = 0.25 total. 43

PAGE 51

and increased dust associated with agricultural practices. Since McCloud Lake is not located in an area of intense agriculture, two estimates of total TP deposition were calculated, based on assumptions that wet deposition represented 25% and 33% of total TP deposition. The estimated total TP loading to McCloud Lake ranged from 2.21 to 2.91 Kg/yr, or 43.9 to 58.0 mg/m 2 yr, This range compares favorably with the mean statewide total TP deposition of 51.0 Kg/hayr reported by Brezonik et al. (1983), but it is more than double the value they found at rural non-agricultural sites (27.0 mg/m 2 yr). Local land use patterns and annual climatic variations can strongly affect nutrient deposition rates, and it is possible that McCloud Lake is atypical of the rural non-agricultural sites monitored by Brezonik et al. (1983). Baker (1984) compared atmospheric nutrient deposition rates at McCloud Lake to nutrient loading criteria established by Vollenweider (1968, 1975) and Shannon and Brezonik (1972). He concluded that atmos pheric nitrogen loading exceeded the minimum input required to sustain mesotrophic conditions, and that atmospheric phosphorus loading was less than half the minimum mesotrophic loading rate. However, his estimate of phosphorus deposition was based on the average rate at rural, non-agricultural sites frcm Brezonik et al. (1983). When atmos pheric phosphorus loading is estimated from precipitation samples collected at McCloud Lake, two of the three minimum mesotrophic loading criteria are exceeded (Table 3-4). Given the oligotrophic status of Mccloud Lake, this suggests that lake acidification may in fact contribute to the low TP and production which are typical of acidic 44

PAGE 52

Table 3-5. Atmospheric deposition of phosphorus at McCloud Lake and loading criteria. Reference Vollenweider (1968) Shannon and Brezonik (1972) Vollenweider (1975) *Calculated by Baker (1984). tThis study. Units mg/m 2 yr mg/m 3 yr mg/m 2 yr Minimum* Mesotrophic Loading 44.0 22. 0 100. 0-110. 0 McCloudt Total Deposition 43.9-58.0 17.3-22.8 43.9-58.0 45

PAGE 53

~ --lakes. However, more detailed studies using data for many lakes would be necessary to provide a satisfactory answer to this question. Mass balance. The seepage contribution to total phosphorus load ing to McCloud Lake was not evaluated in this study because water in the seepage meters became anoxic, thereby promoting solubilization of TP from the sediments. However, since seepage accounted for only 10% of the annual water input, it was assumed that the relative importance of phosphorus input by seepage was minor. This is supported by the fact that SRP tends to be low in groundwater because it adsorbs to clays and hydrous oxide surfaces. The budget summarized in Table 3-5 indicates that phosphorus has a very short residence time in McCloud Lake (O.S-0. 7 yr), which is in keeping with rapid SRP uptake rates and internal phosphorus cycling mechanisms Sedimentation Sediment traps were deployed in McCloud Lake for one 2-month and three 1-month incubations. Gross monthly sedimentation rates were cal culated for total dry sediment and for total phosphorus (Table 3-6). Particulate and phosphorus sedimentation rates were higher during sum mer months when lake productivity was at a maximum. Extrapolation from the 158 days of measured sedimentation to annual figures yields a dry sedimentation rate of 429 g/m 2 yr and a phosphorus sedimentation rate of 370 mg/m 2 yr. These results are probably overestimates for the lake as a whole since more sediment accumlates in the center of the lake than in littoral areas, but several lines of evidence indicate that the trap estimates are reasonable for the pelagic zone First, ZlOpb dating of one profundal McCloud sediment core indicates that 46

PAGE 54

Table 3-6. McCloud Lake phosphorus budget, August 1981 through July 1982. Precipitation Wet(KgP) Dry (Kg P) Total (Kg P) Mean Storage (Kg P) t. S (Kg P) Residence Time (Yr) 0. 73 1.48-2.19 2.21-2.92 1. 54 1. 94 0 5-0. 7 47

PAGE 55

Table 3-7. Sediment trap results from Mccloud Lake (mean+ standard deviation). Dates 10/23-12/21/82 04/28-06/03/83 06/03-07/05/83 07 /05-08 /04 /83 MEAN ANNUAL RATE *g/m 2 yr. tmg/m 2 yr. Dry Sedimentation, g/m 2 mo 25.9 + 0. 78 11.8 + 4.98 54.6 + 1.07 66.5 + 0. 54 36. 9 + 22. 7 429* TP, mg P g dry wt 1. 024 + 0. 064 0. 844 + 0. 084 0.802 + 0.095 0. 724 + 0.131 0.849 + 0.127 Phosphorus Sedimentation, mg/m 2 mo 26.5 + 0. 78 10.0 + 4.20 43. 8 + o. 86 48.2 + 0. 39 31.0 + 15.3 378.St 48

PAGE 56

~--~----------------the recent annual sediment accumulation rate is about 300 g dry wt/m 2 yr (personal communication, Michael Binford, Florida State Museum). The sediment trap estimate is 43% greater than the 2 10pb estimate, but this difference is probably within the range of hori zontal variation within the lake and annual variations due to changing lake levels. The sediment trap measurements were obtained during a lake level rise of 'v 2 ft, when plant litter from formerly terrestrial areas would be incorporated into the expanding littoral zone of the lake. Thus pelagic sedimentation rates would logically be higher dur ing rising water levels than when lake level remains fairly constant. Second, although the annual atmospheric phosphorus loading rate (44-58 mg P/m 2 yr) is much lower than the measured phosphorus sedi mentation rate (370 mg P/m 2 yr), the trap results provide a gross annual rate that is applicable only to an undefined part of the pelagic zone. A correction can be applied to account for phosphorus release from the sedimented particulate matter, based on the difference between TP in surface sediments and TP in particulates retained in the traps. Surface sediments contain about 0.034% phosphorus (dicussed later in this chapter), while the sedimenting material contained approximately 0.085% phosphorus. Thus 60% of the phosphorus in sedimenting material appears to be returned to the water column. Net phosphorus sedimenta tion therefore would be 'vl50 mg/m 2 yr, which is 2.6 to 3.4 times the atmospheric phosphorus input. The third observation that indicates sedimentation rates are higher in the deeper areas of the lake relates to patterns of sediment accumulation in the McCloud Lake basin. Extensive emergent and sub mergent macrophyte communities are found along the shore and in the 49

PAGE 57

shallow littoral areas, although very little organic sediment accumula tion is noted there. On the other hand, pelagic sediments are as much as several meters thick (personal communication, Michael Binford) and highly organic (see Chapter 6). Therefore sedimentation rates measured in the middle of McCloud Lake should not be applied to the lake as a whole. At a minimum, these data apply to the area of lake bottcm where water depth 2_ water depth at which the traps were located (~4.5 m). At a maximum they represent the area of thick organic sediment accumula tion (water depth 2. ~2-2.5 m). Phosphorus Compartments Macrophytes. Figure 3-5 shows the extent of coverage by macro phyte beds in McCloud Lake on September 30, 1982. This figure deline ates the areas of relatively uniform and continuous coverage by Websteria and Eleocharis. Although emergent species (e.g., Leersia, Fuirena, and Xyris) predominated shoreward of the submergent beds, sparse Websteria stands commonly were observed among these emergents. Likewise, Eleocharis beds did not end abruptly at the deeper border indicated in Figure 3-5. The density of the Eleocharis bed decreased with depth until coverage was no longer complete, but scattered "clumps" of various sizes were observed even in the deepest areas of the lake. This figure and subsequent discussions therefore represent conservative estimates of the importance of Websteria and Eleocharis in McCloud Lake. These two species covered about 14,000 m 2 or 26% of the total lake bottom. Eleocharis comprised the majority of macrophyte coverage (93%), and as Table 3-7 shows, this species also contained a much larger proportion of phosphorus than Websteria. Although the density 50

PAGE 58

0 DISTANCE IN METERS 25 50 83 WEBSTERIA 945 rrt E3 ELEOCHARIS 13180 100 Figure 3-5. Extent of submerged macrophyte coverage in McCloud Lake, September 1982. 51

PAGE 59

Table 3-8. Physical characteristics and phosphorus content (mean and standard deviation) of submergent macrophytes in McCloud Lake, September 30, 1982. Dry Weight (g/m 2 ) Volatile Solids (g/m2) Phosphorus (mg/g dry wt) *Not determined. Eleocharis 39.8 + 28.5 25. 6 + 17. 6 7.27 + 1.01 Websteria ND* ND* 1.45 + 0.033 52

PAGE 60

of Websteria beds was not measured separately, these plants tended to be much shorter and to provide a sparser cover than the Eleocharis. Eleocharis therefore constituted a much larger in-lake phosphorus stor age than did Websteria. Using average values from Table 3-7, the total Eleocharis biomass contained about 3.81 Kg of phosphorus, which was 2.5 times the amount of dissolved and particulate phosphorus present 1n the water column of the lake at that time. The significance of macrophyte phosphorus storage to water column processes depends on the source of phosphorus which these plants utilize. If they obtain most of their phosphorus from the water, then Eleocharis and Websteria compete with planktonic primary producers, but if the sediments supply their phosphorus, the macrophytes represent a significant mechanism of sediment phosphorus mobilization. Carignan and Kalff (1980) used 32 P-labeled sediments to determine whether nine rooted aquatic macrophyte species obtained their phosphorus from sediments or water during in situ incubations in Canadian lakes. They found that sediments were the only significant source of phosphorus to the macrophytes in oligotrophic and mildly eutrophic lakes that had relatively high interstitial phosphorus concentrations. Barko and Smart (1980) obtained similar results with laboratory incubations of three species of submersed macrophytes with minor root systems. They further concluded that release of phosphorus from these species to the water column was a result of tissue decay instead of excretory proces ses. It thus appears that Eleocharis and Websteria do not compete with phytoplankton for phosphorus in McCloud Lake. However, based on the phosphorus stored in these macrophytes, they represent a potentially 53

PAGE 61

important mechanism for returning sediment phosphorus to the lake when they senesce and die. Sediments. Surface sediment in the center of Mccloud Lake is highly organic with a large proportion of water (Table 3-8). The phos phorus content of this surficial sediment is about 0.34 mg/g dry weight. If these sediment characteristics are typical where lake depth is 8 feet or greater (the approximate extent of macrophyte beds), the upper 1 cm represents a storage of approximately 7. 74 Kg of phosphorus. This is about 5 times the mean phosphorus storage in the lake water (1.54 Kg P), and 2.7 to 3.5 times the annual total atmospheric phos phorus loading to the lake surface. Phosphorus Uptake Results of the phosphorus uptake experiments using the two major groups of primary producers in Mccloud Lake are summarized in Table 3-9. Even though mid-lake seston showed a first-order uptake rate con stant (k) nearly twice the magnitude of the k value for Websteria, the variability in seston uptake rendered the two mean k values statistic ally indistinguishable (t-test, a< 0.05). The fact that rooted sub mergent macrophytes appear to obtain most of their phosphorus fran the sediments indicates that there is no competition for phosphorus between the macrophytes and planktonic algae. Summary The pH of McCloud Lake has decreased from 4.85 to about 4.55 over the past 15 years (nearly a doubling of H+ concentration), although this has not been accompanied by a reduction of TP, chlorophyll-a, or other nutrient species. The lake exhibits TP, chlorophyll-!!_, and 54

PAGE 62

Table 3-9. Physical characteristics and phosphorus content of surfic ial mid-lake sediment from McCloud Lake. Parameter Mean Standard Deviation W~ter, % 93. 03 0.18 Volatile Sol ids, % 77. 7 1.00 Phosphorus, mg P/g dry wt 0.343 0.027 Phosphorus, % 0.034 0.003 Interstitial TP, mg/L 0.045 0.004 55

PAGE 63

Table 3-10. Phosphorus uptake rate constants (mean+ standard devia tion) for submergent macrophytes and mid-lake seston from McCloud Lake. Component Mid-Lake Seston Websteria sp. t = 1.os; t.os,s = 2.s1. Uptake Rate Constant (k), hr-1 0.486 + 0.364 0. 257 + o. 064 n 4 3 56

PAGE 64

phytoplankton and zooplankton communities that are typical of oligo trophic Florida lakes. Nitrogen-to-phosphorus ratios indicate that primary production is limited by phosphorus. Total phosphorus shows increased concentrations during late spring and summer, but SRP trends are not evident. Nutrient levels during 1980--1982 appear to be related to rainfall patterns and variations in lake stage This trend is cons istent with the finding (Baker 1984) that rainfall to the lake surface accounts for 90% of the total annual water input. Furthermore, atmospheric phos phorus deposition to McCloud Lake appears to approximate the minimum phosphorus loading rate required to sustain mesotrophic conditions, although the lake is oligotrophic. This suggests that McCloud Lake's low pH inhibits water column productivity, or that phosphorus removal mechanisms are faster than in the lakes used to develop nutrient load ing criteria. Rooted submersed macrophytes constitute a significant in-lake storage of phosphorus that is approximately 5.0 times the mean water column phosphorus storage. It thus appears that the littoral zone contributes much of the primary production in McCloud Lake, although the macrophytes do not compete with phytoplankton for water column phosphorus. Dense periphytic growths, which were counnon on the macro phytes, may minimize the role these macrophytes play in recycling sedi ment phosphorus to the water column. The following three chapters discuss the results of experiments designed to test the effect of lake pH on processes that contribute to removal or recycling of water column phosphorus 57

PAGE 65

CHAPTER 4 EFFECT OF PH ON PLANKTONIC PHOSPHORUS DYNAMICS There is ample evidence to suggest that variations 1n aquatic pH are accompanied by changes in phytoplankton and zooplankton conmunities and by changes in phosphorus dynamics in the pelagic environment. However, it is difficult to know how these processes relate to each other. Plankton community changes related to acidification could theoretically affect phosphorus cycling, but conversely pH-related changes in phos phorus availability could also affect planktonic community structure. Rates of phosphorus uptake by algae and bacteria generally increase as cell size decreases (Lean and White 1983), and phosphorus regeneration is faster for small zooplankters than for large species (Johannes 1965; Hargrave and Geen 1968; Peters and Rigler 1973). Thus pH-related trends in body size of phytoplankton and zooplankton communities should alter phosphorus dynamics. On the other hand, the activity of extracellular enzymes used by algae in assimilating phosphorus is influenced strongly by pH. Thus decreased pH could favor phytoplankton species which produce phosphat ase enzymes that are active at the new pH value. This scenario would have the greatest potential impact on phytoplankton community structure in phosphorus-limited lakes.

PAGE 66

A series of experiments was designed to test the effect of pH man ipulation on phytoplankton and zooplankton cormnunity structure and on sestonic rates of phosphorus uptake and turnover. Materials and Methods Mesocosm (Limno-Enclosure) Experiments Littoral mesocosms. Three polyethylene enclosures were con structed according to Landers (1979) and installed in the littoral zone (1 m depth) of McCloud Lake in February 1981. Each mesocosm enclosed a 12-m 2 water column, and the polyethylene material was inserted into the sediment and secured with wooden stakes and rope to ensure a good seal. Before pH adjustment began, the mesocosms were allowed to equilibrate for 4 weeks. The pH of enclosure A was decreased to 3.6 over a 4-week period with 1.0-L additions of 0. 7 N HzS04, while the pH of enclosure B was raised in the same manner to >5.1 with 1.0-L additions of 0.1 to 0.4 N NaOH. Further acid and base additions were made as necessary to maintain the desired pH ranges, although enclosure B never reached the intended pH of 5.6 because of buffering by the sediments (Baker 1984). Enclosure C was left at the ambient pH (4.6 + 0.1) throughout the experiment. These mesocosms were sampled on a weekly basis to follow changes in the littoral chemistry and biology resulting from pH alteration. In addition, a littoral lake station adjacent to the enclosures was sampled on the same schedule. Chemical and biological analyses were performed as described in Chapter 3. Mid-lake mesocosms. Six enclosures were placed in the middle of the lake in July 1982 to evaluate the effects of acidification and 59

PAGE 67

nutrient addition on phosphorus dynamics and phytoplankton and zoo plankton communities. Two groups of three enclosures were installed 1 week apart. These mesocosms were 0.92 min diameter, 2.2 m deep, and each contained 1.2 m 3 of depth-composited lake water (added with a gasoline-powered pump). The pH treatment in these enclosures consisted of duplicates at each of three values: Ml and M4 were left at the lake pH of 4. 7; M2 and MS were lowered to 4.1 with 0.1 N H 2 S04; and M3 and M6 were first lowered to 4.1 and then further acidified to 3.7 1 week later. No additional pH adjustment was required since the enclo sures were isolated from the sediments. During the tenth week after pH adjustment, NH4-N03 and KH2P04 were added to enclosures 1-3 to increase TN and TP each by a factor of about 10. These enclosures and a mid-lake station were sampled twice each month from the end of July through November 1982 using the previously described methods for sample collection and analysis (Chapter 3). Radiolabeled orthophosphorus ( 3 2p) was used to measure sestonic uptake and turnover of phosphorus in the mid-lake enclosures. One-liter samples of seston (unfiltered water) from each enclosure were transported to the laboratory and incubated at room temperature (22C + 2C) in clear glass bottles. Fluorescent lighting was provided from above, and magnetic stirrers slowly mixed the contents. Aliquots of a stock solution containing KzHP04 as a carrier for 32 P04 (obtained from New England Nuclear) were added to the seston samples, and uptake of added 32 P was followed by periodically filtering 5-mL subsamples (0.45 m membrane filters). The volume of stock added in each experi ment was adjusted according to its specific activity, but the range was 25 L to 1000 L stock/L sample. The stock solution contained 3.1 g 60

PAGE 68

P/m.L, and the activity introduced to each sample yielded approximately 12,00G-60,000 counts per minute (CPM) in the 5-mL subsample. The filter and filtrate were stored in separate scintillation vials until they were counted on a Packard Tri-Carb model 4530 liquid scintillation counter. Counts were corrected for background activity and decay so that results corresponded to the time at which samples were collected. Total SRP was calculated as SRP originally in each enclosure plus the amount added in the laboratory. A similar design was used to measure directly the release of phos phorus by seston from the mid-lake enclosures. A 2-L sample was col lected from each enclosure and transported immediately to the labora tory. The seston was concentrated by a factor of 10 by vacuum filter ing all but a small volume (~20 mL) through 0.45 m membrane filters. This concentrated volume was transferred to an Erlenmeyer flask, and seston retained on the filter was resuspended by vortexing three times in 10 mL of filtered enclosure water. The resuspended seston was added to the Erlenmeyer and additional filtered water was used to obtain a final volume of 200 m.L. Stock radiophosphorus was added and the flasks were incubated under fluorescent lighting for 24 h to allow uniform labeling of the seston. After the incubation period, the seston was again concentrated by a combination of centrifugation and vacuum fil tration of the supernatant. The concentrated, labelled seston (~10 mL) was added to unlabelled filtered enclosure water to obtain a volume of 500 mL. Phosphorus release was followed by periodically filtering 5-mL aliquots (0 45 m membrane filters), and storing filter and filtrate in separate scintillation vials for later analysis with the liquid scin tillation counter. 61

PAGE 69

Several methods have been used for quantifying phosphorus uptake and turnover. Many 32 P users have adopted a method described by Zilversmit et al. (1943) for calculating uptake rates and turnover times in radioisotope experiments. The method necessitates three assumptions: 1) Steady state conditions 1n which the rate of appearance of the isotope equals its rate of disappearance; 2) Constant rate of appearance and disappearance; and 3) Random appearance and disappearance of the element and its iso tope. They further define: p = rate of disappearance of B from fluid; x = amount of radioactive B in the fluid at any time; r = total amount of B present in fluid (assumed to be constant); and tt = turnover time, the time required for the tissue to com pletely remove and replacer. The change in x with time is given as dx/dt = -p(x/r). (EQ 1) Integrating EQ 1 yields (EQ 2) and taking natural logarithms, EQ 2 becomes ln x/r = ln c p/rt (EQ 3) 62

PAGE 70

Equation 3 describes a straight line with slope= -p/r which in turn equals -1/tt. Therefore, from a plot of ln x/r versus time and knowledge of r, both uptake rate and turnover time can be calculated. Zilversmit et al. (1943) cautioned that their method of uptake and turnover calculations was valid only during the time interval in which none of the radioisotope is returned from the tissue to the fluid. Fast initial uptake of SRP (and thus quick turnover) coupled with the difficulty of measuring accurately the initial SRP concentration led Lean and White (1983) to reco11DI1end that first-order uptake rate con stants (k) should be used instead. The value of k (time-1) can be obtained as the slope of a plot of percent isotope in the filtrate (or the filter) versus time Diurnal productivity estimates were made in conjunction with both mesocosm studies by following diel DO changes in the water columns of the enclosures with a YSI model 57A dissolved oxygen meter. Oxygen measurements in the littoral enclosures were made at 2to 5-h inter vals over 24-h periods, with shorter intervals used in early morning and early evening, when DO changes occurred most rapidly. Dissolved oxygen was measured every 3 hover 24-h periods in the mid-lake enclo sures. Oxygen changes in the open water of both sets of enclosures were corrected for diffusion across the air-water interface with a dif fusion coefficient experimentally detennined for the lake using the dome method of Copeland and Duffer (1964). Gross primary production (P) and respiration (R) were determined by planimetry frooi plots of corrected rate of DO change over time as described by Odum (1956) and numerous others (Odum and Hoskin 1958; Hall and Moll 1975) for measure ment of community metabolism. 63

PAGE 71

Laboratory Microcosms An experiment similar to the mesocosms involved microcosms set up in the laboratory in March 1983. Three 12-L glass carboys were filled simultaneously by siphoning from a continuously stirred 40-L container of depth-composited lake water. The microcosms were aerated slowly to provide mixing, and overhead fluorescent lighting was timed to mimic the natural day length. One microcosm was left at ambient pH (4.60), and 1 N HzS04 was used to achieve pH values of 4.0 and 3.6 in the other two. SRP, TP, and total dissolved phosphorus (TDP) were measured periodically using previously described methods. Because of an accidental shift in TN/TP ratios, this experiment was terminated 4 weeks after pH adjustment. Three new microcosms were set up 1n the same manner in May 1983. Two weeks after pH adjustment, NH4-N03 and NaHzP04 solutions were added to increase TN and TP to ~1.0 mg/Land 0.1 mg/L, respec tively. TP, TDP, and SRP were measured periodically before and after nutrient addition. The activity of extracellular acid phosphatase enzymes was measured in the microcosm experiments by a fluorometric method developed by Swedish limnologists (Petterson and Jansson 1978; Jansson et al. 1981; Jansson 1981). The procedure involves addition of a buffered fluorogenic substrate (4-methylumbelliferyl phosphate, MUP) to the water sample. Phosphomonoesterase activity is calculated from the rate of liberation of fluorescent hydrolysis product 4-methylumbel liferone (MU). Fluorescence of MU was determined with an American Instrument Co. SPF-125 spectrofluorometer using an excitation wave length of 320 nm and an emission wavelength of 450 nm. A stock solu tion of 102 M MUP was prepared in autoclaved distilled water and 64

PAGE 72

frozen in 1-mL crimp-s e al vials until needed. Working MUP solutions (lo4 M) were obtained by dilution of the stock in 0.1 M acetate buffer at e ach microcosm pH (4.6, 4.0, and 3.6). For the assay, 0.5 mL of working MUP solution was added to 4 mL of the test water. Thus the test could be run at th e pH of e ach sample, or all microcosms could be tested at pH 4.6. The amount of fluorescent MU released was quantified by comparison lo standard MU solutions prepared in acetate buffer. Results and Discussion Littoral Mesocosms Figure 4-1 shows variations in soluble reactive phosphorus in the littoral enclosur e s over 15 weeks (including 4 we e ks of pH adjustment but not the initial 4-week equilibration after installation). Figure 4-2 shows total phosphorus variations over the same period. Peaks in SRP occurred in all three enclosures during initial pH adjustment, although the increase was smallest in the control enclosure (C) left at ambient pH. Thereafter, enclosure B (pH= 5.1) often showed the high est SRP concentration, although no consistent trend was evident. Total phosphorus concentrations also were high initially, and then varied in the range 1-10 g/L after the first 4 weeks. With few exceptions, TP was consistently higher in the high pH enclosure (B) than in the acidi fied one (A), after initial pH adjustment. Table 4-1 summarizes mean nutrient concentrations in the littoral enclosures and at the littoral lake station from late March to July 1981. TP shows a trend apparently related to pH, with the highest mean in the base-treated enclosure (13 g/L), the lowest in the acid treat65

PAGE 73

~ ------------------------------------25 :J 20 -OI :::1.. en :::::, a: 0 I a. en 0 I a. w > i== (J < w a: w -I CD :::::, -I 0 en 15 10 5 0 I 3 I 2 6 MAR 81 5 7 9 WEEKS 6 0 II A (3.7) B (5 1} C (4.6) 13 1 5 I 1 JUL 81 Figure 4 -1 S olu bl e r eact i ve ph osp h orus variatio n s in l i ttoral meso cosms. 66

PAGE 74

---~-----------------..J -....... Cl :l. (JJ ::, 0 a: 0 :c Q. (JJ 0 :c Q. ..J < t0 t60 50 40 30 I 3 5 7 9 I 26 MAR 81 WEEKS C:. 0 II A (3.7) 8 (5.1) C (4.6) 13 I 15 24 JUNE 81 Figure 4-2. Total phosphorus variations 1n littoral mesocosms. 67

PAGE 75

Table 4-1. Nutrient means (g/L except TN/TP) 10 the littoral enclo sures. Enclosure Littoral Parameter A B C Lake SRP 4 4 8 3 TP 7 13 9 12 TN/TP (wt/wt) 54.6 31.0 36.0 45. 5 TON 281 357 258 442 NH4-N 49 27 29 50 N03-N 52 27 37 54 N02-N l 1 1 1 68

PAGE 76

ment (7 g/L), and 9 g/L in the ambient pH enclosure. The means are not significantly different (ANOVA a > 0.05); how e v e r analysis of the average TP di f ferences (calculated for each sampling date) between pairs of enclosur e s (paired -dif ference t-test) shows that TP was significantly higher in enclosur e s Band C than in the acidifi e d enclo sure (TPB TPA, a = 0.0396; TPc TPA, a = 0.0138). The mean difference between Band C is not significant ( a> 0.05) Mean TN/TP ratios in the littoral enclosures ranged from 31.0 in the base treatment to 54.6 in the acid treatment, whil e in the littoral lake the rati o was 45.5 (Table 4-1). Th e se valu e s all indicate phos phorus limitation (TN/TP > 30 according to Huber et al. 1982), and they are similar t o mean annual ratios for the mid-lake station (Table 3-1). The trend of increasing TN/TP values as pH decreases suggests that phosphorus becom e s more limiting as pH is lowered. This reflects the fact that TN variations in the littoral enclosures were insignificant, while TP decreased with decreasing pH. Although TN means did not vary significantly in these enclosures, means of some of the nitrogen forms did suggest a pH effect TON levels did not appear to be related to pH, and the differences among enclosure means were not significant CANOVA, a> 0.05). However, ammonium and nitrate means were slightly higher in the littoral lake than in the enclosures. Both ions tended to increase as enclosure pH decreased, which could indicate increased mineralization or decreased utilization rates, although the differences b e tween treatments were not significant. Among biological parameters measured in the littoral enclosures, only zooplankton abundance varied significantly. Differences in the 69

PAGE 77

mean values of log-transformed chlorophyll-~ and total phytoplankton abundance were not significant (ANOVA, a > 0.05), although total phyto plankton means did decrease with decreasing pH. Means of total zoo plankton and cope pod abundances (log-transformed) al so dee reased with pH. In both cases the differences between enclosures Band C were not significant, while the values in enclosure A were significantly lower (Duncan's Multiple Range Comparison Test, a= 0.05). The variations in total zooplankton numbers were due to the reduction in copepods as pH decreased, as evidenced by the fact that cladoceran and rotifer abund ances did not show a treatment effect. Mid-Lake Mesocosms Nutrient trends. Figure 4-3 shows TP variations in the unfertil ized mid-lake enclosures. TP stayed relatively constant in Ml (con trol) while it increased in acidified enclosures M2 and M3. In the other set of enclosures however, TP was consistently lower in M6 (pH= 3. 7) than in the higher pH enclosures M4 and MS. TP trends were not similar in the control enclosures (Ml and M4). Table 4-2 summarizes the mean nutrient concentrations in the mid-lake enclosures by pH treatment. Total phosphorus was lowest at pH 3. 7 (6 g/L), and was nearly the same at 4.1 (10 g/L) and 4. 7 (9 g/L). According to Duncan's Multiple Range Test (a= 0.05) the two higher TP means did not differ significantly, while both were statistically higher than TP at the low pH of 3. 7. SRP concentrations were consistently low in these enclosures and mean SRP was 2 g/L for all three pH treatments. TN/TP ratios in the mid-lake enclosures (Table 4-2) were lower than in the littoral enclosures or the lake, and they were indicative 70

PAGE 78

..J ..... ti ::t. I..J ..... C) ::t. 15 10 5 a 15 10 0 M1 (4.6) 0 M2 (4.1) M3 (3.7) AUG SEP OCT NOV M4 (4.6) 0 MS (4.1) c::. Me (3.7) AUG SEP OCT NO V Figure 4-3. Total phosphorus in mid-lake mesocosms excluding period of nutrient addition. 71

PAGE 79

Table 4-2. Nutrient means (g/L except TN/TP) for mid-lake enclosure pH groups (excluding data from Ml to M3 after nutrient addition. pH Group Parameter 4.6 4.1 3. 7 SRP 2 2 2 TP 9 10 6 TN/TP (wt/wt) 15.4 11.5 17. 3 NH4-N 11 11 13 N03-N + NOz-N 12 7 8 TN 139 115 104 72

PAGE 80

of nutri e nt-balanc e d conditions according to criteria propos e d for Florida lakes (10 < TN/TP < 30; Huber e t al. 1982). \<.lhil e th e lowest pH enclosures did show th e highest TN/TP ratio (17.3), th e inverse relationship found b e tw e en TN/TP and pH in th e littoral enclosures was not seen in th e mid-lake mesocosms. Total nitrogen in the unfertiliz e d mid-lak e enclosures was highest at the ambient pH (139 g/L) and decreased with enclosure pH. ANOVA showed a significant treatment (pH) effect ( a < 0.05), and Duncan's Multiple Range Test yielded the following relationship among mean TN values: pH 4. 7 pH 4 .1 pH 3. 7 (TN means at underlined pH values are not significantly different, a = 0.05). Means of ammonium and nitrate/nitrite were all less than 15 g/L, and no pH-related trends were evident for these nitrogen forms. Figure 4-4 illustrates the effect of nutrient addition on th e phosphorus fractions in mid-lake enclosures Ml, M2, and M3. SRP removal rates were similar in all three pH treatments, and SRP concen trations decreased to pre-fertilization levels within 13 days after nutrient addit ion. Particulate organic phosphorus (POP) also responded in a similar manner at the three pH valu e s POP reached maximum levels ('v 30 g/L) about 2 weeks after nutrient addition, and had returned to pre-fertilization concentrations within 33 days. The response of solu~ ble organic phosphorus (SOP) did indicate a pH effect, however. SOP in Ml (pH 4.6) and M2 (pH 4.1) stayed below 10 g/L after nutrient addi tion, but at the lowest pH (3.7), SOP increased to nearly 30 g/L 73

PAGE 81

50 40 ':! ........ c,t 30 a. 20 0 a. 10 0 !50 40 ':! ....... 30 c,t ,3. a. 20 0 (/J 10 0 100 80 c,t 60 :l a. 40 cc (/J 20 0 0 10 0 10 0 1 0 20 20 20 DAYS 30 30 e M1 (4.7) Q M2 (4 1) 6 M3 (3 6) 30 40 40 40 Figure 4-4. Changes in phosphorus forms after nutrient addition to Ml, M2 and M3 74

PAGE 82

These results suggest a reduced ability of the biota to utilize SOP at a pH of 3. 7. Biological trends. Total phytoplankton densities initially decreased in all six mid-lake enclosures, but in the intermediate pH (4.1) enclosures, phytoplankton then began to increase relative to the other enclosures. After nutrient addition to Ml, M2, and M3, phyto plankton in both pH 4.1 enclosures (M2 and MS) declined to levels com parable to those in the other enclosures. The addition of nutrients was not followed by an increase in phytoplankton numbers in Ml, M2, and M3, although chlorophyll levels did reflect the increased nutrient concentrations. This phenomenon was apparently the result of intensi fied zooplankton grazing in the fertilized mesocosms, which kept phyto plankton numbers low while production was high. Chlorophyll-_! concen trations did increase because the analysis included material extracted from phytoplankters and zooplankters. Total zooplankton populations also increased markedly during the same period in the fertilized enclosures. Species composition of phytoplankton and zooplankton communities responded to increased acidity in the mid-lake enclosures. Ultraplank ton (1-10 m) comprised 8~100% of total phytoplankton numbers at the lake pH (4.6), but decreased in importance at lower pH values. The green alga Oocystis gloeocystiformis increased in abundance at reduced pH in the unfertilized enclosures, while after fertilization, Crypto monas marsonii dominated at all three pH values. Zooplankton response to reduced pH was similar to that seen in the littoral enclosures. Copepod numbers (predominantly Diaptomus missis sippiensis) decreased markedly as pH was lowered, so that this group 75

PAGE 83

accounted for l ess than 1 % of total z oop lankton in the pH 3. 7 m e so cos ms shortly a fte r initial pH adjus t m e nt. The acid-tol era nt cladoceran genera Eubosrnina and Diaphanosoma becam e incr e asin gly important at more acidic pH values. Rotifer p e rcent composition in the low pH e nclosures increased shortl y after pH r e ducti o n, but after about 6 mor e w ee ks, their imp o rtance d ecre ased to lev e ls comparable t o the hi g h e r pH e nclosures. In contrast to phytoplankt o n composition, zooplankton species composition was not affected by nutrient addi tion. Radiophosphorus ex p e riments. Phosphorus uptak e by seston in the mid-lake enclosures was measured on six dat es using radi ol ab e ll e d phos phorus in the laborator y In Figur es 4-5 through 4-10, uptak e o f 32 P is repres e nted as the increase in radioactivity on filters versus tim e durin g the incubations. (Not e variations in abscissas and ordinates on diff erent dat es ) The first two dat e s (Figures 4-5 and 4 6) were bef o r e nutrient addition to Ml-413, while the last four wer e after the addition. Missing data for Ml, M4, and MS in Figures 4-9 and 4-10 (the last two dat e s) were because bird droppings had caused increases in pH and TP in those enclosures. Mesocosm Ml (pH 4.6) showed the fastest uptake of 3 2p befor e nutri e nt addition but M6 (pH 3.7) consistently ex hibited th e fastest uptake aft e r the addition. Table 4-3 gives uptake rat e s and turnover times calculated accord ing to Zilversmit et al. (1943) from the data presented in Figures 4-5 through 4-10. Uptak e rat es ranged from 0.1 to 12 ~g/Lh, while the range of turnover times was 0.3 h to 12.2 h. It is inter esti ng to note that the unfertilized mesocosms with the low es t TP values (Ml before nutrient addition, and M6) showed the fastest uptak e rates. 76

PAGE 84

----------------------------------77 25 20 'o 15 ... >( 5 a. C a: w I10 i -' u: 4 5 ---------:3 0 I 120 240 TIME (min) Figure 4-5. 32 p uptake by mid-lake enclosure seston, 8/24/82.

PAGE 85

,.., 0 ..... X ::E Q. C cc w _J u: 60 50 40 30 20 10 0 0 120 TIME (min) 6 5 4 Figure 4-6. 32p uptake by mid-lake enclosure seston, 9/14/82. 78 240

PAGE 86

,., 0 .... )( Q. C a: w u: 8 6 4 2 0 0 6 / 6 6 60 TIME (min) Figure 4-7. 3 2p uptake by mid-lake enclosure seston, 11/9/82. 79 5 2 3 4 120

PAGE 87

"b .... )C 40 30 20 10 0 6 6 2 I I I 0 60 120 TIME (min) Figure 4-8. 3 2p uptake by mid-lake enclosure seston, 11/11/82. 80

PAGE 88

'"b ,.. )( C. C a: w ... ..J i:i: 20 15 10 5 0 ~ ---------s /" 6 3 5 5~35 6 3 -;::::::: 3 --------2 ~Li-2----~ O 60 I 120 TIME (min) Figure 4-9. 32 P uptake by mid-lake enclosure seston, 11/18/82. 81 I

PAGE 89

12 10 8 "b 9"" )( C 6 a: w ... ...J i! 4 2 0 6 6 6 6 3 2 3~: ~2 6 3 .,.......~!-s----5 ---------s 0 60 TIME (min) 120 Figure 4-10. 32 P uptake by mid-lake enclosure seston, 11/23/82. 82

PAGE 90

Table 4-3. Phosphorus uptake rates and turnover times by seston from mid-lake enclosures (underlined values indicate enclosures with nutrient addition). Date Var. Ml M2 M3 M4 MS M6 08/24 p 12 1.8 0.6 2.4 4.2 2.4 tt 0.4 3.9 10 0 3.3 1. 2 1. 7 09 /14 p 6 3.0 3.6 1.2 2.4 4.2 tt 0.5 1.0 0.8 1. 7 1. 2 0.5 11 /09 p 0.6 4.8 1.8 0.2 2.4 4.2 tt 6.2 1.2 2.0 9.4 0. 7 0.6 11/11 p 0.4 0.2 1.2 0.1 1. 2 1. 2 tt 9.7 9.2 5.8 28.0 3.3 2. 7 11 /18 p 0.2 0.6 0.6 1.8 11. 0 tt 12.2 11.6 2.8 2.2 0.3 11 /23 p 0.5 0.4 6 tt 7.4 5 .6 0.7 p = Uptake rate ( g/L"h). t t = Turnover time (h). 83

PAGE 91

The same procedure was used to calculate phosphorus release rates from the seston concentration experiments, which were carried out on two dates before nutrient addition to enclosures Ml, M2, and M3. These release rates, corrected to reflect the original seston concentrations, are given in Table 4-4. No effect of enclosure pH is evident in these data, although the range of values is comparable to the uptake rates in Table 4-3. Table 4-5 lists the means and ranges of uptake rate constants cal culated by the method of Lean and White (1983) ran the same experi mental data presented in figures and tables above. Results using this method were similar to those of the previous method. In the group con sisting of M4, MS, and M6 there was a clear trend of increasing rate constants as pH decreased. However, this was not repeated in the pre fertilization data for Ml, M2, and M3, where the high pH enclosure (Ml) showed the largest rate constants. ANOVA revealed no significant effect of pH on k values (a> 0.05). It thus appears that phosphorus availability has more effect on the planktonic uptake and turnover of phosphorus than does any direct effect of hydrogen ion concentration. Furthermore, phosphorus availability (as indicated by SRP concentra tions) did not appear to be related to pH in these mesocosms. CoIIDllunity Metabolism The diel oxygen technique was used to obtain two estimates of com munity metabolism in the littoral enclosures, while the mid-lake enclo sures were monitored on three dates. Table 4-5 presents means of these productivity (P) and respiration (R) measurements for the littoral and 84

PAGE 92

Table 4-4. Phosphorus release rates (g/Lh) by seston from mid-lake enclosures. Date 08/27 09/16 Ml 2.0 2. 1 M2 10.2 0.7 Enclosure M3 0.5 3.2 M4 0.7 MS 0.6 2. 1 M6 1. 7 0.5 85

PAGE 93

Table 4-5. Phosphorus uptake rate constants (h1 ) for the mid-lake enclosures. Enclosure pH Mean k Range Ml (Pre) 4.6 0. 74 o. 52-0. 96 Ml (Post) 4.6 0.11 0.09-0.13 M2 (Pre) 4.1 0.25 0.14-0.36 M2 (Post) 4.1 0.25 0.13-0. so M3 (Pre) 3. 7 0.32 0.02-0.61 M3 (Post) 3.7 0.28 0.19-0.37 M4 4.6 0.14 0. 02-0 32 MS 4.1 0.40 0.01-1.30 M6 3.6 1.12 0.17-1.99 86

PAGE 94

mid-lake mesocosms, and a mid-lake station adjacent to the open-water enclosures. The data are expressed on a volumetric basis (g o 2 /m 3 day) to allow comparison of metabolism in connnunities of different depth (Lind and Campbell 1970). P and R means were similar in any given enclosure. Mean produc tivity in the mid-lake mesocosms ranged from 0. 70 to 0.84 g Oz!m 3 day, while mean respiration varied from 0. 75 to 1.01 g o 2 /m 3 day. Average P/R ratios were close to unity in these communities and at the mid-lake station. Two-way ANOVA and Duncan's Multiple Range Comparison showed no significant pH effect on P, R, or P/R ratios for the enclosures, and no significant differences in these parameters between the open lake and any enclosure. Total community P and R were much higher in the littoral enclo87 sures than in the open lake or the mid-lake enclosures. Mean P ranged from 3.8 to 6.1 g 02/m 3 day and the range in R was 3.8 to 7.1 g Ozlm 3 day. The difference between littoral and open lake conmunities probably reflects the contribution of submersed macrophytes and periphytic algae to littoral productivity. It is interesting to note that the higher littoral primary productivity was balanced by a comparable community respiration, so that P/R ratios were near 1.0 for littoral and plank tonic communities. A P/R value near unity is considered indicative of a balanced ecosystem (Lind and Campbell 1970). Two-way ANOVA of the littoral metabolism data showed no significant pH effect ( ~> 0.10 in all cases) on productivity, respiration, or P/R values. Although areal or volumetric rates of primary productivity are higher in the littoral zone of Mccloud Lake than in its pelagic waters, the area represented by both habitats would have to be determined to

PAGE 95

Table 4-6. Volumetric productivity (P), respiration (R), and P/R means in littoral and pelagic mesocosms. Connnunity pH n P* R* P/R Open-water Ml 4 6 3 0. 70 0.75 1.05 Open-water M4 4.6 3 0.82 0.92 1.01 Mid-lake 4.6 3 0.83 1.01 0. 87 Littoral C 4.6 2 6.07 7. 12 0.86 Littoral B 5.6 2 4 98 5.36 0.93 Open-water M2 4.1 3 0.84 0.88 1. 12 Open-water MS 4.1 3 0.76 0.94 0.83 Open-water M3 3 7 3 0 83 0.88 1. 12 Open-water M6 3.7 3 0. 77 1.01 0. 77 Littoral A 3. 7 2 3.82 3.83 1.02 *g 02/m 3 day. 88

PAGE 96

evaluate the overall importance of each zone to total lake productiv ity. Variations in primary productivity are matched by similar changes in respiration, so that P/R ratios remain close to unity. In addition, variation of pH between 5.6 and 3.6 did not significantly affect the metabolism of littoral or planktonic communities originally adapted to a pH of 4. 6. Laboratory Microcosms Total phosphorus concentrations in the unfertilized microcosms were too low (generally 2--4 g/L) to permit determination of a pH effect. Initial TP levels were higher (~6 g/L), but filamentous algae growing on the glass microcosm walls soon reduced the phosphorus avail able to planktonic biomass. SRP uptake after nutrient addition to the second set of microcosms is shown in Figure 4-11. Removal of SRP from the water columns appeared to follow first order kinetics for 5--6 days, but it was essentially linear from that point through day 15. This could be accounted for by a S--6-day lag in the increase of zooplankton numbers, after which zooplankton grazing would encourage a constant rate of algal productivity. There was no indication that microcosm pH influenced the rate of SRP uptake, which was essentially identical at the ambient pH (4.6) and the low pH (3. 7), and was only slightly slower at the intermediate pH. In addition, the shapes of SRP loss curves were the same at all three pH values. After fertilization, these microcosms did not exhibit the SOP increase at pH 3.7 that was seen in the mid-lake mesocosms. The large surface-to-volume ratio in the microcosms increased the importance of 89

PAGE 97

90 0 120 6 4 6 6 4 0 0 3 6 100 80 60 Q. a: "' 40 20 0 0 5 10 15 DAYS Figure 4-11. SRP 1n microcosms after nutrient addition

PAGE 98

attached algae, so that the response of water column TP to fertiliza tion was much less at all pH values than in the mid-lake enclosures. Despite the inherent differences between the microcosm and enclosure experiments, microcosm fertilization did indicate that the SOP increase seen in the pH 3. 7 enclosure was not a universal effect of low pH. Both experiments also showed little or no effect of pH on SRP uptake. Acid phosphatase activity. Table 4-7 presents typical phosphatase assay results from the first set of microcosms. Activity decreased as assay pH decreased, regardless of initial pH of the sample, and the potential activity (all assayed at pH 4.7) decreased as microcosm pH decreased. Activity in the two low pH microcosms was 80--150% higher when assayed at pH 4. 7 than at the ambient microcosm pH. This indi cates that the enzymes were adapted to the ambient pH of the lake, and also suggests that production of the extracellular enzymes decreased at the lower pH values. While pH definitely affected the potential activ ity of the enzymes, many factors can influence production of phosphat ase enzymes. Potential phosphatase activity in the second set of microcosms showed the same pH effect (Table 4-7). Samples frcm the acidified microcosms showed higher enzyme activity when assayed at pH 4.7 than at ambient pH. However, enzyme production did not follow the same pat tern. When samples from all three microcosms were assayed at pH 4. 7 the activity of the pH 3.7 microcosm was frequently equal to or higher than that of the pH 4.7 microcosm, and activity of the intermediate pH microcosm was higher than in either of the other two. These results indicate that some factor (or factors) other than pH affects the pro duction of acid phosphomonoesterase enzymes. 91

PAGE 99

Table 4-7. Effect of pH on phosphatase enzyme activity (n mole/Lmin) in microcosm experiments. FIRST EXPERIMENT SECOND EXPERIMENT Microcosm EH Microcosm EH Assay pH 4.7 4.0 3.7 4 7 4.0 3.7 4 7 30.2 16.1 9. 1 14.5 24.1 15. 1 4.0 6.1 9.3 3. 7 5 1 7.8 92

PAGE 100

CHAPTER 5 EFFECT OF PH ON PHOSPHORUS RELEASE DURING PLANT DECOMPOSITION Grahn et al. (1974) hypothesized that acidification of fresh waters causes reduced rates of organic matter decomposition and thus slower rates of nutrient remineralization. Experiments to test this hypothesis have taken several forms. Most researchers have examined the effect of pH on loss of particulate or soluble organic substrate, and little attention has been focused on the release of nutrients during decomposition of naturally occurring particulate organic matter. This chapter presents experiments designed to follow release of soluble reactive phosphorus from plant matter decomposing at different experi mental pH values. Because the submergent macrophytes Websteria sp. and Eleocharis sp. represent a large reservoir of phosphorus within the lake, they were used as the organic substrate in these experiments. Experimental Methods Live, freshly collected Eleocharis or Websteria plants were used 1n all experiments in order to simulate as closely as possible the conditions of decomposition in McCloud Lake. Intact plants were collected from the littoral zone, brought to the laboratory, and gently washed to remove attached algae from the leaves and organic sediment from the roots. The plants were blotted dry with paper towels to obtain fresh 93

PAGE 101

weights. Representative samples were dried and digested at the begin ning of each experiment to obtain water and phosphorus contents. Preliminary Experiments 94 In the first experiment Websteria plants were blotted dry and 2.0 g (+ 0.01 g) of intact plant material were added to each of 18 glass stoppered, 300-mL clear glass bottles. The bottles were overfilled with filtered lake water (0.45 m membrane filters) to eliminate air. No bacterial seed was introduced, as it was assumed that sufficient bacteria were present on the plants. Additions of 1 N H2S04 or NaOH were used to obtain three rep licate bottles at each of six pH values (Table 5-1). The bottles were incubated 1n a dark BOD incubator at 20C (+ 1C) for 170 days to allow a natural senescence and decay of the plants. Samples ( "'5 mL) were removed and filtered for SRP analysis about every 10-15 days, and initial and final pH values were measured. The sample volume was not replaced, and a noticeable sulfide odor after 30 days indicated anoxic conditions. Some oxygen was introduced during the sampling process, but the bottles remained unstoppered for a minimal length of time (usually 30 s/bottle). The second preliminary experiment involved intact, live Eleocharis plants incubated 1n aerated glass bottles maintained in the dark at pH 3.7, 4.8, or 5.5 (Figure 5-1). Based on the first preliminary experi ment, it was not anticipated tht continued pH adjustment would be necessary. However, the pH began to increase 1n all three groups after 1 month of incubation After 6 weeks of incubation the pH in all bottles was near 7.0, and the experiment was terminated.

PAGE 102

Table 5-1. Initial and final pH values in first preliminary decomposi tion experiments. Addition 2 mL Initial Final Group 1 N NaOH 1 NHzS04 pH pH B 2.0 11.50 8.37 C 5.61 3.93 Al 0.02 4 94 4.07 A2 0.20 3 35 4.04 A3 2.0 2.31 2.37 A4 4 0 2.02 2.08 95

PAGE 103

A. ELEOCHARIS PLANTS 800 ml WATER 8 GANG VALVE .. 00 0 00 0 0 0 0 TQ AIR FILTERS AND AIR SUPPLY 4 L WATER. Figure 5-1. Incubation set-ups in aerated decomposition experiments (A. preliminary; B. final). 96

PAGE 104

Final Experimental Design The final decomposition experiment was carried out under similar conditions. Freshly washed Eleocharis plants were blotted dry and 0.5 g (+ 0.01 g) of intact plants were added to each of eight experimental bottles. These 1-gal screw-cap plastic Nalgene bottles were filled with 4 L of lake water that had been passed through glassfiber filters to remove large particulates while allowing small planktonic forms to be introduced. Additions of 1 N H 2 S04 or NaOH were used to achieve duplicate bottles at four pH values: 3. 7, 4.1, 4. 7, and 5. 7. After pH adjustment, the bottles were incubated at room tempera ture under fluorescent lighting (ambient light-dark cycle) for 1 week to allow the plants and bacterial populations to respond to the new pH conditions. At that time plastic screw caps were installed and the bottles were transferred to a dark storage cabinet. Aeration was pro vided from a glass pipette inserted through a hole in the cap of each bottle (Figure 5-1), in order to ensure oxygenated conditions and a well-mixed system. The pH of these systems was checked weekly, and acid or base was added as necessary to maintain the desired pH values. Aliquots were removed periodically for SRP, TP, and TDP analysis using previously described methods. TP and TDP were followed for the first 44 days, and release of SRP was followed for 227 days. 97

PAGE 105

----------------Results and Discussion Preliminary Experiments Table 5-1 shows the initial and final pH values in the un-aerated decomposition bottles from the first preliminary experiment. The con trol group (C) received no pH adjustment; NaOH was used to increase the pH in group B; and the symbols Al through A4 indicate increasingly lower pH values. The phosphorus content of the Websteria plants was 0.15% initially (Table 5-2). Figure 5-2 shows changes 1n SRP in those decomposition bottles during the 170-day incubation period. A rapid initial release (~10 days) of SRP occurred at all pH values except the two closest to the lake pH (Group C, pH 5.6; and Group Al, pH 4.94). This suggests that pH adjustment killed the plants at pH 11.55 (B) and pH <3.35 (A2, A3, and A4), resulting in rapid release of cellular SRP to the water. This release was followed by a period of uptake (to day 41) in A2, A3, and A4, which appears to correspond to an increase in heterotrophic bio mass. Those bottles exhibited extensive fungal and/or bacterial growth on the decomposing plants (Table 5-3). SRP in A2 and A3 stayed rela tively low during the remainder of the incubation, although it did fluctuate between 10 and 60 g/L. The lowest pH (A4) showed a second SRP increase after day 41. At the highest pH (B) SRP continued to increase for 55 days, after which the concentration decreased. Never theless, SRP was consistently higher than 150 g/L at the highest pH after day 10. SRP did not begin to increase in the bottles without pH adjustment (Group C) until day 41, but it reached 100 g/L by day 94. In the 98

PAGE 106

~ ----------------------------------Table 5-2. Water and phosphorus content of plants used 1n decomposi tion experiments. Experiment Plant 1st Prelirr 1nary Websteria 2nd Prelirr 1nary Eleocharis Final Eleocharis *Mean+ standard deviation. tn = number of observations i. HzO in* fresh wt 93. 0 + o. 4 89. 7 + 0.8 88.9 + 1.1 mg p /g* nt dry wt nt 4 1. 45 + 0.03 4 3 12.86 + 2 42 8 3 14. 82 + 3.63 9 99

PAGE 107

300 ...J -200 C, ::t tL a: en 100 0 Figure 5-2. I NO GROUP pH I 1 B 11.5 2 C 5.6 3 A1 4.9 I 4 A2 3.4 I 5 A3 2.3 I 6 A4 2.0 I INITIAL I SAP 6 ./2 2 3 2 5 4 1 3\~\ 5 45 3 :/'~. 2-2 / aji2 --2 0 50 1 00 150 200 DAYS OF INCUBATION Changes in SRP in unaerated decomposition bottles (each point represents the mean of three bottles). 100

PAGE 108

Table 5-3. Qualitative observations during unaerated decomposition of Websteria sp. Group B C Al A2 A3 A4 21 Days Plants green, floating; water colored; no odor. Plants pale, on bottom; water turbid, foul (non sulfide) odor. Plants decomposing on bottom; water turbid; strong sulfide odor. Plants brownish, on bottom, covered by mucouslike sheath; strong sulfide odor. Plants discolored, on bottom; apparent fungal growth; water clear; no odor. Plants discolored, on bottom; apparent fungal growth; water clear; no odor. Incubation Period 30 Days Plants same as Day 21; odor of cut grass. Plants same as Day 21; bacterial or fungal growths in water; sweetish odor. Same as Day 21. Plants same as Day 21; water clear; strong sulfide odor. Plants same as Day 21; sour-sweet odor. Plants same as Day 21; fungal growths; sweet ish odor. 41 Days Plants still green, floating; water colored; odor of hay. Plants appear decomposed; water cloudy; faint, non sulfide odor. Plants matted; water slightly colored; strong sulfide odor. Plant debris covered by bac teria or fungi; water clear or slightly turbid; sulfide odor. Plants matted; water turbid; fungal filaments; faint sweet odor. Plants appear decomposed, colorless, but not matted; fungi; sweetish odor. I-' 0 I-'

PAGE 109

bottles that received the smallest acid addition (Al) SRP release showed a similar delay, and by the end of the incubation SRP also was near 100 g/L. The initial total phosphorus in the Websteria plants (Table 5-2) would have yielded 680 g/L with complete conversion to SRP Final SRP values represented only 21% and 26% of this potential at the pH extremes (A4 and B), 14% in C and Al, and a mere 1.5% in the A2 group. These results are much lower than those reported by Foree and McCarty (1968), who found that after 200 days of anaerobic decomposition only 40% of the initial particulate phosphorus in cultured algae remained as refractory solids. Extensive growths of bacteria and fungi observed in the Websteria decomposition bottles appear to have reduced SRP release to the water, although the lack of consistent anoxic conditions may have inhibited the decomposition process. It is interesting to note that the final pH vaues in groups C, Al, and A2 were all close to 4.0, even though initial pH ranged from 3.35 to 5.61 (Table 5-1). There was a net pH decrease in C and Al during the incubation, and a net increase in A2 during the same period. This phenomenon strongly suggests two concurrent mechanisms in these groups: 1) all of the sulfate added to groups Al and A2 (as H2S04) ultim ately was reduced, resulting in consumption of all of the added H+; 2) the decomposition process resulted in an equal production of acidity in groups C, Al, and A2. The lowest pH values (2.02 and 2.31) were too extreme for sulfate reducing bacteria, and thus there was no significant pH increase in those bottles during the incubation Furthermore, qualitative observa tions recorded during the incubation also suggest that different 102

PAGE 110

processes occurrred at the different pH levels (Table 5-3). While the bottles adjusted to pH 2.02 and 2.31 exhibited an odor, it was sweetish rather than a sulfide smell A sulfide odor was noticed in groups Al and A2, but not in group C (no H2S04 addition). In addition to these differences in odor, there was a trend of increased fungal growth as pH decreased These results indicate that decomposition (and phosphorus release) can occur as low as pH 2, although the decomposer organisms and end products may change drastically at such low pH. Since some plants appear to have been killed initially by the acidification while others closer to the lake pH were not, no conclusions can be drawn about the effect of pH on phosphorus release rate during this experiment. How ever, near the lake pH (roughly between pH 3.4 and 5.6) decomposition seemed to produce acidity, and sulfate reduction apparently consumed essentially all of the H+ (added as H2S04) necessary to reduce pH from 5.6 to 3 4. Figure 5-3 shows the changes in SRP concentration during the dark incubation period in the second preliminary experiment, and the corre sponding pH values Significant SRP release occurred in all three treatments. Initial concentrations ranged from 20 to 70 g/L, and final values were about 400--550 g/L. Although SRP was consistently and significantly higher (a< 0.05) in the low pH treatment than at high pH, the rates of SRP increase were essentially the same. The pH increases were unexpected because oxidation reactions typ ically produce acid rather than consuming it (e.g., oxidation of NH4 to N03, or HSto S04). A subsequent test without plants 103

PAGE 111

400 -J -Q :t. Q. a: a, 200 0 10 20 30 DA VS OF INCUBATION Figure 5-3. Changes in pH and SRP during second preliminary decomposi tion experiment with Eleocharis. 104

PAGE 112

indicated that the glass incubation bottles were not the source of the observed buffering. Therefore the decomposition process appears to have resulted in a net decrease in [H+] concentration. Final Experiment The final decomposition experiment used a larger water:plant ratio to reduce the effect of this pH increase, and pH was checked and adjusted on a weekly basis to maintain the desired ranges. Figures 5-4 through 5-7 show changes in SRP in duplicate microcosms maintained at four pH values (3.7-5.5) during 227 days of aerobic senescence and decomposition of Eleocharis. Initial SRP concentrations all were less than 5 g/L, and final values ranged from about 80 to 115 g/L. Sol uble reactive phosphorus values did not differ significantly by pH groups on the last two sampling dates (ANOVA, a> 0.05), although the initial release of SRP (50-60 days) tended to be fastest at the lowest pH (Figure 5-8). The bottles maintained at pH 5.5 showed a lag of 30 days in which SRP remained <10 g/L, while levels increased to 15---40 g/L at the lower pH values during that same period. Soluble organic phosphorus (SOP) and POP also were measured during the first 44 days of incubation in this experiment. Soluble organic phosphorus values generally were near 1 g/L during that period, and never exceeded 5 g/L. Particulate organic phosphorus concentrations also were low; initial POP was ~l g/L, and levels gradually increased to 3-8 g/L by day 44. Thus most of the phosphorus released ran the Eleocharis plants was SRP, and the lag in SRP increase at pH 5.5 cannot be attributed to a slower conversion of SOP to SRP. Nichols and Keeney (1973) observed an initial release of SOP which preceded an increase in 105

PAGE 113

100 75 ..J C, 3, 50 a. cc (/J 25 0 REP A e REP B .A 106 0 50 100 150 200 250 DAYS OF INCUBATION Figure 5-4. Changes in SRP at pH 3.7 during 227-day aerobic decomposi tion of Eleocharis.

PAGE 114

100 75 ..J C, ,.:50 ll. a: "' 25 0 107 0 50 100 150 200 250 DAYS OF INCUBA T10N Figure 5-5. Changes in SRP at pH 4.1 during 227-day aerobic decomposi tion of Eleocharis

PAGE 115

..J ....... C, 100 75 :t 50 Q. a: (/) 25 0 e REP A .& REP B 108 0 50 100 150 200 250 DAYS OF INCUBATION Figure 5-6. Changes in SRP at pH 4.6 during 227-day aerobic decomposi tion of Eleocharis.

PAGE 116

a. a: "' 100 75 50 25 0 e REP A A REP B 109 0 50 100 150 200 2 0 DAYS OF INCUBATION Figure 5-7. Changes in SRP at pH 5.5 during 227-day aerobic decomposi tion of Eleocharis.

PAGE 117

..J ........ a 100 75 :1. 50 a. cc (/J 25 0 0 50 100 150 DAYS OF INCUBATION Figure 5-8 Composite of Figures 5-4 through 5-7. 110 SYMBOL pH A 3.7 4.1 0 4.6 D 5.5 200 250

PAGE 118

SRP during decomposition of Myriophyllum exalbescens. However, their macrophytes had been killed by herbicide treatment, which would cause a rapid release of the cellular contents (including SOP) before the decomposition process began. Based on the initial phosphorus content of the Eleocharis plants (Table 5-1), complete decomposition would have resulted in a final SRP concentration of 200 g/L in the bottles. The observed range of final SRP values (80-115 g/L) constitutes 40-58% of that potential, which means that 42-60% of initial plant phosphorus remained as refractory particulate material or heterotrophic biomass. These results are con sistent with those of Foree et al. (1970), who found that 50% of the initial particulate phosphorus 1n cultured algae was released during l year of aerobic decomposition. The fate of the remaining 40-60% of original particulate phos phorus in the macrophytes would depend on rates of sediment accumula tion and internal sediment processes. Refractory sedimentary materials continue to decompose at slow rates (Foree et al. 1970), and therefore constitute a potential source of phosphorus to overlying water for years after the initial ('\J. yr) period of rapid SRP release. These experiments demonstrate that the release of SRP from decom posing aquatic macrophytes is not affected by pH (over the range 3.7 to 5 5), and they suggest that other aspects of decomposition are s1m1larly unaffected by low pH. These findings contradict the "oligotroph ication" hypothesis of Grahn et al. (1974), but they support the results of many subsequent studies of the effect of acidification on decomposition. Sediment oxygen demand and glucose turnover show no effect of reduced experimental pH (Andersson et al. 1978; Gahnstrcm et 111

PAGE 119

al. 1980), while Schindler (1980) found no significant change in TP and no evidence of decreased decomposition during the 3-year experimental acidification of a Canadian lake from pH 6.6 to 5.6. 112

PAGE 120

CHAPTER 6 SEDIMENT-WATER INTERACTIONS Introduction Sediment adsorption and release reactions can affect phosphorus concentrations 1n lake water. The relative importance of these pro cesses in any given lake depends on sediment characteristics, water chemistry, lake morphometry (e.g., ratio of sediment surface to lake volume), and other factors that affect mixing within the lake. This chapter describes the physical and chemical characteristics of McCloud Lake sediments as well as experiments designed to evaluate the effect of pH on the capacity of those sediments to adsorb or release phos phorus. The experimental design included well-mixed batch systems and intact cores of surficial sediment. Background Both sediment characteristics and conditions in the overlying lake water have been shown to affect phosphorus adsorption. MacPhearson et al. (1958) found that maximum phosphorus adsorption occurred over the pH range 5.5 to 6.5 with sediments from lakes of several different types, and that more phosphorus was released frcxn the sediments at higher and lower pH values. Other workers have shown greater release 113

PAGE 121

of sediment phosphorus at high pH than at low pH (Andersson et al. 1978; Gahnstrom et al. 1980). As discussed in Chapter 2, sediment iron and aluminum are impor tant agents in the adsorption of inorganic phosphorus. Surface charge is another sediment characteristic that affects phosphorus adsorption. According to Laverdiere and Weaver (1977), surface charge characteris tics of certain soil types are strongly pH dependent. These include highly weathered tropical soils and spodosols, both of which are rich 1n hydrous oxides of Fe and Al. The sign of the surface charge depends on pH, and the magnitude of the charge at any given pH is determined by the ionic strength of the bulk solution. The authors further described a potentiometric acid-base titration method that can be used to deter mine the pH at which the sign of the net surface charge reverses C pHzpc) Langmuir adsorption isotherms offer an empirical approach to the interpretation of phosphorus adsorption data. The Langmuir equation is derived from theoretical considerations of gas adsorption onto solids, but it 1s also commonly used to model adsorption of phosphorus onto soils. A linear form of the equation 1s m/x = l/Q 0 + l/Q 0 bC 1n which, m = weight of dry sediment (g); x = amount of phosphorus adsorbed (mg); Q 0 = maximum adsorption with monolayer coverage (mg P/g dry sediment); and (EQ 4) 114

PAGE 122

b = constant that reflects energy of interaction between solute and adsorbent. Therefore, if experimental data fit the Langmuir model, a plot of m/x versus 1/C should give a straight line with y intercept at l/Q 0 and a slope of l/Q 0 b Methods Sediment Characterization Surficial mid-lake sediments were collected in July 1983 by com positing replicate petite Ponar grab samples from six widely spaced locations in the open water area of the lake. This composite sample was homogenized and standard procedures were used to determine physical characteristics (APHA 1980). An ashing-HN03 digestion procedure (Delfino and Enderson 1978) was used to measure total Fe and Al in the pelagic sediment. Aliquots of the sample also were extracted with 0 1 M Na-pyrophosphate (NaP 2 0 7 H 2 0), which has been shown to extract organically bound Fe and Al in soils without dissolving crystalline forms (McKeague 1967). Total and extractable Fe and Al were determined by flameless AAS. The charge characteristics of the pelagic sediment were evaluated by potentiometric titration (Laverdiere and Weaver 1977). The pro cedure involved titration of 10 g of soil in 100 mL of 0.01, 0.1 or 1.0 N NaCl with dilute HCl or NaOH. A magnetic stirrer provided constant m1x1ng, and approximately 0.01 meq of acid or base was added at 2-min intervals. To avoid drying the sediment, this procedure was modified by using 50 g wet sediment (3.48 g dry weight) and 50 mL of 0.02, 0.2, 115

PAGE 123

or 2.0 N NaCl. These mixtures were titrated with 0.1 N HCl or NaOH. Batch Adsorption/Desorption Experiments Short (~l wk) incubations of well-mixed sediment-water systems were used to evaluate the potential of Mccloud littoral and mid-lake sediments to adsorb or release SRP over the pH range 3.5-7.0. For the first experiment (June 1982), a sediment sample was composited from four petite Ponar grabs collected in littoral areas (0.5-1.0 m depth) free of macrophytes. Aliquots (25 g wet weight) of this homogenized sample were added to 100 mL of membrane-filtered (0.45 m) lake water in 30 125-mL screw-cap polyethylene bottles. Solution pH was measured after shaking for a 1-day equilibration period and 0.1 N H 2 S04 or 0.1 N NaOH was added to obtain six replicates at each of five pH values over the desired range. Solution pH was measured again after 2 more days of mixing, and 1 mL of a stock KHzP04 solution (to give 1 mg/L in 100 mL of water) was added to three replicates in each pH group. After 5 more days of mixing, final supernatant pH values were measured, and a solution sample from each bottle was filtered through a 0.45 m membrane filter. SRP was determined for the filtered supernatant samp les and the initial filtered lake water on an AutoAnalyzer II system. Similar procedures were used with pelagic sediments. A mid-lake surface sediment composite sample was collected as previously described (six locations) in February 1983. It was homogenized, and 50-g (wet weight) portions were mixed with 200-mL aliquots of membrane-filtered (0.45 m) lake water in 48 250-mL screw cap polyethylene bottles. Twenty additional bottles (used as controls) contained only 200 mL of 116 the filtered lake water with no sediment. After an initial equilibration

PAGE 124

period (1 day) and pH check, 1 N solutions of NaOH and H2S04 were used to obtain eight sediment-water replicates at five pH values over the range of 3.5 to 6.9. The remaining eight sediment-water bottles gave four replicates at each of two intermediate values in the same pH range. Four control replicates (water only) were titrated to each of the five pH values represented by eight sediment-water replicates, and no controls were used at the pH values with only four sediment-water replicates. After another day of shaking for equilibration to the new pH conditions, control and experimental flasks at each pH were spiked with four levels of stock SRP solution: 0.0, 0.5, 1.0, and 2.0 mg P/L above the ambient equilibrium SRP concentration. The final experimental matrix was as follows: H 3.5 3.8 4.2 4.7 5.5 6.4 6.9 No. Sediment-Water Bottles/SRP Level 2 2 2 l 2 l 2 No. Controls/SRP Level l l I 0 l 0 l After 2 days of mixing, final solution pH and aluminum and SRP concen trations were measured. Undisturbed Core Experiments Well-mixed batch systems can demonstrate the potential of a lake's sediments to adsorb or release phosphorus, but the extreme experimental 117

PAGE 125

conditions make it difficult to extrapolate the result to the ambient lake situation. Therefore undisturbed sediment cores were used to sim ulate more closely phosphorus adsorption and release in McCloud Lake. Sixteen cores consisting of ~10 cm undisturbed mid-lake sediment and ~30 cm of overlying lake water were collected manually using SCUBA equipment and pre-cut lengths of clear cellulose acetate butyrate tub ing (4.13 cm ID) stoppered with parafilm-covered rubber stoppers. Dis turbance of the interface was minimal, as indicated by clarity of the overlying water, and the presence of chironomid tubes and filamentous algal growth at the sediment surface when the cores were raised to the boat. The cores were transported carefully to the laboratory, where they were installed upright in a wooden rack covered with black plastic to exclude light. The volume of overlying water was adjusted to 400 mL, and the tubes were covered with parafilm to minimize evaporation. Each core was aerated through a small glass pipette inserted to about one-half the depth of the water column. Aeration provided vertical mixing of the water without suspending sediment, and maintained an oxy genated water column (anoxic conditions have not been detected in the McCloud Lake water column). After the cores had been allowed to equilibrate for 3 days, 0.1 N H2S04 or NaOH additions were used to obtain four replicate cores at water column pH values 3.7, 4.1, 4. 7, and 6.0. Sediment consumption of H+ and OHnecessitated pH measurement and acid or base addition every 2-3 days to maintain water column pH within +0.2 units of the nominal value. After 1 month of pH adjustment, SRP was added to two cores in each pH group to increase water column concentration by 1 mg P/L. Controls consisted of one aerated tube with 400 mL lake water 118

PAGE 126

adjusted to each pH and the same SRP spike. Water column SRP concen trations 1n the spiked cores were followed to evaluate the effect of pH on the capacity of undisturbed lake sediment to adsorb phosphorus. In addition, periodic SRP measurements in all replicates during the !-month pH adjustment, and in the unspiked cores after that, provide an estimate of pH effects on SRP release from undisturbed sediments. Results and Discussion McCloud Sediment Characteristics Sediments in the littoral area of McCloud Lake are distinctly different from those in the open-water part of the lake (Table 6-1). The flocculent, highly organic profundal sediments are fine textured and homogeneous. Littoral sediments, however, are more variable in texture and composition. Wave action in some areas has exposed alter nating layers that are predominated by sand or coarse, peaty organic matter. The nature of the littoral sediment surface depends on which layer is exposed in any particular area. Total Fe and Al (Table 6-1) were comparable to values reported from similar lakes in this area (Thompson 1982) and the organically bound fractions represented about 30% (Al) and 50% (Fe) of respective totals. Theoretical considerations indicate that at the pHzpc, surface potential and surface charge of constant potential surfaces will be zero. At pH values other than the pHzpc, the magnitude of a positive or negative surface charge will vary with salt concentration, while at the pHzpc, net charge will be independent of ionic strength. Figure 6-1 shows the results of the potentiometric titration 119 ____ __ __ _________ __ _ ___ _.

PAGE 127

Table 6-1. Some physical and chemical characteristics of McCloud Lake sediments. Water, Percent Volatile Sol ids, Percent Dry Weight Sand, Percent Dry Weight Total Al, Percent Dry Weight Extr. Al, Percent Dry Weight Total Fe, Percent Dry Weight Extr. Fe, Percent Dry Weight Interstitial Al, mg/L Interstitial Fe, mg/L Littoral 68. 7 11. 6 18. 3 Sediment Profundal 93.3 77. 7 0 2.0 0.57 0.36 0.18 0.078 0.025 120

PAGE 128

--------------------------------pH 3.0 O-t------------------;::;----~----:::-9'--4 0 5.0 Cl 0 0 ,.. ..... 5 C w m a: 0 "' C < + 10 1N NaCl i 0 0.1N NaCL E 0.01N NaCl 15 Figure 6-1. Potentiometric titration of McCloud mid-lake sediment (in NaCl solution) with 0.1 N HCl. 121

PAGE 129

of mid-lake sediments. The three titration curves intersect at a unique pH value (pHzpc) that is indenpendent of electrolyte con centration and locates the point at which the net surface charge reverses. Although the pHzpc is not precisely identified, it does occur in the pH range 3.5 to 3. 7. Laverdiere and Weaver (1977) pointed out that highly organic soils often have low pHzpc values, and noted that the occurrence of pHzpc below the zero point of titration (no acid or base addition) indicates the presence of a permanent negative charge on the surface. McCloud sediments thus appear to have a low pHzpc because of their high organic content, and they display a permanent negative charge that is independent of pH. Net charge does not become positive until solution pH is "1.,).6 or lower. Batch Adsorption/Desorption Experiments Table 6-2 summarizes the results obtained in the batch experiment using littoral sediments. No significant release of SRP from the sedi ments occurred over the pH range 3.5 to 6. 7. SRP adsorption ranged from 6.8 to 11.9 g P/g dry sediment, which constituted a removal of about 55% to 95% of the added SRP. The effect of solution pH on the amount of SRP adsorbed can be seen in Figure 6-2. Maximum adsorption occurred at pH 4.4, and only slightly less was removed at other pH values between 3.5 and 4.4. Adsorption decreased markedly at pH values above 4. 4. Table 6-3 presents the results obtained in batch experiments with profundal sediments. The amount of SRP adsorbed per gram dry sediment increased as the initial concentration increased, although at a given pH value the percent of added SRP retained stayed relatively constant. 122

PAGE 130

r 123 Table 6-2. SRP adsorption/release by littoral Mccloud sediments SRP Sediment SRP Adsorbed Added, Final SRP, Dry Wt, pH Group mg/L g/L* g g P /g dry wt % Added p 3.5 1.0 88 + 27 8 .02 11. 2 89 7 3.8 1.0 60 + 5 8.01 11. 6 93. 0 4.4 1.0 43 + 2 8.02 11. 9 95.0 5.5 1.0 156 + 15 8.02 10. 2 81. 7 6.7 1.0 388 + 36 8.02 6.8 54.6 3.5 0 1.0 + 1. 7 8.01 3.8 0 0.0 + 0.0 8.02 4.4 0 4.0 + 1. 7 8.01 5.5 0 3.6 + 0.6 8.01 6.7 0 3.0 + 0.0 8.02 Initial SRP in filtered lake water 3.0 + 0.6 *Mean+ standard deviation.

PAGE 131

-------------------..J -Cl E C? 'P"' ..J c:( j::: z C w ID cc 0 "' C < a. 100 80 60 40 20 3 0 4.0 5.0 pH 6 0 7.0 Figure 6-2. Effect of pH on SRP retention by littoral McCloud Lake sediments. 124

PAGE 132

125 Table 6-3. SRP adsorption/release by profundal McCloud sediments. SRP Sediment SRP Adsorbed Added, Final SRP, Dry Wt, pH Group mg/L g/L* g g P/g dry wt % Added p 3 5 0 6 + 1 4.04 0.51 23 + 4 3.97 29.2 95. 5 1.02 45 + 1 3.98 60.1 95:5 2.03 97 + 1 3.97 117. 6 95.2 3.8 0 0.5 + 0. 7 3.97 0.51 8 + 0 3.97 30.3 98.4 1.02 16.5 + 0.7 3.97 61.9 98.4 2.03 38 + 6 3.98 125.0 98. 1 4.2 0 0 + 0 3.97 0.51 7. 5 + 0. 7 3.98 30.3 98. 5 1.02 13 + 1 3.97 62.5 98. 7 2.03 40 + 9 3.98 123.0 98.0 4.7 0 0 3.97 0.51 3 3.98 31. 3 99.4 1.02 11 3.97 62.3 98.9 2.03 42 3.96 123.6 97.9 5.5 0 0 + 0 3.97 0.51 5 + 1 3.98 31.1 99.0 1.02 22.5 + 2 3.98 61.5 97. 7 2.03 103 + 4 3.98 119. 0 95.0 6.4 0 16 3.98 0.51 43 3. 96 28.9 91. 6 1.02 136 3.98 54.4 86.6 2.03 680 3.96 84.0 66.5 6.9 0 20 + 1 3.99 0.51 217 + 4 3.99 18.0 57.4 1.02 480 + 13 3.99 33.0 52. 7 2.03 1290 + 0 3.97 46.0 36.5 Initial SRP in filtered lake water 2 + 0. 7 *Mean+ standard deviation.

PAGE 133

Adsorption ran ged from 33 to 62 .5 g P/g dr y s e dim e nt in th e g r o up with ~ 1 m g /L SRP additio n whi c h amounted to 53 % to 99 % of add ed SRP. Th e range of p e r ce nt SRP r e moval was similar for b ot h litto ral and pr o fundal s e diments, although adsorption p e r gr am dry sediment was about five times hi g her for the profundal sediment. Thi s diff e r e nc e appears to be r e lated to the high e r o r g anic content of profundal sediments, which sug ge sts that the org ani c fraction is lar gely r espo nsibl e f o r SRP adsorption b y McCl o ud sediments. When removal is n o nnaliz ed to organic co ntent, th e littoral SRP ads o rption range 1s 58.6 to 102 g P/g dry org anic sediment, and th e corr es p o nding pro f undal (1 m g/L added SRP) range 1s 43 to 80.5 g P/ g dr y organic sedim e nt. The effect of pH on SRP adsorption by pr of undal sediments (Figur e 6-3) was ver y much like the pattern seen for li t toral sediments Maxi mum removal occurred at pH 4. 7. At low e r pH values adsorp t ion decreased slightl y while abov e pH 4. 7 the de cre as e was much more extreme. In bottles that received no addition, final SRP concentra tions showed a similar but opposite effect o f pH (Figure 6-4a). The low initial SRP concentration (2 /L) was reduced below detectable levels between pH 4.2 and 5.5, while rel e as e from the sediment occurred at pH 3.5 and at pH 5.5 Two possible mechanisms f o r the observed effect of oH on sediment SRP adsorption include changes in sediment characteristics with pH (e.g., surfa ce charge) and changes in SRP sp ec iation with pH. Theoret ical considerations indicate b o th process e s could influ e nc e anion adsorption. As mentioned previously, most surfaces ha ve a net ne g ative charge under nonnal e nvironmental conditi o ns. For constant p ot ential surfaces, decreas e s in pH r e duce the ma g nitud e of this net char g e until 126

PAGE 134

a. a: "' 0 w 0 C < L&. 0 ?Jl z 0 .:: a. a: 0 en C < 80 60 40 20 4.0 INITIAL P, mg/ L 5 0 pH 6.0 0 0.5 e 1.0 2.0 7.0 Figure 6-3. Effect of pH and initial concentration on SRP retention by profundal McCloud Lake sediment. 127

PAGE 135

20 A. QO .g P/L AOOED 10 0 250 B. 500 .g P/L AODED 125 ..J CD :t a. 0 a: (/J 500 ..J < C. 1000 .g P/L AOOEO z 250 ol___ ..:...._....:.. __ ~----------2000 D. 2000 .g P/L AOOED 1000 o.L--==~=;:::::::!:::::=:::::::;:===:..._----,r--__ ___, 3D 40 5D ~O lO pH Figure 6-4. Effect of pH on SRP release or adsorption by McCloud Lake profundal sediments. 128

PAGE 136

the pHzpc is reached, at which point further reduction in pH increases the net positive charge. Anion adsorption thus should show a continual increase as pH is reduced, within limits imposed by the solu bility and physical characteristics of the solid surface. SRP adsorp tion by McCloud sediments, however, reached a maximum at about pH 4.5 and decreased at lower pH values. Sediment charge characteristics can explain the observed effect of initial concentration on the percent SRP removed (Figure 6-3). At a given pH ~4.2, the percent added SRP that was adsorbed was independent of initial SRP concentration (within the range 0.5 to 2.0 mg P/L). As pH increased above 4.2, however, percent SRP removal showed an inverse relationship to initial concentration, and this effect bee rune more pronounced as pH increased. Figure 6-1 shows that near the sediment pHzpc ( 3.5 to 3.8), 1on1c strength of the bulk solution did not affect surface charge, but as pH increased above pHzpc, a higher ionic strength caused a more negative net surface charge. Since the phosphorus addition was as a salt (KH2P04), larger SRP additions resulted in higher ionic strength bulk solutions (and thus a greater net negative surface charge), which reduced the capacity of the sedi ment to adsorb SRP as pH increased above ZPC. Solution pH also has a minor effect on the distribution of ionic species of SRP over the pH range 3.0 to 7.0, as shown in Figure 6-5. H2P04 predominates over the entire range, with small proportions of undissociated H3P04 at low pH and HPOi at higher pH values. Hingston et al. (1967) demonstrated that HzP04 is the species most readily adsorbed, and its distribution corresponds remark ably well to the pattern of phosphate adsorption by littoral and pro129

PAGE 137

130 100 ~PO; __ .. ,,, H P04 .,, -/ 75 I a. a: (/) ..J < 50 t0 t'ilt I I 25 I / , ... -.-5 6 7 8 9 pH Figure 6-5. Effect of pH on speciation of SRP.

PAGE 138

fundal sediments (Figure 6-6). Over much of the experimental pH range, the profundal sediment removed essentially all of the added H2P04. Deviations at higher pH values occurred because of generally increasing net negative surface charge and because of increasing ionic strength effects on surface charge. Littoral sediments removed a smaller per centage of the available H2P04, but this was probably because of their lower organic content. The pattern of SRP release by pro fundal sediments shows the same relationship to changes in SRP specia tion. No SRP release occurred over the pH range in which H2P04 constitutes ~99-100% of SRPT, but SRP was released at higher and lower pH values. Table 6-4 summarizes the regression parameters and isotherm coef ficients that were obtained by applying the Langmuir equation (EQ 4) to data from the phosphorus adsorption experiment using mid-lake sediment. A limited number of data points was available for these calculations since only three levels of phosphorus addition were used. Data for the two pH values without duplicate bottles (4.7 and 6.4) were not included because each offered only three data points. Although the scarcity of data points dictates caution in interpreting these results, the data do seem to fit the Langmuir equation fairly well; regression r 2 values ranged from 0 943 to 0.997. Values of Q 0 represent theoretical maximum adsorption of phos phorus with monolayer coverage of the solid surface. The three lowest pH values showed the highest Q 0 values, while the pH 5.5 value was intermediate and the lowest Q 0 was at the highest pH. This trend reflects again the pH effect seen in phosphorus adsorption at specific levels of phosphorus addition. 131

PAGE 139

100 80 Q. a: "' C w C C < LI. 60 0 'l/l z 0 i= 40 Q. a: 0 "' C < 20 3.0 0 6. LITTORAL SEDIMENT 0.5 mg/ L 0 1.0 mg/L 2.0mg/L 4 0 PROFUNDAL SEDIMENT 5.0 pH SRP ADDITIONS 6.0 Figure 6-6. Composite of Figures 6-2, 6-3, and 6-5. 132 H po2 4 0 7.0

PAGE 140

Table 6-4. Regression parameters and isotherm coefficients obtained from Langmuir isotherm plots of mid-lake sediment phos phorus adsorption data. Regression, pH r2 Slope Intercept qo' mg/ g b, L/mg 3.5 0.943 o 70 1. 58 0.633 2.26 3.8 0.997 0.25 1.47 0. 680 5.88 4.2 0.950 0.22 1. 53 0.653 6.95 5.5 0. 980 0 15 8.13 0.123 54.8 6.9 0.988 8.97 13.49 0.074 1.50 133

PAGE 141

Sediment Core Experiments Results obtained with the undisturbed sediment cores were similar 1n some respects to trends seen with the batch systems. Figure 6-7 shows mean water column SRP concentrations in the unspiked cores over a 46-day period aft er desired nominal pH values were attained. Avera ge SRP concentrations varied in the range of 4 to 14 g/L, but no consis tent pH-related trend was evident. Based on results of the batch incu bations, the columns maintained at the pH extremes (3.7 and 6.0) would have been expected to exhibit the highest SRP values. This was not the case. However, several differences in experimental conditions possibly account for the fact that the low pH columns showed relatively low SRP concentrations. First, SRP release from undisturbed sediment should be slower than from well-mixed sediment because of decreased sedimentwat e r contact. Secondly, while the complete system (sediment and water) was titrated to the desired pH in the batch experiments, only the water column and an unknown depth of sediment were affected in the cores. Finally, during the longer incubation of the cores, biological activity could have been sufficient to mask a physical or chemical pH-mediated release of SR r from the sediments. The cores were incu bated in the dark to eliminate the influence of primary producers, but no effort was made to control heterotrophic activity. Bacterial and fungal growths were observed on the tube walls; zooplankters were observed in some water columns; and, as mentioned earlier, chironomids were present at the sediment surface of some cores. Figure 6-8 shows the removal of SRP from solution 1n the spiked cores and controls. Some initial loss of SRP occurred in the controls, but over the 30-day period, this loss amounted to only 10% to 16% of 134

PAGE 142

---135 12 I pH 3.7 8 f 4 t 0 12I pH -4.1 ,, aI 4 ..J Cl 0 =-1.. ,, Q. 12' f cc "' pH 4.8 a 4l 0 12! pH 5.5 8 4 f 0 ' 10 20 30 40 DAYS AT NOMINAL pH Figure 6-7. SRP variation in unspiked cores (means and ranges).

PAGE 143

1.0 0 8 0.6 _, 'A. Q e a. a: en 0!4 0 2 0 0 10 20 30 DAYS AFTER SRP ADDITION 1.0 _, ......... Q B e B. a. a: en 0 8 0 10 20 30 Figure 6-8. SRP concentration 1n spiked sediment-water columns (A) and controls (B). 136

PAGE 144

the phosphorus adsorbed in the sediment-water columns. The effect of pH on phosphorus adsorption in these cores was simi_lar to the trends seen in the batch adsorption experiments. The rates of SRP disappear ance seemed to be influenced by HzP04 distribution. Phosphorus removal was slowest in the cores maintained at pH 6.0, and fastest 1n the pH 4.1 cores, while intermediate adsorption rates were obtained 1n the cores at 3.7 and 4. 7. The curves in Figure 6-9 indicate that first-order kinetics can be used to model adsorption by the cores. Empirical rate constants (k) obtained from the slopes of regressions of the natural logarithm of SRP versus time are given in Table 6-4. They range from 0.036/day to 0.090/day and mirror the pH trend seen 1n Figure 6-8. Analysis of covariance (ANCOVA) demonstrates that there are significant differences among the slopes (k values) of the four regression equations, and further ANCOVA comparison shows that the rate constants for cores at pH 3.7 and 4. 7 are not significantly different (a= 0.81). However, those rate constants do differ significantly frcxn the k values obtained for the cores at pH 6.0 and 4.1 (a< 0.05). The net result is the follow ing relationship among the four rate constants: The rate of phosphorus adsorption 1s fastest at pH 4.1; slightly slower at pH 3.7 and 4. 7; and slowest at pH 6.0. This trend is similar to the pH effect seen in SRP adsorption in the batch experiments. Further reduction of pH in McCloud Lake thus could increase slightly the rate of phosphorus adsorption onto sediments, but the most dramatic change 1n adsorption rate would occur over the pH range 4.7 to 6.0. 137

PAGE 145

11. C/J C, 0 ..J I 1.25 1.00 0 75 0 50 0.25 0 3.7 10 DAYS 20 30 Figure 6-9. First-order SRP uptake plot for sediment-water columns. 138

PAGE 146

Table 6-5. Phosphorus adsorption rate constants for undisturbed sedi ment cores. Core pH Rate Constant, day-l Regression, r2 3. 7 0 061 0 93 4.1 0 090 0.86 4.7 0.063 0.97 6.0 0.036 0 99 139

PAGE 147

In summary, pH does affect adsorption of phosphorus by littoral and pelagic McCloud sediments, and the same pH trends are seen in well mixed or undisturbed sediment-water systems. Amounts of phosphorus adsorbed and rates of adsorption vary only slightly between the present pH of the lake (~4.7) and pH 3.5. Compared to the magnitude of varia tion possible 1n other processes that affect phosphorus concentrations in lakes, the differences 1n adsorption over this pH range seem insig nificant. However, as pH is increased above approximately 5.0, phos phorus adsorption decreases significantly with sediments from McCloud Lake. This implies that the effect of pH on phosphorus adsorption may contribute to the observed trend of low TP values in acidic lakes by increasing rates of phosphorus removal as lakes are acidified over the pH range 7.0 to 5.0. 140

PAGE 148

CHAPTER 7 SUMMARY AND CONCLUSIONS Laboratory and in situ experiments as well as an examination of the historical data base were used to characterize phosphorus dynamics in acidic, soft-water McCloud Lake and to evaluate the effect of acid ification on phosphorus cycling processes. The lake presently exhibits nutrient and chlorophyll-~ concentrations typical of oligotrophic Flor ida lakes. A 15-year decline in pH from 4.85 to 4.55 has not been accompanied by significant changes in TP, chlorophyll-~, or nitrogen to phosphorus ratios, which indicate phosphorus-limited primary produc tion. Total phosphorus and SRP both show maximum levels during late spring and summer. These variations appeared to be related to rainfall patterns and lake levels during 1980-1982. Rainfall to the lake surf ace contributes 90% of the water input to McCloud Lake, and atmospheric phosphorus deposition approximates loading rates required to maintain mesotrophic conditions. This emphasizes the importance of atmospheric nutrient loading to small lakes similar to McCloud Lake, but it also suggests that low pH may contribute to the low TP in the water column of McCloud Lake. Rooted submergent macrophytes constitute a significant in-lake storage of phosphorus that is approximately 2.5 times the average water column phosphorus storage. These plants appear more likely to mobilize sediment phosphorus than to compete with phytoplankton for water column 141

PAGE 149

phosphorus, although periphytic algae may limit the impact of mobilized phosphorus on phytoplankton. In situ littoral and open-water mesocosms indicated that acidifi cation (from 4.6 to 3.7) does lead to reduced water column TP values, although the trends were not consistent in the open-water enclosures. The pH treatments did not affect chlorophyll-a concentrations or phyto plankton densities. However, copepods were almost eliminated at pH 3. 7 in littoral and open-water enclosures. No relation was seen between pH and rates of phosphorus uptake by mesocosm seston. Laboratory micro cosms similarly showed no consistent relation between pH and the activ ity of extracellular acid phosphatase enzymes. The amount of phosphorus released from decomposing submersed macrophytes was independent of pH (over the range 3.7 to 5.5) after 227 days of aerobic dark incubation. Initial rates of release were somewhat faster at the lowest pH, although it was not established whether this was due to a difference in the times required for the plants to I die (live, intact plants were used). Nevertheless, these experiments indicated that acidification does not inhibit phosphorus release during decomposition of organic matter, as suggested by Grahn et al. (1974). Adsorption of phosphorus by McCloud Lake sediments was affected by pH, apparently through changes in the charge characteristics of the sediment surfaces and, to a lesser extent, by pH-controlled changes 1n SRP speciation. Maximum phosphorus adsorption occurred near pH 4.7 with littoral and profundal sediments, although adsorption rates and amounts varied only slightly between pH 5.0 and 3.5. However, phosphorus adsorption decreased significantly as pH was increased above approximately 5.0. This suggests that the effect of pH on retention of 142

PAGE 150

phosphorus by sediments may contribute to the observed trend of low TP values in acidic lakes. The effect would be most dramatic over the pH range 7.0 to 5.0, and further acidification of lakes near the pH of McCloud Lake would have little effect on phosphorus adsorption. Of the processes investigated in this study, only sediment adsorp tion demonstrated a response to pH that is consistent with a reduction of TP levels in acidic lakes. Sedimentation rates measured in McCloud Lake were high, but the effect of pH was not evaluated. Similar phosphorus adsorption experiments with sediments from many additional lakes would demonstrate the extent of the phenomenon and would clarify the significance of the process to reduced TP in acidic lakes. A more detailed measurement of the phosphorus budget terms in Mccloud Lake (total phosphorus deposition instead of wet only; seepage phosphorus inputs) would help clarify the relation between phosphorus loading to the lake and its trophic status. 143

PAGE 151

LITERATURE CITED Andersen, J. M. 1976. An ignition method for determination of total phosphorus in lake sediment. Water Res. 10:329--331. Andersson, G., S. Fleischer, and W. Graneli. 1978. Influence of acid ification on decomposition processes in lake sediment. Verh Internat. Verein. Limnol. 20:802-807. American Public Health Association. 1980. Standard methods, 15 ed. American Public Health Association, Washington, D.C. Armstrong, D. E. 1979. Phosphorus transport across the sediment-water interface. Lake restoration. Proceedings of a national confer ence. U.S. Environmental Protection Agency, Report 440/5-79-001, Washington, D.C. Baker, L. A. 1984. Mineral and nutrient cycles and their effect on the proton balance of a softwater, acidic lake Ph.D. disserta tion, University of Florida, Gainesville. 151 pp. Baker, L. A., P. L. Brezonik, and C. R. Kratzer. 1981. ing-trophic state relationships in Florida lakes. 56, Water Resources Research Center, University of Gainesville. Nutrient load Publication No. Florida, Barko, J. W., and R. M. Smart. 1980. Mobilization of sediment phos phorus by submersed freshwater macrophytes. Freshwater Biol. 10:229--238. Blomqvist, S., and L. Hakanson. 1981. A review on sediment traps in aquatic environments. Arch. Hydrobiol. 91(1):101-132. Bradley, D. B., and D. H. Sieling. 1953. Effect of organic ions and sugars on phosphate precipitation by iron and aluminum as influ enced by pH. Soil Science 76:175---179. Brezonik, P. L., T. L. Crisman, and R. L. Schulze. 1984. cormnunities in Florida softwater lakes of varying pH. Fish. Aquatic Sci. (in press). Plank tonic Can. J. Brezonik, P. L., E. S. Edgerton, and C. D. Hendry. 1980. Acid precip itation and sulfate deposition in Florida. Science 208:1027-1029. Brezonik, P. L., C. D. Hendry, Jr., E. S. Edgerton, R. L. Schulze, and T. L. Crisman. 1983. Acidity, nutrients, and minerals in atmos

PAGE 152

pheric precipitation over Florida: deposition patterns, mechan isms, and ecological effects. EPA project completion report for Grant No. 805560. U.S. EPA, Office of Research and Development, Corvallis, Ore. 170 pp. Brezonik, P. L., W. H. Morgan, E. E. Shannon, and H. D. Putnam. 1969. Eutrophication factors in north central Florida lakes. Bull. Series No. 134, Water Resour. Res. Ctr. Publ. No. 5, University of Florida, Gainesville. Buechler, D. G., and R. D. Dillon. 1974. Phosphorus regeneration 1n fresh-water Paramecia. J. Protozool. 21(2):339-343. Canfield, D. E., Jr. 1981. Chemical and trophic state characteristics of Florida lakes in relation to regional geology. Center for Aquatic Weeds, IFAS, University of Florida, Gainesville. 444 pp. Carignan, R., and J. Kalff. 1980. Phosphorus sources for aquatic weeds: water or sediments? Science 207:987-989. Carritt, D. E., and S. Goodgal. 1954. Sorption reactions and some ecological implications. Deep-Sea Res. 1:224-243. Chen, C. W. 1969. Concepts and utilities of an ecologic model. In Modeling the eutrophication process. Workshop proceedings, St. Petersburg, Florida, Federal Water Quality Administration. Coleman, R. 1944a. Phosphorus fixation by the coarse and fine clay fractions of kaolinitic and montmorillonitic clays. Soil Sci. 58: 71-77. Coleman, R. 1944b. The mechanism of phosphate fixation by montmoril lonitic and kaolinitic clays. Soil Sci. Soc. Amer. Proc. 9:72-78. Confer, J. L., T. Kaaret, and G. E. Likens. 1983. sity and biomass in recently acidified lakes. Sci. 40: 36--42. Zooplankton diver Can. J. Fish Aquat. Copeland, B. J., and W. R. Duffer. 1964. The use of a clear plastic dome to measure diffusion of natural waters. Limnol. Oceanogr. 9:494-495. Cowling, E. B. 1982. Acid precipitation in historical perspective. Environ. Sci. Technol. 16(2):110A-122A. Davis, R. B., D. L. Thurlow, and F. E. Brewster. 1975. Effects of burrowing tubificid wonns on the exchange of phosphorus between lake sediment and overlying water. Verh. Internat. Verein. Limnol. 19:382-394. Delfino, J. J., and R. E. Enderson. 1978. Comparative study outlines: methods of analysis of total metal 1n sludge. Water and Sewage Works 125(RN):R32-R34 and R47-R48. 145

PAGE 153

Dillon, P. J., D. S. Jeffries, W. Snyder, R. Reid, N. Yau, D. Evans, J. Moss, and W. Scheider. 1978. Acidic precipitation in south central Ontario: recent observaions. J. Fish. Res. Bd. Canada 35:80~15. Dillon, P. J., and R. H. Rigler 1974. A test of a simple nutrient budget model predicting the phosphorus concentration in lake water. J. Fish. Res. Bd. Canada. 31: 1771-1778. Dillon, P. J., N. D. Yan, W. A. Scheider, and N. Conroy. 1979. Acidic lakes in Ontario, Canada: characterization, extent and responses to base and nutrient additions. Arch. Hydrobiol. Beih. 13:317-336. Fillos, J., and W. R. Swanson. 1975. The release rate of nutrients from river and lake sediments. J. Water Poll. Contr. Fed. 47:1032-1042. Fleming, W. M. 1975. A model of the phosphorus cycle and phytoplank ton growth in Skaha Lake, British Columbia, Canada. Verh. Int. Ver. Limnol. 19:229-240. Fogg, G. E. 1975. Algal cultures and phytoplankton ecology. Univer sity of Wisconsin Press, Madison. Foree, E. G., and P. L. McCarty. 1968. The decomposition of algae 1n anaerobic waters. Tech. Rept. No. 95, Fed. Water Poll. Contr. Admin. Res., Grant WP-1037. Foree, E. G., W. J. Jewell, and P. L. McCarty 1970. The extent of nitrogen and phosphorus regeneration from decomposing algae. Fifth Water Poll. Res. Conf. Fried, M., and L. A. Dean. 1955. Phosphorus retention by iron and aluminum in cation exchange systems. Soil Sci. Soc. Amer. Proc. 19:14~147. Gahnstrom, G., G. Andersson, and S. Fleischer. 1980. Decomposition and exchange processes in acidified lake sediment. Pages 306-308 in D. Drablos, and A. Tollan (eds.), Ecological impact of acid precipitation. SNSF, Oslo, Norway. Gallepp, G. W. 1979. Chironomid influence on phosphorus release in sediment-water microcosms. Ecol. 60(3):547-556. Gallepp, G. W., J. F. Kitchell, and S. M. Bartell. 1978. Phosphorus release from lake sediments as affected by chironomids. Verh. Internat. Verein. Limnol. 20:458-465. Grahn, 0., H. Hultberg, and L. Landner. 1974. Oligotrophication: a self-accelerating process in lakes subjected to excessive supply of acid substances. Arnbio 3:93-94. Hall, C. A. S., and R. Moll. 1975. Methods of assessing aquatic pri mary productivity. Pages 19--53 in H. Lieth and R. H. Whittaker 146

PAGE 154

(eds.), Primary productivity of the biosphere. Springer-Verlag, New York. Hargrave, B. T., and G. H. Geen. 1968. Phosphorus excretion by zoo plankton. Limnol. Oceanogr. 13:332-342. Harter, R. D. 1968. Adsorption of phosphorus by lake sediments. Soil Sci. Soc. Amer. Proc. 32(4):514-518. Haseman, J. F., E. H. Brown, and C. D. Whitt. 1950. Reactions of phosphate with clays. Soil Sci. 70:257-271. Hayes, F. R., and J. E. Phillips. 1958. Lake water and sediment, IV. Radiophosphorus equilibrium with mud, plants, and bacteria under oxidized and reduced conditions. Limnol. Oceanogr. 3:459-475. Hendrey, G. R., K. Baalsrud, T. S. Traaen, M. Laake, and G. Ruddum. 1976. Acid precipitation: some hydrobiological changes. Ambia 5:224-227. Hendry, C. D., and P. L. Brezonik. 1984. Chemical composition of softwater Florida lakes and their sensitivity to acid precipitation. Water Resour. Bull. (in press). Hingston, F. J., R. J. Atkinson, A. M. Posner, and J. P. Quirk. 1967. Specific sorption of anions. Nature 215:1459--1461. Huber, W. C., P. L. Brezonik, J. P. Heaney, R. E. Dickinson, S. D. Preston, D. S. Dwornik, and M. A. DeMaio. 1982. A classification of Florida lakes. Final Report to Fla. Dept. Env. Regulation, No. ENV-05-81-1. Hynes, H. B. N., and B. J. Greib. 1970. Movement of phosphate and other ions from and through lake muds. J. Fish. Res. Bd. Canada 27:653-668. Jansson, M. 1981. in acid lakes. Induction of high phosphatase activity by aluminum Archiv. Hydrobiol. 93(1):32-44. Jansson, M., H. Olsson, and 0. Broberg. acid phosphatases in the acidified Hydrobiol. 92(3):377-395. 1981. Characterization of Lake Gardsjon, Sweden. Arch. Johannes, R. E. 1965. Influence of marine protozoa on nutrient regen eration. Limnol. Oceanogr. 10:434~42. Kamp-Nielsen, L. 1974. Mud-water exchange of phosphate and other ions in undisturbed sediment cores and factors affecting the exchange rates. Arch. Hydrobiol. 73:218-237. Kitchell, J. F., J. F. Koonce, and P. S. Tennis. 1975. Phosphorus flux through fishes. Verh. Int. Verein. Limnol. 19:2478-2484. 147

PAGE 155

Landers, D. H. 1979. A durable, reusable enclosure system that compensates for changing water levels. Limnol. Oceanogr. 24(5):991--994. Landers, D. H. 1982. Effects of naturally senescing aquatic macrophytes on nutrient chemistry and chlorophyll a of surrounding waters. Limnol. Oceanogr. 27(3):428--439. Laverdiere, M. R., and R. M. Weaver. 1977. Charge characteristics of spodic horizons. J. Soil Sci. Soc. Amer. 41:505---510. Lean, D. R. S. 1973a. Movement of phosphorus between it biologically important forms in lake water: a community excretion mechanism. J. Fish. Res. Bd. Can. 30:1525-1536. Lean, D. R. S. 1973b. Phosphorus dynamics in lak e wat ~r Science 179:678-680. Lean, D. R. S., and E. White. measurements of phosphorus Aquat. Sci. 40:147-155. 1983. Chemical and radiotracer uptake by lake plankton. Can. J. Fish. Leivestad, H., G. Hendrey, I. P. Muniz, and E. Snekvik. 1976. Effects of acid precipitation on freshwater organisms. Pages 87-111 in F. H. Braekke (ed.), Impact of acid precipitation on forest and freshwater ecosystems in Norway. SNSF Proj. FR6/76, Aas, Norway. Li, W. C., D. E. Armstrong, J. D. H. Williams, R. F. Harris, and J. K. Syers. 1972. Rate and extent of inorganic phosphate exchange in lake sediments. Soil Sci. Soc. Amer. Proc. 36:279-285. Likens, G. E., R. F. Wright, J. N. Galloway, and T. J. Butler. 1979. Acid rain. Scientific Amer. 241:43---51. Lind, 0. T., and R. S. Campbell. 1970. Community metabolism in acid and alkaline strip-mine lakes. Trans. Amer. Fish. Soc. 3:577-582. Lopez-Hernandez, I. D., and C. P. Burnham. 1974. The effect of pH on phosphate adsorption in soils. J. Soil Sci. 25:207-216. Lund, J. W., C. Kipling, and E. D. LeCren. 1958. microscope method of estimating algal numbers basis of estimations by counting. Hydrobiol. The inverted and the statistical 11:143-170. MacPhearson, L. B., N. R. Sinclair, and F. R. Hayes. 1958. Lake water and sediment: III. The effect of pH on the partition of inorganic phosphate between water and oxidized mud or its ash. Limnol. Oceanogr. 3:31S--326. Mortimer, C. H. 1941. The exchange of dissolved substances between mud and water in lakes. J. Ecol. 29:280-329. 148

PAGE 156

Mortimer, C. H. 1942. The exchange of dissolved substances between mud and water in lakes. J. Ecol. 30:147-201. Mortimer, C. H. 1971. Chemical exchanges between sediments and water in the Great Lakes--speculations on probable regulatory mechanisms. Limnol. Oceanogr. 16:387--404. National Research Council of Canada (NRCC). 1981. Acidification of the aquatic environment: scientific criteria for assessing the effects of acid deposition on aquatic ecosystems. Publ. No. 18475. Nichols, D. S., and D. R. Keeney. 1973. release from decaying water milfoil. Nitrogen and phosphorus Hydrobiologia 42:509-525. Odum, H. T. 1956. Primary production of flowing waters. Limnol. Oceanogr. 2:85-97. Odum, H. T., and C. M. Hoskin. 1958. Comparative studies on the metabolism of marine waters. Publ. Inst. Marine Sci., University Texas 8:159-170. Peters, R. H., and D. R. S. Lean. 1973. The characterization of soluble phosphorus released by limnetic zooplankton. Limnol. Oceanogr. 18:270-279. Peters, R. H., and F. H. Rigler. 1973. Phosphorus release by Daphnia. Limnol. Oceanogr. 18(6):821-839. Pettersson, K., and M. Jansson. 1978. Determination of phosphatase activity in lake water: a study of methods. Verh. Internat. Verein. Limnol. 20:1226--1230. Phillips, J. E. 1964. The ecological role of phosphorus in water with special reference to microorganisms. Pages 61-79 in H. Heukelekian and N. C. Dondero (eds.), Principles and applications in aquatic microbiology. John Wiley and Sons, New York. Pomeroy, L. R., E. E. Smith, and C. M. Grant. 1965. phosphate between estuarine water and sediments. Oceanogr. 10:167-172. The exchange of Limnol. Porcella, D. B., J. S. Kumagai, and E. J. Middlebrooks. ical effects on sediment-water nutrient exchange. Div., Proc. Amer. Soc. Civil Eng. 96:911--926. 1970. Biolog J. San. Eng. Reichardt, W. 1975. Responses of phosphorus remobilizing Cytophaga species to nutritional and thermal stress. Verb. Int. Verein. Limnol. 20:2227-2232. Rigler, F. H. 1964. The phosphorus fractions and the turnover time of inorganic phosphorus in different types of lakes. Limnol. Oceanogr. 9:511-518. 149

PAGE 157

Rigler, F. H. 1966. 1n lake water. Rigler, F. H. 1968. hypothesis that 1.n lake water. Radiobiological analysis of inorganic phosphorus Verh. Int. Verein. Limnol. 16:465~70. Further observations inconsistent with the the molybdenlDil blue method measures orthophosphate Limnol. Oceanogr. 13:7-13. Rigler, R. H. 1973. A dynamic view of the phosphorus cycle in lakes. Page 539 in E. J. Griffith, A. Beeton, J. M. Spencer, and D. T. Mitchell (eds.), Environmental phosphorus handbook. John Wiley and Sons, New York. Schindler, D. W. 1980. Experimental acidification of a whole lake: a test of the oligotrophication hypothesis. Pages 370-374 in D. Drablos and A. Tollan (eds.), Ecological impact of acid precipita tion. SNSF, Oslo, Norway. Shannon, E. E., and P. L. Brezonik. 1972. Relationships between lake trophic status and N and P loading rates. Environ. Sci. Technol. 6: 719-725. Shukla, S. S., J. K. Syers, J. D. H. Williams, D. E. Armstrong, and R. F. Harris. 1971. Sorption of inorganic phosphate by lake sedi ments. Soil Sci. Soc. Amer. Proc. 35:244-249. Singer, R., D. A. Roberts, and C. W. Boylen. 1983. Effects of the benthic algal mat of an acidic Adirondak lake on phosphorus cyc ling and other ions. U.S. EPA/NCSU Annual Review, Acid Precip. Prog. Sprules, W. G. 1975. Midsummer crustacean zooplankton communities 1n acid-stressed lakes. J. Fish. Res. Bd. Canada 32:389-395. Struthers, P. H., and D. H. Sieling. 1950. on phosphate precipitation by iron and pH. Soil Sci. 69: 205-213. Effect of organic anions aluminlDil as influenced by Stumm, W., and J. J. Morgan. 1981. Aquatic chemistry. John Wiley and Sons, New York. 780 pp. Swenson, R. M., C. V. Cole, and D. H. Sieling. 1949. Fixation of phosphate by iron and aluminum and replacement by organic and inorganic ions. Soil Sci. 67:~22. Syers, J. K., R. F. Harris, and D. E. Armstrong. 1973. Phosphate chemistry in lake sediments. J. Environ. Qual. 2(1):1-14. Thompson, D. M. 1981. Distribution of heavy metals in selected Florida lakes. Master's thesis, University of Florida, Gainesville. 160 pp. Traaen, T. S. 1980. Effects of acidity on decomposition of organic matter in aquatic environments. Pages 340-341 1n D. Drablos and 150

PAGE 158

A. Tollan (eds.), Proc, Int. Con. Ecol. Impact Acid Precipita tion. SNSF project, Norway. U.S. Environmental Protection Agency. 1979. Methods for chemical analysis of water and wastes. U.S. Environmental Protection Agency, Cincinnati, Ohio. Vollenweider, R. A. 1968. Scientific fundamentals of the eutrophica tion of lakes and flowing waters, with particular reference to nitrogen and phosphorus as factors in eutrophication. OECD, Paris. Vollenweider, R. A. 1975. Input-output models with special reference to the phosphorus loading concept in limnology. Schweiz. Z. Hydrol. 37:53-84. Wetzel, R. G. 1983. Limnology, 2 ed. Saunders College Publishing, New York. Wetzel, R. G., and G. E. Likens. 1979. Limnological analyses. W. B. Saunders Co., Philadelphia. Yeasted, J. G., and M. M. Morel. 1978. Empirical insights into lake response to nutrient loadings, with application to models of phos phorus 1n lakes. Environ. Sci. Technol. 12(2):195--201. Zilversmit, D. B., C. Enternnan, and M. C. Fishler. 1943. On the cal culation of "turnover times" and "turnover rate" from experiments involving the use of labelling agents J. Gen. Physiol. 26:325-331. 151

PAGE 159

BIOGRAPHICAL SKETCH Reuben Walter Ogburn, III, was born and raised in Mobile, Alabama. He obtained a B.S. in Biology from Southwestern at Memphis 1n June 1970 and entered active duty in the U.S. Navy in September of that year. During his 2-year tour as an electronics technician aboard the U.S.S. Shelton (DD-790), he visited several countries in the western Pacific Ocean and saw combat duty off the coast of North Vietnam. After an Honorable Discharge from the Navy, he returned to Alabama to enter an M.S program in Marine Science at the University of Alabama. Walt and the former Marlyn Wadley were married in 1974, and he obtained the M.S. in 1976. The Ogburns were Peace Corps Volunteers in Talcahu ano, Chile, where Walt taught Marine Ecology and Intertidal Ecology at the Catholic University of Chile from 1976 to 1978. Their first son, Reuben Walter Ogburn, IV, was born in Concepcion, Chile, in 1977. After they returned to the U.S., Walt worked as a hydrographer/biolo gist for an environmental consulting firm in Mobile, Alabama, until 1980, when he entered the Ph.D. program in the Environmental Engineer ing Sciences Department at the University of Florida. A second son, Douglas Rayfield Ogburn, was born in Mobile in 1980. Walt has worked for Breedlove Associates, Inc., a Gainesville consulting firm, since August 1983. 152

PAGE 160

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Patrick Adjunct Environmental Engineering Sciences I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. 4me0 Thomas L. Crisman Associate Professor of Environ mental Engineering Sciences I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Bob G. Volk Professor of Soil Science

PAGE 161

This dissertation was submitted to the Graduate Faculty of the College of Engineering and to the Graduate Council, and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. April 1984 AluaL~., dean,College of Engineering Dean for Graduate Studies and Research

PAGE 162

UNI V ERSIT Y O F FLORIDA II I II IIIIII Ill Ill lllll lllll II IIIIII IIII II llll llll 111111111111111 3 1262 08553 3791