UIRNVE SIT YLBRARY

IIII III III IIIIIIII III III I III I

3 1293 105838118

 

 

 

 

 

 

This is to certify that the

thesis entitled

AQUATIC NITROGEN CYCLING:
THE PROCESSING OF NITRATE BY CHLORELLA VULGARIS
AND THE POSSIBILITY OF AMMONIA LOSS
TO THE ATMOSPHERE

 

presented by

Jim Edward Galloway

has been accepted towards fulfillment
of the requirements for

PhoDo degreein Fisheries & Wildlife

 

éfifié

Major professor

Date May 15, 1980

 

0-7639

 

OVERDUE FINES;
_ 25¢ per day per item
In
(Lb-M t; W:

I.\ -‘ , Place in bookm mto remove
._ “may; .4. charge from circulation records

'94

 

‘&1}?’:g "’ 61%,; b lam;
sum ~3f85m¢

I II
300 A 19 1

 

 

 

AQUATIC NITROGEN CYCLING:
THE PROCESSING OF NITRATE BY CHLORELLA VULGARIS
AND THE POSSIBILITY OF AMMONIA LOSS
TO THE ATMOSPHERE

BY

Jim Edward Galloway

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Fisheries and Wildlife

1980

-/

I/u

ABSTRACT

AQUATIC NITROGEN CYCLING:
THE PROCESSING OF NITRATE BY CHLORELLA VULGARIS
AND THE POSSIBILITY OF AMMONIA LOSS
TO THE ATMOSPHERE

 

BY

Jim Edward Galloway

This study investigated three segments of the aquatic nitrogen
cycle: (1) nitrate assimilation by algae and the possibility of
increased algal nitrogen requirements at high pH levels, (2) algal
loss of dissolved organic materials, and (3) ammonia volatilization
from aquatic systems to the atmosphere.

The first two points were addressed by growing Chlorella vulgaris
in closed, batch culture microcosms and monitoring their uptake of
nitrogen and carbon and loss of organic matter. This was accomplished
by periodically analyzing samples of media for the concentrations of
the various forms of inorganic and organic carbon and nitrogen. Ammo-
nia volatilization was examined by measuring the decline in the ammonia
concentration of buffered solutions of ammonium chloride placed under
a laboratory exhaust hood. Data collected were evaluated and combined
with data found in the literature to generate an equation predicting
ammonia loss rates.

The study revealed that Chlorella vulgaris takes up and stores

 

nitrogen in excess of its immediate needs when abundant nitrogen and

carbon are available in the growth medium. As external nitrogen

Jim Edward Galloway

concentrations declined, algal stores of nitrogen decreased resulting
in lower cellular N/C molar ratios. Similar but less drastic declines
in cellular N/C molar ratios were observed as the carbon reserve of

the medium was depleted. Algal specific growth rates were controlled
by light intensity when all nutrients were supplied in excess. The
relationship between algal specific growth rate and the free carbon
dioxide concentration of the medium of carbon limited cultures followed
a rectangular hyperbolic curve. A similar relationship was displayed
between algal specific growth rate and cellular N/C molar ratios for
nitrogen limited microcosms. No evidence was found for any interaction
between these two limiting factors or for any increase in algal nitro-
gen requirements at high pH levels.

The percentages of the total nitrogen and total carbon taken by
the algae which appeared as dissolved organic matter in the media were
approximately the same (about 15%). It is possible that this material
is largely released through cell lysis.

Ammonia volatilization rates were predicted by the dual boundary
layer model as the product of the ammonia concentration of the water
in excess of atmospheric equilibrium and the mass transfer coefficient
for ammonia. The mass transfer coefficient was described by an equa-
tion dependent upon pH, temperature, and wind speed. It was determined
that ionized ammonia played an important role in increasing the mass
transfer coefficient over intermediate pH ranges when wind speeds were‘
low. Ammonia loss to the atmosphere is likely to be significant in

(

shallow eutrophic lakes, especially those with low alkalinities.

ACKNOWLEDGMENTS

I would like to acknowledge the love and companionship of my
wife, Deborah. Her support, both personal and financial, made my
graduate studies possible.

I wish to extend my sincere thanks to my advisor, Dr. Darrell L.
King, who provided encouragement and guidance well in excess of his
responsibility. Thanks also go to the other members of my doctoral
committee: Dr. Niles Kevern, Dr. Thomas Burton, and Dr. Eric Goodman.

Appreciation is extended to Dr. C.D. McNabb and the Institute of
Water Research for the use of analytical facilities. The technical
aid of John Craig, Chuck Annett, and Rosita Cabrera were also of great
help.

This study was supported by a Graduate Fellowship from the

National Science Foundation.

ii

TABLE OF

LIST OF TABLES . . . . .
LIST OF FIGURES . . . .
INTRODUCTION . . . . . .

MATERIALS AND METHODS .

Experimental Approach

Apparatus and Procedures

Algal Microcosms .

Nutrient Media
Algal Cultures

Sampling Procedure .
Ammonia Volatilization

Analytical Methods

Glassware . .
Temperature .
Alkalinity . .
pH . . . . . .

CONTENTS

Procedure

Ammonia, Nitrate, Nitrite . . .
Organic and Total Nitrogen . . .
Carbon - Direct Measurement . .

Carbon - Calculated

Growth Rate Calculations . . . .

RESULTS AND DISCUSSION .
Algal Microcosms .
Microcosm pH .

Carbon Fixation
NitrOgen Uptake

N/C Molar Ratios

Cell Leakage .
Growth Kinetics

iii

Ammonia Volatilization . . . .

Fick's Law . . . . . . . .
ANH3_w Mediated Flux . . .
AXNH3_w Mediated Flux . .
Dual Layer Model . . . . .
Enhancement . . . . . . .
Magnitude of Enhancement .
Partitioning of Resistance
Predicting KENH3-w . . . .
Effects of Wind . . . . .
Importance of Enhancement

Effects of Temperature on Diffusivity .
A Predictive Equation for Ammonia Loss .
A Predictive Equation for Concentration Change

ECOLOGICAL IMPORTANCE . . . . . . .

CONCLUSIONS 0 O O O O O O O O O O O

Algal Microcosms . . . . . . .

Ammonia Volatilization . . . .

LIST OF REFERENCES . . . . . . . . .

APPMDICES O O O O O O O O O O O O 0

Appendix A:

Appendix B:

iv

Algal Microcosms -

Raw

Ammonia Volatilization

Data

- Raw

121

12h

.131

15h

10.

LIST OF TABLES

Measured initial values of the variable nutrients in
the algal microcosms and their microcosm abbreviation

COdeSo O O O O O O O O O O O O O O O I 0 O O I O O O O O 0

Results of linear regressions of ammonia volatilization
rates (mmoles/cme/hr) against deviations in the unionized
ammonia concentration of the water from atmospheric
equilibrium (moles/l). . . . . . . . . . . . . ... . . . .

Results of multiple linear regressions of ammonia
volatilization rates (mmoles/cm2/hr) against deviations
in the ionized and unionized ammonia concentrations of
the water from atmospheric equilibrium (moles/2) . . . . .

Multiple linear regressions of flux as a function of

the concentration of unionized and ionized ammonia in
excess of atmospheric equilibrium for selected pH

groups . . . . . . . . . . . . . . . . . . . . . . . . . .

Total ammonia concentration (£NH3 moles/£)* in water

at equilibrium with an atmospheric concentration of

2.35 x 10'10 moles/2 under one atmosphere total

pressure . . . . . . . .V. . . . . . . . . . . . . . . . .

Variations in mass transfer coefficient (KZNH _w) with
temperature and pH at a wind speed of 0.66 me ers/
second C O O O O C O C O C O C O O O O O O O O O O I O O 0

Variations in mass transfer coefficient (KXNH3-w) with
temperature and wind speed at pH 9.0 . . . . . . . . . . .

Variations in mass transfer coefficient (KZNH3.V) with
pH and Wind at 20 C O O C . C O C O C C O C C O O O O O O 0

Comparisons of mass transfer coefficients (KZNH3-w)
calculated for Equation 36 with those calculated
from available data. . . . . . . . . . . . . . . . . . . .

Time (days) required to lose one half of the ammonia

in the water in excess of atmospheric equilibrium from

a 5 meter deep impoundment subjected.to a 2 meter/

second wind. . . . . . . . . . . . . . . . . . . . . . . .

ll

67

71

80

83

100 "

101

102

105

115

Composition of inorganic nutrient medium used for
algal microcosms O O O O O O O O O O O O O O O O O O I O I 132

Raw data collected from algal microcosms . . . . . . . . . 133

Ammonia volatilization data from experimental laboratory
vessels 0 O O O O O 0 O O 0 O O O O O O O O O O O O O O O 155

vi

10.

ll.

l2.

13.

LIST OF FIGURES

Experimental microcosm used for Chlorella vulgaris
growth studies 0 O '0 O O I I O O O O O O O O O O I O O 0

Time related pH responses of Chlorella vulgari
Clutllres dwing Trial I O O O 0 O O O O O O O O O I O O 0

Time related pH responses of low nitrogen cultures of
Chlorella vulgaris during Trial II . . . . . . . . . . .

Time related pH responses of replicate cultures of
Chlorella vulgaris during Trial II . . . . . . . . . . .

Variations in microcosm inorganic and fixed carbon
with time 0 O O O O O I O O O O O O O O O O I O O O O O 0

Variations in nitrogen and carbon assimilation of
selected high alkalinity microcosms. . . . . . . . . . .

Variations in nitrogen and carbon assimilation of
selected low alkalinity microcosms . . . . . . . . . . .

Variations in the nitrogen to carbon ratio of algae
from selected microcosms . .‘. . . . . . . . . . . . . .

Algal carbon to nitrogen molar ratios at maximum
microcosm pH levels. . . . . . . . . . . . . . . . . . .

Carbon to nitrogen molar ratios at maximum.microcosm
pH levels as reported by Atherton (197h) and Garcia
(1971‘) O O O O 0 O O O O O O O O O O O O O O O O O O O O

Corrected carbon to nitrogen molar ratios at maximum
microcosm pH levels as calculated from the data of
Atherton (197A) and Garcia (197a). . . . . . . . . . . .

Variations in algal carbon to nitrogen molar ratios at
maximum culture pH levels with media free carbon dioxide
concentration. . . . . . . . . . . . . . . . . . . . . .

Percentage of carbon fixed which leaked in experimental
microcosms as a function of the media free carbon
dioxide concentration at maximum culture pH. . . . . . .

27
29
3o
32
33.
3h

37

no.
2a
2+3
an

A9

Figure
1b.

15.

16.

17.

18.

19.

20.

21.

22.

23.

2h.

25.

26.

Variations in algal specific growth rates with media
free carbon dioxide concentrations during Trial I . . . . .

Variations in algal specific growth rates with media
free carbon dioxide concentrations in low alkalinity
microcosms Of Trial 1:1. 0 O O O O O O O O O O O O O O O O 0

Variations in algal specific growth rates with media
free carbon dioxide concentrations in high alkalinity
microcosms of Trial II. . . . . . . . . . . . . . . . . . .

Variations in algal specific growth rates with cellular
nitrogen to carbon molar ratios for high nitrogen,
Trial I microcosms O O I O O O O O O O O O O O O O O O O O 0

Variations in algal specific growth rates with cellular
nitrogen to carbon molar ratios for high nitrogen,
Trial II microcosms . . . . . . . . . . . . . . . . . . . .

Variations in algal specific growth rates with
cellular nitrogen to carbon molar ratios for low
nitrogen microcosms . . . . . . . . . . . . . . . . . . . .

Relationship between cellular carbon to nitrogen
molar ratios and inorganic nitrogen in the media
for selected microcosms . . . . . . . . . . . . . . . . . .

Relationship between ammonia loss rate and the
concentration of unionized ammonia in the media in
excess of atmospheric equilibrium . . . . . . . . . . . . .

Linear regressions of the log of ammonia flux as a
function of the log of the unionized ammonia concentration
of the water in excess of atmospheric equilibrium . . . . .

Dual layer model of a gas-liquid interface. . . . . . . .2.

Mean mass transfer coefficients at each pH level tested
in laboratory experiments . . . . . . . . . . . . . . . . .

A comparison of various linear regression equations
with experimental mean mass transfer coefficients at
different pH levels . . . . . . . . . . . . . . . . . . . .

Resistance to movement of ammonia across the air-water

interface as a function of pH for the current laboratory
conditions 0 O O O O O O I O O O O O O O O O O O O O O O 0 0

viii

53

Sh

55

56

57

6O

65

69
73'

79

81

87

Figure
27.

28.

29.

Calculated air phase resistance of data from Weiler

(1977) as a function of wind speed. All data was

adjusted to pH 12 to eliminate pH as a cause of

variance, and temperature was between 20 C and 21 C

unless otherwise noted . . . . . . . . . . . . . . . . . . 93

Variation in mass transfer coefficient with and
without enhancement as a function of media pH. . . . . . . 95

Resistance as related to wind speed at 10 cm
above the water surface. . . . . . . . . . . . . . . . . . 96

INTRODUCTION

As seen by the general public, the problem of cultural
eutrophication is composed of two major factors: (1) increasing
primary productivity accompanied by an increase in the visible plant
population, and (2) fish kills associated with subsequent periods of
anoxia resulting from the decay of the generated biomass. These
changes result from an increase in the amount of nutrients available
and a shift in the relative abundance of these nutrients. Because
phosphorus most often limits primary productivity in unperturbed
fresh waters, and diatoms and green algae most often dominate phos—
phorus limited systems, most studies and control efforts have been
focused on this nutrient. Despite these efforts, many natural and
man-made systems have moved beyond a phosphorus limit to a carbon
or nitrogen limit. Understanding the mechanisms and rates of these
changes and how these factors can be incorporated into management
strategies has become an important area in limnological and water
quality research.

A few investigators have looked at carbon and nitrogen as
limiting nutrients in highly enriched waters. King (1972) discussed
the evolution of a carbon limit under the hypereutrophic conditions
Of a sewage lagoon. Gavis and Ferguson (1975) proposed a model of
the kinetics of CO uptake under carbon rate limiting conditions.

2
King (1970), Sievers (1971), Young (1972), and Hill (1977) all

2
reported on the interaction of carbon and some other resource as a
limiting factor.

Nitrogen has been considered as the nutrient limiting algal
productivity in aquatic systems by Gerloff and Skoog (195h, 1957),
Lund (1965), Keeney (1973) and others. Recent emphasis on nitrogen
has been largely in relation to marine systems where it is often in
short supply (Yentsch, Yentsch, and Strube, 1977; Eppley, Renger, and
Harrison, 1979; Eppley g§_§l,, 1979). Additional investigations into
the freshwater nitrogen cycle also have been conducted. The need for
better information on both the rates and extents of essentially all
the pathways in the fresh water nitrogen cycle was pointed out by
Brezonik (1972). Although many investigations have been made since
that time,vfiiflisome of the most recent being those of McElroy §t_al,
(1978), Vlek and Stumpe (1978), and Rosenfeld (1979), questions about
several segments of the aquatic nitrogen cycle remain to be answered.

Algal assimilation of and requirements for nitrogen have been
a focal point of investigations on nitrogen metabolism in plants for
nearly thirty years (Syrett, 1953, 1962; Morris, l97h). However,
only studies by Atherton (197M) and Garcia (l97h) have subjected algal
cultures to the wide range of carbon and pH conditions which they .
might be expected to encounter in an eutrophic lake throughout the
course of a summer. In these studies, data from algae grown in
closed microcosms suggest that increases in pH may require algae to
use abnormal amounts of nitrogen for each unit of carbon fixed.
This phenomenon raises questions about the validity of carbon to
nitrogen ratios reported in the literature, and hence the relative

rate of cycling of these nutrients.

3

Absent from the work of Atherton and Garcia was any considera-
tion of the ultimate fate of the nitrogen taken by the algae. It
seems reasonable that this nitrogen would have been leaked from the
cells while they were still living or would have been released when
the cells broke down at their death (Brezonik, 1972). Which of these
pathways was dominant may be important in determining the time and
place of ammonification of the nitrogenous materials. The fate of
this nitrogen, after being cycled to ammonia, is also unclear.

Ammonia nitrogen in a fresh water system could move in any one
of four directions. It could be used as a nitrogen source by algae
or other aquatic organisms. This pathway has been extensively studied
over the years (Syrett, 1962). If algal uptake was slow enough and
conditions were favorable, ammonia could be converted by bacteria to
nitrite and then nitrate. This would be apparent in an increase in
the nitrate concentration of the water. The ammonia also could be
sorbed to the sediments, but this process would be expected to be
largely reversible over short time periods (Rosenfeld, 1979), and
therefore in a nitrogen limited system ammonia would reappear at a
later time. The only pathway which would represent a direct and
permanent sink of nitrogen is ammonia volatilization. This mechanism
has received some attention in the past (Stratton, 1968, 1969; Blain,
1969; Folkman and Wachs, 1973; Weiler, 1977; Vlek and Stumpe, 1978),
but is still poorly understood and often ignored in the calculation
of the nitrogen budgets of aquatic systems.

This study was aimed at providing additional information on
three of the segments of the nitrOgen cycle by (l) investigating the

possibility of increased algal nitrogen requirements at high pH levels,

h

(2) determining the significance of cell leakage, and (3) estimating
ammonia volatilization rates. To carry out these investigations two
types of experiments were conducted. The first two objectives were
addressed by taking periodic samples from a series of closed, batch
culture microcosms. These samples were used to follow changes in both
inorganic and organic nitrogen and carbon in the media of the micro-
cosms, thus allowing nutrient uptake and loss to be calculated.
Initially it was hoped that the third objective could be reached by
estimating ammonia volatilization rates from lakes directly, using an
apparatus similar to that of Stratton (1969). This device proved
inadequate to handle the flux rates encountered. Therefore, all
ammonia loss estimates were made by measuring decreases in the ammonia
concentration of ammonium chloride solutions placed under a laboratory

exhaust hood for a specified period of time.

MATERIALS AND METHODS

Experimental Approach

Whenever possible, multiple means were used to measure all
nutrient transformations within the microcosms. Nitrogen assimilation
was estimated by three means: (1) the depletion of nitrate and nitrite
nitrogen in the nutrient medium, (2) the increase in total organic
nitrogen plus ammonia (total Kjeldahl nitrogen), and (3) the difference
between measured total nitrogen (total persulfate nitrogen) and nitrate
plus nitrite nitrogen. Nitrogen retained by algal cells was estimated
by the difference between total persulfate nitrogen and dissolved per-
sulfate nitrogen, or the difference between total KJeldahl nitrOgen and
dissolved Kjeldahl nitrogen.

Carbon assimilation, used as an indicator of algal growth, also
was measured by two methods. Total carbon fixed was determined by
calculations based on the measured initial alkalinity and changes in
pH, and directly measured by a total carbon analyzer. Particulate
organic carbon was determined by difference between total organic
and dissolved organic carbon.

The extent of leakage of organic carbon and organic nitrogen
from cells was determined by measuring ammonia, dissolved Kjeldahl
nitrogen, and dissolved organic carbon in the nutrient media. This,
coupled with the measurements described in the previous paragraphs

provide a carbon and nitrogen budget for each microcosm.

6

Multiple initial nitrate levels were used in the microcosms so
that different amounts of carbon would be extracted from the alkalinity
system, and the termination of growth would occur over a wide range of
pH levels. Two widely different initial alkalinities were also used
at each of these nitrogen levels. This arrangement provided data
from which carbon to nitrogen ratios under nitrogen limiting condi-
tions at a variety of pH levels could be determined, independent of
inorganic carbon levels. Some microcosms also were provided with
additional nitrate after algal growth ceased, allowing the calculation
of multiple "final" C/N ratios from individual microcosms. The
nitrate levels selected for the first of two separate experiments per-
formed using the same apparatus (Trial 1) were similar to those used
by Garcia (l97h). For the second set of experiments (Trial II),
selected nitrate levels were repeated from Trial I.

Three independent experiments (Trials A, B, and C) were carried
out to estimate ammonia volatilization rates.‘ All ammonia loss
Vestimates were made under a laboratory exhaust hood using small
plastic containers which were filled with an ammonium.chloride solu-
tion buffered at various pH levels. Rates of loss were determined
by periodically removing a water sample and analyzing it for ammonia

nitrogen.'

Apparatus and Procedures

 

A1ga1_Microcosms

The batch culture microcosms used in this study were similar to
those employed by Young (1972), Atherton (197k), and Garcia (l97h).
They consisted of four-liter Erlenmeyer flasks, each containing a

2h mm teflon coated magnetic stir bar and sealed with a rubber stopper

(Figure 1). The stir bars were used for agitation of the cultures
prior to sampling. Each stopper was fitted with a water filled
manometer to allow the internal pressure to equilibrate with the
atmosphere, andrismall (6 mm) rubber serum cap through which samples
were removed using a hypodermic syringe. Both of these devices were
designed to minimize possible recarbonation from the atmosphere.

The microcosms were arranged in a row on a wooden rack and
illuminated by banks of two to watt Sylvania "Gro Lux" fluorescent
lights placed approximately 21 cm away. The selection of an appro-
priate light intensity was achieved through preliminary experiments
in which the lights were covered with combinations of black nylon
cloth and fine mesh wire screens painted black to prevent alteration
of the light spectrum (Luebbers and Parikh, 1966).

Light intensity reaching a flat sensor placed against the
illuminated inside of a test microcosm flask filled with deionized
water was measured between the first and second trials using a LI—COR
light meter (model LI-SlO integrator and model LI-l925 sensor). This
unit measures only "photosynthetically active radiation" which is
defined as those wave lengths between hOO and 700 nanometers. The
average illumination at the ten microcosm stations for Trial I, as
measured using the test flask, was 26.78 microeinsteins - 1n”2 . sec'l
with a range of 2h.87 to 30.h7 microeinsteins . m‘2 - sec‘l. The
high readings were from stations in the middle of a given light bank
(stations 3 and 8) and the low values were from stations at each end
of the rack (stations 1 and 10). Prior to the second trial, the posi-

tions of the stations were changed slightly to reduce the differences

which occurred between stations in the first trial. The average

 

 

 

 

 

 

—-d

 

 

 

 

 

 

 

 

 

Figure l. Ebcperimental microcosm used for Chlorella vulgaris
growth studies.

9

illumination measured at stations 1 through 10 for the second trial
was 25.76 microeinsteins - m‘2 - sec"l with a range of 25.59 to 25.9h
microeinsteins . m’2 - sec‘l. Station 11 was added at a later date
and undoubtedly received less light than the original ten stations.
It also should be noted that stations on the end of the rack (l, 10,
ll) probably received less refracted radiation than those in the mid-
dle. This could not be measured adequately with the available equip-

ment.

Nutrient Media

The inorganic nutrient media used were derived from one originally
formulated by Kevern and Ball (1965). This medium was dominated by
monovalent cations, a situation which reduced the chance of carbonate
precipitation and subsequent reductions in alkalinity (Young and King,
1973). The suitability of modifications of this medium for batch
culture experiments has been demonstrated under similar conditions
by King and King (197M), Ziesemer (l97h), Atherton (197A), Garcia
(l97h), Hill (1977), and others. For the present work, the medium was

modified by reducing the CaCl2 concentration in an effort to further

reduce the chance of carbonate precipitation, and substituting Na2MoOh'
for (NHh)6MOTO2h to reduce extraneous ammonia (Table A1).. Despite
these changes, calcium carbonate precipitation prevented autoclaving
the media prior to seeding. Iron flocculation also occurred under
normal conditions in the microcosms. Attempts to eliminate this
problem by adding iron in the form of a Fe-EDTA complex prepared as

described by Soeder (l97h) were fruitless. This precipitation is not

believed to have had any effect on the extent of algal growth.

10

The media were prepared for each of the alkalinity levels by
mixing the non-varying macronutrients and an appropriate volume of
micronutrients in a large polyethylene carboy. After dispensing media
into the microcosm flasks and recarbonating them by bubbling with lab
air, an appr0priate amount of KNO3 dissolved in media was added to
reach the desired nitrogen concentration. The measured initial

alkalinity and nitrogen concentration of the medium in each of the

microcosms is listed in Table l.

Algal Cultures

Chlgzglla,xulgg;i§_was obtained from the University of Texas Type
Culture Collection (Culture No. 260). Prior to the start of each
trial, a portion of the culture was aseptically transferred from,its
original protease slant to the liquid medium previously described
containing l.h mg N/£ and h meq alkalinity/l. The algae were allowed
to grow in this medium until it was noticeably green. At this time
a small sample was removed and reseeded into new medium containing
h meq/g alkalinity and approximately 1.0 mg N/l. This culture was
allowed to grow until there was no further carbon extracted from the
carbonate-bicarbonate alkalinity System, at which time the cells
were assumed to be nitrogen starved. The culture was then centri-
fuged, and after decanting the supernatant media, the remainder was
resuspended in nitrogen free media and used to seed the experimental
microcosms.

Short lag times in the experimental microcosms and low dissolved
organic carbon (<l.7 mg C/Z) and dissolved organic nitrogen (<O.l2 mg
N/l) in the seed media suggest that the seed cells were in good condi-

tion. The total and dissolved organic carbon content of the seed was

11

Table 1. Measured initial values of the variable nutrients in the
algal microcosms and their microcosm abbreviation codes.

 

 

Abbreviation Alkalinity Nitrate Concentration
Code* meq/l NO3-N mg/l
Trial I
1:1 1.031 0.093
1:1' 1.116 0.056
1:2 n.288 0.093
1:2' h.33h 0.055
1:3 1.031 0.350
1:3' 1.051 0.290
lzh h.288 0.350
1:h' h.320 0.271
1:5 1.031 l.h7
1:6 h.288 1.57
1:7 1.631 2.92
1:8 h.288 2.92
1:9 1.031 5.77
1:10 h.288 5.80
Trial II

2:1 1.015 0.076
2:2 b.30h 0.078
2:3 1.015 0.330
2:h h.30h 0.330
2:5 1.015 0.900
'2:6 h.30h 0.890
2:7 h.30h 1.h5
2:8 h.30h 1.h6
2:9 h.30h 2.87
2:10 h.30h 2.85
2:11 9.30h 1.h2

 

The first digit of each abbreviation code designates whether the
microcosm was from Trial I or Trial II. The second digit identi-
fies the station on the light rack which the microcosm occupied.
These stations were numbered sequentially from one end to the
other. Those codes followed by a prime (') indicate that the
particular microcosm was started in the middle of the trial,
after cessati6n of growth of the initial microcosm at that sta-
tion.

12

determined using a Beckman model 915A Total Carbon Analyzer. Total
and dissolved organic nitrogen was determined for the first trial
using the alkaline persulfate digestion method and for both trials
using a micro-Kjeldahl digestion. This information was necessary for
the construction of a total nitrogen budget for each microcosm. All
experimental microcosms were seeded to an algal concentration of 0.20
mg particulate C/fi. The total organic nitrogen added during the ini-

tial seeding of the Trial I microcosms was 59.h ug N or 0.015 mg N/l

of media while those cultures which were initiated in the middle of
Trial I received 5h.9 us N or 0.015 mg N/£ of media. The Trial II micro-
cosms started with 71.7 us N or 0.019 mg N/2 of media. No inorganic

nitrogen was measured in any of the seed.

Sampling Procedure

As a number of analyses were to be performed on each sample, it
was necessary to develop a detailed sampling procedure to insure that
a portion of each sample would be handled properly for each analysis.
While thoroughly mixing the culture with a magnetic stirrer, samples
were withdrawn through the serum cap using a large (30 or 50 mi)
syringe equipped with a 30 cm cannula. A small aliquot of the sample
was transferred directly to a vial for later total carbon and total
organic carbon analysis. When both analyses were to be made, the
sample was immediately capped to prevent any atmospheric recarbona-
tion. When only organic carbon was to be considered, the sample was

acidified below a pH of 24.5 with H Po,4 (Gordon and Sutcliffe, 1973)

3
prior to capping. A second aliquot was injected into a beaker for
pH determination. During Trial I, subsamples were removed after pH

analysis for total nitrogen and total organic nitrogen determinations;

but in Trial II, a separate aliquot of the original sample was used.

13
The remaining portion of the sample was acidified slightly (to
about pH 7) with HBPOh to prevent any volatilization of ammonia, and
filtered through a pre-ignited (2 hours at 500 C) Whatman GF/F glass
fiber filter under a vacuum of less than 10 cm of mercury. The fil-
trate was then pipeted into appropriate containers for the determina-
tion of ammonia, nitrate, nitrite, total dissolved nitrogen, dissolved

Kjeldahl nitrogen, and dissolved organic carbon.

Ammonia Volatilization Procedure

Ammonia volatilization studies were carried out under a labora—
tory ventilation hood by measuring ammonia loss from small, open,
plastic vessels. These vessels, approximately 15.5 cm in diameter
and slightly under 6 cm in depth, were filled with 1010 m2 of buffer
containing a given concentration of ammonia nitrogen added as ammonium
chloride. An initial 10 m1 sample was removed for ammonia analysis
after thorough mixing with a magnetic stirrer and initial pH deter-
minations. This sample was acidified with concentrated sulfuric acid
and sealed for later ammonia analysis which was accomplished using the
same procedure employed for digested Kjeldahl samples. Next, a needle
fixed to a steel rod suspended over the vessel was positioned so it
just touched the surface of the liquid. The vessels were then allowed
to sit without agitation from 1 to 2% hours. Throughout this period,
the hood fan was operated creating a wind velocity of about 0.66 m/s
at the surface of the liquid. This insured that ammonia concentrations
in the air did not build up to inhibit volatilization. At the end of
the exposure period, deionized water was added to each vessel until
the liquid level again just touched the suspended needle. This com-

pensated for any evaporation which had taken place. A final sample

1h
was then removed for ammonia analysis and a final pH reading was
taken.

Background ammonia concentrations in the lab air were determined'
using Gelman A glass fiber filters impregnated with 5 N_sulfuric acid
prepared as described by Okita and Kanamori (1971). After pumping
lab air through a series of these filters at a known rate for 1h hours,
they were each soaked in a known volume of deionized water which was
then analyzed for ammonia. The buffers used to control the pH levels
in the volatilization vessels were made from various combinations of
potassium dihydrogen phosphate, borax, sodium bicarbonate, sodium
hydroxide, and hydrochloric acid. They were all dilute versions of

those described by Robinson and Stokes (1978).
Analytical Methods

Glassware

Prior to its initial use, all glassware was cleaned in chromic
acid cleaning solution and then subjected to the standard cleaning
procedure. This procedure included washing with soap and water,
rinsing with distilled water, soaking in dilute hydrochloric acid,
rinsing twice with distilled water, and finally rinsing three times
with deionized water. Between uses all glassware was subjected to the
standard cleaning procedure. During the first trial, the strength of
the hydrochloric acid used was approximately 10% by volume. In the

second trial, the strength of this acid was increased to 50%.

Temperature
One microcosm in each trial had a general laboratory thermometer

With 1 C graduations inserted through a third hole in its stopper.

15
Prior to each sampling period, the temperature of the media of that
culture was recorded, and this temperature was assumed to hold for each

of the other microcosms.

Alkalinity

Titratable alkalinity was determined for both the high and low
alkalinity media at the beginning of each trial. Samples of 50.0 ml
were titrated with 0.02 §.H230u and pH was determined after each acid
addition. The equivalence points of these titrations were taken to be
the inflection points of the respective curves. These were found as
outlined by Dick (1973). A11 alkalinities reported are the average

of values obtained from two or three titrations.

pH

A Corning model 12 research grade pH meter equipped with a Corning
semiémicro glass electrode was used for pH determinations throughout
the study. The meter was calibrated daily at two points using commer-
cial standard buffer solutions.

A procedure similar to that of Young (1972), Garcia (l97h), and
Hill (1977) was used to overcome the problem of atmospheric recarbona-
tion of the sample during pH measurement. A 50 m2 beaker was sealed
with a number 9% st0pper through which two holes were bored. The
beaker was continuously flushed with nitrogen entering through one
of the holes. After thorough flushing of the beaker, the sample was
introduced through the second hole which also acted as a port for the
introduction of the pH electrode. Measurements were taken while the
sample was being agitated with a magnetic stirrer. No corrections
were made for any changes in the temperature of the sample which may

have occurred during pH measurements.

16
Ammonia, Nitrate, Nitrite

These three nitrogen compounds are grouped here because analysis
for them was carried out simultaneously. Due to the large number
of samples to be analyzed and the desire to remove as little media as
possible from the microcosms, a Technicon Autoanalyzer II was used for
these determinations. An attempt was made to carry out these analyses
as soon as possible after sampling, and in most instances it was accom-
plished in less than 2h hours. All samples were stored under refriger-
ation until analyzed.

Ammonia analysis was carried out using a method similar to that
described by Solorzano (1969) based on the indophenol-blue reaction
between ammonia, phenol, and hypochlorite. This procedure was reported
to be sensitive down to 0.01 mg N/z by the U.S. Environmental Protec-
tion Agency (l97h).

Total nitrate plus nitrite was analyzed by converting nitrate
to nitrite via cadmium reduction and measuring it as the colored
diazonium compound which is a product of the sulfanilamide reaction.
This same reaction was used for nitrite analysis by omitting the
reduction step. The procedures used are those described by the U.S.
Environmental Protection Agency (197k) and are reported as being

satisfactory down to 0.05mg N0 -N«ar NOZ-N/l. Based on the current

3

study, these estimates of reliability appear to be conservative.

Organic and Total Nitrogen

A large number of methods exist for the digestion of organic
samples and subsequent nitrogen analysis (see Strickland and Parsons,
1968; Golterman, 1969; Allen, l97h;enx1reviews in Analytical Chemistry,

April, 1973, 1975). Photo-oxidation or photochemical combustion has

17
received increasing attention recently, with several papers appearing
in the past 12 years describing various methods of ultraviolet combus-
tion (Armstrong gp_al,, 1966; Armstrong and Tibbits, 1968; Manny pp_
51,, 1971; Semenov gp_al,, 1976). As advances in procedure occur and
equipment becomes more accessible, this will probably become the
method of choice. However, currently the most commonly used methods
involve wet combustion of the sample and subsequent analysis for
nitrate, nitrite, or ammonia.

The classical method for determining organic nitrogen is the
Kjeldahl digestion followed by ammonia analysis. Traditionally, this
has involved digestion in sulfuric acid with a metal catalyst, distil-
lation and nesslerization. When done on untreated samples, the results
include NH3-N and organic-N. Modifications of this method are numerous
with some of the more recent ones being presented by Adamski (1976),
Nicholls (1975), and Scheiner (1976). By using the indophenol-blue
method of ammonia analysis each of these methods avoids the distilla-
tion process. This improves precision, but some authors report that
it requires variations in the digestion process to avoid interferences
by the Kjeldahl reagents. A different method of analysing water sam-
ples for organic nitrogen has recently been proposed by DiElia,
Steudler, and Corwin (1977). This method involves persulfate diges-
-N and N0 -N. This

3 2
method yields what is called total persulfate nitrogen (TPN) which

tion and determination of total nitrogen as NO

should be equivalent to total Kjeldahl nitrogen (TKN) plus N03-N and

N02-N. Organic nitrogen can be determined by difference when NH

N03-N, and N0

3-H,

2-N are measured in parallel samples.

In this study, quantification of organic nitrogen was accomplished

both by difference and by direct measurement. The alkaline persulfate

18
digestion of D'Elia gp_g;, (1977) for total nitrogen was used on both
filtered and unfiltered samples. Any changes which occurred in the
unfiltered samples could be attributed to gaseous losses of nitrogen
or to changes in the completeness of the digestion. After correction
for dissolved N03-N, N02 3

samples provided an estimate of total particulate nitrogen, presumed

-N, and NH -N, the filtered and unfiltered
to be the total nitrogen contained by the algal cells. It should be
noted however that this procedure was not recommended by D'Elia gp
g. (1977).

Duplicate measurements of organic nitrogen plus ammonia were
obtained by a modification of the Kjeldahl digestion. Three different
procedures were tested in preliminary studies in order to select the
method giving most complete digestion, best precision, and greatest
distinction between organic plus ammonia nitrogen and nitrate plus
nitrite nitrogen. The methods used were restricted by the sample size
and digestion time required. Comparisons were based upon the repro-
ducibility of results obtained from the digestions of ammonium.chloride
standards, standards spiked with sodium nitrate, and algal cultures.
The methods tested were those outlined by Harris (personal communi-
cation), Adamski (1976), and Nicholls (1975).

Of these methods, the method of Harris most closely approaches
the traditional micro-Kjeldahl digestion. It employs a sulfuric acid,
potassium sulfate, mercuric oxide digestion carried out in test tubes
on a block digestor. Digestion is followed by ammonia determination
using the indophenol-blue method for acidic samples on a Technicon
Autoanalyzer (Technicon Autoanalyzer Industrial Method 15h-71W).

This method appears to have several advantages over traditional

Kjeldahl methods, particularly at low nitrogen levels.

l9

Adamski (1976) pr0posed the use of a semiautomated persulfate
digestion in which the samples are concentrated in 1 mi of sulfuric
acid to which 1 gram of potassium persulfate is added to oxidize the
sample. A similar method was proposed by Nicholls (l975),but 50%
hydrogen peroxide was used as the oxidizing agent rather than potas-
sium persulfate. The advantages of these methods were the avoidance
of both the possible conversion of nitrate to ammonia (Vitula, 1971),
and the possible interference of Kjeldahl catalysts in the indophenol-
blue determination of ammonia (Nicholls, 1975; Scheiner, 1976).

Trial runs using the hydrogen peroxide method of Nicholls, modi-
fied by omitting neutralization and determining ammonia by the
indophenol-blue method for acidic samples, produced linear standard
curves with high precision. When standards were spiked with sodium
nitrate an unexplained interference occurred which depressed the absor~
bance of the colorimetric product and was more pronounced with
increasing concentration of both ammonia and nitrate. Sodium chloride
spiked samples showed no interference. In general, it was found that
the persulfate method of Adamski gave higher results with identical
algal samples, particularly at low algal concentrations. This evi-
dence of incomplete digestion using hydrogen peroxide also was found
in comparisons with the catalyzed Kjeldahl method. Repeated additions
of hydrogen peroxide served to increase ammonia production éslightly,
but it appeared that no number of additions could cause complete
breakdown of organic nitrogen to ammonia. The hydrogen peroxide
method produced higher results only when the Kjeldahl digestion did

not go to completion as evidenced by a lack of clearing of the sample.

20

Trials of the persulfate method revealed much less precision
among the standards than was found with either the hydrogen peroxide
method or the later trials of the catalyzed Kjeldahl method. Incom-
plete conversion of algal nitrogen to ammonia is apparent in this
method when compared to the catalyzed Kjeldahl, and a nitrate inter-
ference similar to that demonstrated by the hydrogen peroxide method
is suggested by some of the results.

The nitrate interference in these two methods is difficult to
explain. Scheiner (1976) noted a nitrite interference in the
indophenol-blue reaction, but the strong oxidizing conditions of the
reactions make the conversion of nitrate to nitrite unlikely.
Bradstreet (1965) reported that strong oxidants such as H202 may cause
conversion of ammonia to nitrate and therefore low results, but this
does not explain why the presence of a nitrate spike reduced measur-
able ammonia.

The more complete digestion achieved by the catalyzed Kjeldahl
is probably the result of higher digestion temperatures provided by
the addition of potassium sulfate and the catalytic action of the mer-
curic oxide. Bradstreet (1965) cites some investigators who claim
complete digestion requires catalytic mercury, and as the amount of
mercury used in the catalyzed reaction investigated here did not
appear to interfere with the ammonia analysis, it was determined the
method of choice. Early trials of this method, carried out in test
tubes using a block digestor, lacked reproducibility. This was
apparently due to contaminated boiling chips and the problem was
remedied by two hours of boiling in concentrated hydrochloric acid.

Later trials were carried out in 50 m2 volumetric flasks with larger

21

samples and more digestion solution (a mixture of 13h g K2SOh, 2 g
ago, 25 mifSN.HéSOh, and 200 m2 conc. H2SOh diluted to one 2). These
trials produced linear standard.curves with high correlation coeffi-
cients. The results also failed to show any conversion of nitrate
,to ammonia based on mixed nitrate ammonia standards.

During the first algal microcosm experimental trial, the alkaline
persulfate digestion procedure of D'Elia gp_a;, (1977) was used on
both filtered and unfiltered samples. Final determination of N0 -N

3

plus N0 -N was made using the cadmium reduction procedure previously

2
described. This analysis was discontinued prior to the second experi-
mental trial due to somewhat erratic results (Table A2).

Kjeldahl digestions were performed on samples taken from both
trials. During Trial 1, dissolved and total Kjedlahl analyses were
performed on.samples taken about every third day. In order to fit
both sets of samples on a single standard curve, the filtered samples
were concentrated threefold prior to analysis. This procedure was
validated by treating some low standards in a similar fashion and
noting that they fell where expected on the standard curve. Total
Kjeldahl samples of 30.0 mi were digested in 50 ml volumetric flasks
using 3.0 mi of digesting solution. After the sample began to fume,
it was digested for an additional 30 minutes at the highest setting on
the hotplate and then diluted with 30.0 ml of deionized water. The
filtered samples used for dissolved organic nitrogen were treated in
exactly the same manner except only 1.0 mi of digesting solution was
used and the final dilution was only 10.0 ml. This procedure concen-
trated the sample threefold without changing the final pH of the

digested sample.

22

During Trial II, the same digestion procedure was used but the
total Kjeldahl samples were reduced to 10.0 mi. All samples received
1.0 m2 of digesting solution and were diluted after digestion to 10.0
ml. In this trial, separate standards were used for the total Kjeldahl
samples and dissolved Kjeldahl samples.

Despite the fact that preliminary examinations of possible
nitrate interference in Kjeldahl tests showed that no problem existed,
such a problem did arise during the course of the experiment. Tests
of mixed nitrate-ammonia standards during the second experimental
trial showed high nitrate concentrations greatly reduced measured
Kjeldahl nitrogen whenever digestion was carried out long enough or at
a high enough temperature to insure complete digestion. It is believed
that the discrepency between these results and those found in the pre-
liminary studies are due to changing the location of the hotplate.
This move resulted in increased line amperage and subsequently higher
digestion temperatures. The problem of nitrate interference was con-
firmed by two other laboratories which routinely run some type of

Kjeldahl digestion.

Carbon - Direct Measurement

A Beckman model 915A Total Carbon Analyzer was used as an inde-
pendent measure of both organic and total carbon. The analyzer con-
sists of a sample injection-furnace module and a model 865 Non-
Dispersive Infrared Analyzer. Analysis is based upon conversion of
sample carbon to carbon dioxide in the furnace module and subsequent
infrared determination. Although this machine provides separate

furnace units for inorganic and total carbon fractions, all analyses

23
reported here are based upon combustion in the total carbon furnace
(950 0).

Samples analyzed for total carbon required no pretreatment.
Those analyzed for organic carbon were acidified to a pH below h.5
with concentrated phosphoric acid (Gordon and Sutcliffe, 1973) and
purged with nitrogen to remove all inorganic carbon prior to analysis.
Inorganic carbon was calculated by the difference between total and
organic carbon. It was assumed that all inorganic carbon was soluble
due to the monovalent character of the media.

All values reported in Table A2 are the average of three injec-
tions. Standard curves were prepared from solutions of anhydrous

potassium biphthalate.

Carbon - Calculated

Previous microcosm experiments carried out by Atherton (l97h),
Garcia (197k), and Hill (1977) have based all carbon kinetics on cal-
culations of the total inorganic carbon present in the media as deter-
mined by pH, temperature, and alkalinity considerations. Young and
King (1973) demonstrated the utility of this approach in batch culture
microcosms. In this study, an effort was made to measure organic car-‘
bon directly in order to correct for any errors introduced by the
algal loss of organic matter. However, as there appeared to be more
variation in direct organic carbon measurements than in calculated
carbon values, calculated values were used as the basis of calcula-
tions of carbon uptake kinetics.

Carbon calculations are based on equations derived by Harvey

(1957). Park (1969), and King and Novak (197k). Total inorganic

2h
carbon was obtained from tables generated by an equation based on pH,

alkalinity, and the dissociation constants of carbonic acid.

—K-l—+H+K2
ECO = a (l)

2
H + 2K2

 

where: 2C02 Total inorganic carbon, moles C/l

no
II

Carbonate-bicarbonate alkalinity, corrected for
hydroxyl ion concentration, eq/£

211
ll

Hydrogen ion concentration, moles/2

First dissociation constant of carbonic acid

K1

K2

Second dissociation constant of carbonic acid

The total carbon fixed over any period of time can be calculated

from the change in total inorganic carbon over the same time period.

Cfixed = AECO2 = ECO2 initial - ECO2 final (2)
As it is thought that algal cells exclusively use free carbon
dioxide as an inorganic carbon source (King and Novak, l97h), the con-
centration of this form of carbon was also calculated

H2
CO =

2f a K1 (H + 2K2) (3)

 

where: C04 Free carbon dioxide concentration, moles C/z

a = Carbonate-bicarbonate alkalinity, corrected for
,hydroxyl ion concentration, eq/l
H = Hydrogen ion concentration, eq/£
Kl = First dissociation constant for carbonic acid
K s Second dissociation constant for carbonic acid

Growth Rate Calculations

Throughout this study, algal growth rate refers to the rate of
algal fixation of carbon, and total carbon fixed will be taken as
equivalent to algal biomass. Algal biomass accrual is usually

described by the first order growth equation.

Mt = M0 eugt (u)
where: Mt = Mass at time (t)
MO = Initial mass
ug = Specific growth rate

In practice, the specific growth rate (”8) is often determined
by normalizing the growth rate for the average biomass present

during a given time interval.

___“f___ (5)

where: “g = Specific growth rate, hours"l

ACf = Change in carbon fixed, mmoles C/z
At = Time increment, hours
x C = Average carbon fixed (average biomass) over the time

increment, mmoles C/z

RESULTS AND DISCUSSION

Algal Microcosms
Microcosm pH
The microcosms used in this study were designed to minimize
possible atmospheric recarbonation of the media so that carbon
kinetics could be accurately measured. As free CO2 was removed
from the microcosms,the following equilibrium equations were shifted

to the right.

__> +
HCO ‘— co2 H20

.. ..
HCO —-‘*‘__ 002 + OH

.____> '

Thus, the carbonate-bicarbonate alkalinity system provided the carbon
source needed to maintain photosynthetic activity. As illustrated by
the equilibrium equations, the liberation of free CO2 was accompanied
by an increase in 0H concentration with a resultant increase in pH.
Figure 2 illustrates the pH response with elapsed time of the micro-
cosms of Trial I. The two microcosms with the lowest initial nitrogen
concentrations at each’alkalinity were rerun using new media after
about hOO hours, thus the curves presented are discontinuous. With
the exception of the microcosms containing low alkalinity media and the
three highest nitrate levels, it is apparent that the extent of pH
change was related to the initial nitrate level of the media. The

three excluded microcosms all contained nitrate concentrations in

26

27

 

.H adage mzzzo mothaso mfihdwdS dzohoano do momanmop =2 acetate 05:. .m “thaw:

 

 

 

 

 

2.3: .2...» . 3.3: .23:
3o..3o.8o.o_ovtoo«.. o@.ooo.o.ootomvsomuab
..od .66
ed M O 6.0 M
O O
.... . ....
o .. .. . .o.a m... .. :3 w.
. D M
a a u-
x v
4\ W .. I.
c a. m...
rod m . .06 .II:
0 m...
..a. . M
w I. I. w
-92 a - 6.2 a
4.4 a a D. W
4 m. 0
9o. 3 - 92.
as. 5... . . n.1,.
. .. . . .0
o 3.. -4 , I I . . ....\o .2... -4
.36 do»... ..a . Seasoned . .
noodfiaod ..o .o: *7 83.390 .0 a:
(as .252. BE... <2... .2 he: .23...

28
excess of what the algae required to deplete the available carbon
(Table A2), as did the fifth microcosm of the higher alkalinity series
which attained the highest pH recorded during this study (11.225).
Algal growth in each of the other microcosms appears to have been
nitrogen limited, but until that limit was approached, the rate of
change of pH was nearly uniform among the cultures. Declines in pH
towards the end of each trial were the result of recarbonation from
respiring algae and bacteria, and leakage of cellular materials into
the media.

Figures 3 and h illustrate the time related pH changes in the
microcosms of the second trial. The stepwise increases in the pH of
the cultures with low initial nitrogen concentrations displayed in
Figure 3 are the result of periodic additions of nitrate to the media.
In three of the four cases, the maximum pH attained by these micro-
cosms was essentially the same as that reached by microcosms contain—
ing medium of the same alkalinity and receiving the same total amount
of nitrate at the beginning of the trial. In the one case which was
an exception (low alkalinity: 0.327, 0.607, 0.887 mg N/l), the algae
failed to assimilate the final nitrogen addition. This is apparently _
because the culture was stressed to the point where the cells were
unable to respond to the more favorable nutrient conditions.

The sudden plunges in pH illustrated in Figures 3 and h are the
result of recarbonating the cultures by bubbling them with laboratory
air. This was done to determine if significant additional growth
would occur in cultures which appeared to be nitrogen limited when
free CO2 was added to the microcosm and the pH dropped. Some increase
in pH occurred in each of the recarbonated microcosms, but the extent

was minimal and inconclusive.

29

 

.HH Adana wsuuao magmmHs> ddaouoazo Mo moanedso somoauw: 3oa mo monsoamoh =9 ompmaon as“? .m whom“:

 

 

  

owcw fluvimw Am :0 Hd

 

:3: .22; . . 2.6: . 22:.
8o..o€.ooc.opwioow.oo o§.ooo.ooo.3w.3u o
.3 ed
10.0 AU." 3.0
O
I,
r . H r.
co m. do
u.
. w
.26 m... I;
m...
n...
‘
5.9m . 66.
D.
m. i
3.2 ad.
Bod -b . 3.6.5
Sod . 8a “and - a .9. some .2890 ..Rnd 6 d.
33 .omd Eod-o 5nd .36 Sad ..o 2
to... 2...o2 .22... .xcs.2-.oz .22...

 

 

3O

 

.HH Hague madame mfiudmas> daaoaoano mo mopzuaso opaofiadon do noncommoa no abandon use? .: manna:

:20: . 2......

ems. eager. .2... .2...

 

scan--
2.....-4
e . .. 31-0
.\ s 2 oz 3......

 

.66.

0mm Knuuoxlv ubm :0 Hd

10.0—

no...

 

 

‘

2:0: .22....
000 . 00¢ . owv . Gina

4 ‘ 1

so: .. o
2;... .. o
.\ a... 2.32 .....e.

for

.0...

 

mow Knunmuv ubm :0 Hd

31

As previously mentioned, the final pH levels of those microcosms
which received stepwise nitrogen additions were similar to those micro-
cosms receiving the same total amount of nitrogen as a single initial
addition. Figure h illustrates that replicate microcosms in Trial II
also showed good reproducibility, suggesting variability resulting
from.slight differences in individual microcosms was small. Care
must be taken when comparing changes in pH between cultures as the
alkalinity of the media determines the extent to which the hydroxide
ions released during carbon assimilation affect the pH. This precludes
direct comparison of microcosms of differing alkalinity from a single

trial or direct comparison of microcosms from two different trials.

Carbon Fixation

Throughout this discussion, growth is considered as carbon
accumulation or net productivity in terms of carbon. For the purpose
of these calculations, carbon accumulation was determined by changes
in the amount of inorganic carbon present in the media as calculated
using Equations 1 and 2 of the Analytical Methods Section. Using
the high alkalinity microcosm with the highest initial nitrate con-
centration, the conversion of pH data to inorganic carbon in the media~
and organic carbon accumulated is illustrated by Figure 5. It should
be emphasized that carbon uptake figures created from reductions in
the media inorganic carbon represent the total organic carbon of the
microcosm, not just the carbon within the algal cells.

Data showing net carbon fixed for selected microcosms from both
trials are presented in Figures 6 and 7. As would be expected from
the way the data were generated, these curves closely resemble those

presented earlier for pH changes. However, because these figures deal

32

2:2. .22:

l

com

.033 :33 .5930 cox: EB canning: 28900.82. 5 noggin;

CD
1.0
(0

flow;
-ON.N

unufi

waxy

 

‘

o

.oo...

ICON

..OOfi

 

.m mummHm

P“!J

udqmo

mmpaw ug

uoqaoo

l/«qomw

l/ulomm

33

.nsnooououa 2323333 am“: 35038 Mo coupons—«nod 259.30 can ammonia 5 n:o«aa:d> .w chow;

 

 

 

   

2:22.22... 23: .223.
8...... one. Bo 8o 8.. 21 .00 one. 8o 8... 3.. atom -oo
r00. .80.
w w
s 5. m .o. w
m . ..m.
0 .N:
6.. m .2. .m.
u w
o .... o. ..u
. WM. I.
a . D
. .o... P . . . .2... a.
. U
I'll-IIIII‘
. . mum: -
.o.~ timid ram.
0 mimic.
o: f
.o.n 530322 on.

 

 

3h

.mamooouowa hawcfiddxad 30H cmuomaom Ho negadaaaummd conndo cad sewage“: :u acofiuaand> .> muzwfik

 

 

uum uaDomN l/anom u:

2.3: .35.. 2.5: .2...»
8w. owe Be So ooh pom oo com.— com 8o 8.. com . o
. . -..... . . . 68.
. . . w
w . o u n
o a .
gen. m. :80
V .
c O 4 4 4 q 4 a 4 4 4 \
a a . p . m
a . 6... q 68.
o
. U
. u
x
4 g .
. . . . .oc. D. o .09.
< O .
o ..N-..
. o o mawuu
.. rub. 0.9.4 .69.
. . 3-4
0 o h..no
w m....
. :om. 830222 .59.

 

 

3h

.mamooouowa huacfiddxad 30d empowaom no sodadaaafimmd conndo can :mwowuuc :H muoHpafiad> .P mpzwam

cow. coo.

b

.52. .22....
00.9 com

o

o

1....

6?.

6h.

 

valoww

pox” uoqwa

9:3: .2...»
one 00¢ com

b

Q.

o . TN..-

0 o mfil a

n.~..<
0....4
.... no
9.-.
£323.23

fine.

.6. ~..

 

uoqm unboum l/auom LU

35
with carbon changes and not pH,it is possible to directly compare
growth rates of cultures with different alkalinities. Again it is
apparent that until the cultures approach their nutrient limitation,

or stationary phase of growth, they all show similar growth rates.

Nitrogen gptake
The rate of assimilation of nitrogen by algae in selected micro-
cosms is also illustrated in Figures 6 and T. The data used in con-

struction of the figures were determined by changes in NO -N plus

3

N02-N measured in the media. Ammonia was ignored throughout these
computations as the measured values were variable and low, usually
below 0.03 mg NH3-N/1 (Table A2).

Examination of Figures 6 and 7 reveals that inorganic nitrogen
uptake is completed prior to the termination of carbon fixation in each
of the microcosms. Rapid nitrate uptake, especially by nitrogen"
starved cells,has been reported by many investigators (Syrett, 1956;
Dugdale and Goering, 1967; Eppley, Rogers, and McCarthy, 1969;
Yentsch, Yentsch, and Strube, 1977). McCarthy and Goldman (1979) have
recently suggested that uptake rates in marine algae may be rapid
enough to allow phytoplankton exposed to intermittent pulses of
nitrogen to exhibit near maximal growth rates,even when found in
nitrogen depleted waters.

Rates of nitrate assimilation for some species of algae have
been found to be related to the nitrate concentration of the media and
are adequately described by the Michaelis-Menten equation (MacIsaac
and Dugdale, 1969; Caperon and Meyer, 1972b). Efforts to examine the
rate of uptake of nitrogen as a function of external nitrogen concen-

trations in this study resulted in figures which were generally the

36
shape of the expected rectangular hyperbola with a maximum uptake rate
of about 0.0018 mmoles N/mmole C/hour but displayed a very large
degree of scatter. Part of this variance may have been due to the
effect of internal nitrogen concentrations on nitrogen uptake (Droop,
l97h), but a large amount of the scatter is likely due to measurement
errors. Due to the variance and the fact that uptake rates were still

relatively high at media N0 -N concentrations of only 0.01 mg N/l,

3
very near the limit of reliable determination for the current study,
it was decided that the data available would make further analysis of

nitrogen assimilation kinetics fruitless.

N/C Molar Ratios

Due to differing rates of nitrogen and carbon uptake, the nitrogen
to carbon molar ratio of algal cells varies throughout the time course
of an experiment. The N/C molar ratio of selected nitrogen limited

microcosms, as calculated from changes in NO -N plus NO -N and inor-

3 2

ganic carbon, are presented in Figure 8. The shape of the curves

shown for microcosms 2:5 and 2:9 are characteristic of those micro-
cosms which received all of their nitrogen at the beginning of a trial.
The initial point of each curve is the N/C ratio of the seed as deter-'
mined using Kjeldahl digestions and total organic carbon analysis.
Initial rapid uptake of nitrogen results in a high N/C ratio which
continuously declines until some minimal value is reached. Slight

increases after that point are the result of respiratory CO lowering

2
the calculated carbon content of the cells. These curves are similar
to those reported by Atherton (l97h) for Q, vulgaris and other species

of algae which took up nitrogen in excess of their needs.

37

mcofiucaam>

 

 

.mSmooouoHa copooaom scum woman no Canon sonhdo Ou somoaufi: one :a ..c ohsmum
2.3: .23....
com com one 8.. com one on? com 09m 00. o
- A.
.No.
.Vnw
F8. w
..I If} . ”o. m
Oillllli OI:/: \\\o|..||o/p/ \MC. .1; . . o w
.Illlllllllll‘.l'.l' .p\.|lllll 0"" so /I o N
C . nooCoou. lulll.’lll‘l\.l‘/ / .0 .‘T9. 0
.. / .. / .. . . m
/ \. ‘ / . . ..N_. W.
. k c / I/ o o u
. , . . ... .. m
/ / ... r... m
n l / a/ o. o R m
- ooo‘. lQ—o m.
. ..
. u: ..
gm: 5 € ION.
o.~-< // ..
...m-.. 4/... .N...
..N... 11*
52.2.3.2 rem.

 

38

The curves shown for microcosms 2:1 and 2zh are typical of those
microcosms which received periodic additions of nitrate. In addition
to the original nitrogen, nitrate was added at 350 and 635 hours.

Each addition was followed by an increase in N/C ratio, with each
increase being of lesser magnitude than the preceding one as the addi-
tional nitrate became a smaller fraction of the total nitrogen taken.

Cultures which were ultimately carbon limited displayed trends in
N/C ratio which were similar to those shown by the nitrogen limited
cultures in Figure 8. Although the N/C ratio of carbon limited micro-
cosms was higher (mean 0.llh) at the cultures' maximum.pH than that of
nitrogen limited cultures (mean <0.089), they were much less than the
maximum N/C ratio exhibited by the carbon limited microcosms (mean
~0.2). Thus, although large amounts of excess nitrate (as much as
3.85 mg N/Z) remained in the media, N/C ratios did fall considerably
prior to cessation of growth. These findings directly contradict
those of DrOOp (1973) who reported that the cell contents of a given
nutrient remained high when cultures were limited by a second nutrient,
and negate the possibility of using cellular concentrations to determine
the degree to which a nutrient is in excess of requirement (Droop,
l9Th).

Various explanations for this discrepancy can be proposed. One
is that in these bottles, nitrate uptake was inhibited by a lack of
free C02 through the need for carbon compounds during the first step
of nitrate reduction (Kessler, l96h; Morris, lQTh). A similar inhibi-
tion of nitrate uptake was reported by Ketchum (1939) when phosphate
was the limiting nutrient. This pattern also appeared with respect to
vitamin B in the phosphorus limited chemostats of Droop (l9Th).

12

Another possible explanation is that the cells in these microcosms

39
were not exposed to the high nitrogen concentrations they were thought
to be exposed to. This is possible if local nutrient depletions
occurred around the cells which had settled to the bottom, not an
unlikely event considering the quiescent conditions present in the
microcosms (Gavis and Ferguson, l975; Gavis, 1976). A third possible
explanation is that cells which were once nitrogen starved store
large amounts of excess nitrogen while those with a history of expo-
sure to adequate nitrate concentrations store nitrogen to a smaller
degree. No evidence has been found to support any one of these pos-
sibilities, but the last of the three seems most unlikely.

One of the major objectives of this study was to verify the con-
clusions of Atherton (l97h) and Garcia (l97h) concerning the effect of
media pH on algal C/N molar ratios. A graph of C/N molar ratio, as
calculated from seed carbon and nitrogen content and uptake of these
two elements, against maximum media pH is presented in Figure 9.
Ekamination of this figure does show a trend of decreasing C/N ratio
with increasing pH, even when carbon limited cultures are omitted;
but it also displays a large amount of variation in the ratio at any
given pH. The data presented by Atherton (l97h) and Garcia (l97h)
showed much less variation in C/N ratio for nitrogen limited cultures
and much higher C/N ratios at low pH levels (Figure 10) than does the
.data of the present study. Closer examination of the work presented
by Atherton and Garcia revealed that both investigators failed to
consider the nitrogen introduced into their microcosms with their
micronutrient media when calculating nutrient uptake. When this nitro-
gen is included in the calculations and estimates of C/N molar ratios

are made from the data available in their reports, the plot of C/N

 

hO

O-Trlal I Low Alkalinity

 

 

pH levels.

26m o-Trial I ngh Alkallnl‘ly
A-TrialII Low Alkalinity
A-TrlalII quh Alkallnlly
24- o
221?
20-
. 0
l84
\ ”l A. o
l I I 3'6?
:; o
I '1 m“:
3
‘7. 312. ' 'A A A
\‘ :E A. .A9 '0
m 8 /
0., . ’,/'
R? l 2‘0" A ‘ A/ o
; \ O
o / O
.‘l 8-» A '
3 Carbon limited
.7 2 6- Mncrocosma
.v 4-“
g 24
as $0 9T5 lo.o lO'.5 ”To
Maxlmum Mlcrocosm pH
Figure 9. Algal carbon to nitrogen molar ratios at maximum microcosm

Lil

 

 

DATA APPROXIMATE
M— ALHALLNJL‘L.
so. 0 o- Atherton 4 meq/l
o- Garcia 4 meq/i
A- Garcia 3 maq/ l
a- Garcia 2 meq/ l
O- Garcia l 111qu l
50~
.9.
‘6
a: A
E 40d
o
2
3
a. a
g 30‘ "e
E
\
3
.0 O
3 20~
o
. .
O o O
A D A a O 0
K3. . r- (3 o 51'
I Carbon 0 a I
L__':_""_'.’_"’_ _ __ 2. _ _l
o F ' U1 I T I
8.5 9.0 9.5 l0.0 IO.5 l LG I |.5

Maximum Microcosm, pH

Figure 10. Carbon to nitrogen molar ratios at maximum.microcosm pH
levels as reported by Atherton (l97h) and Garcia (197h).

A2
molar ratio versus pH changes completely (Figure 11). There is no
trend in C/N molar ratio with pH apparent from their revised data.

In addition to the trend in C/N ratio with pH, Figure 9 displays
a distinct difference in the response of the low and high alkalinity
microcosms. Almost all of the low alkalinity microcosms (closed
symbols) displayed lower C/N ratios at a given pH than did the high
alkalinity microcosms (open symbols). Reexamination of Figure ll
shows a weak correlation between the relative C/N ratio and media
alkalinity at a given pH, with higher alkalinity microcosms displaying
higher C/N ratios. This correlation between C/N ratio and alkalinity
suggests that perhaps the free CO2 concentration of the media is a
determining factor.

Figure 12 displays the relationship between C/N molar ratio and
the free carbon dioxide concentration of the media for the nitrogen
limited microcosms. The data show slightly less variation than seen
in Figure 9, and the clear distinction between high and low alkalinity
microcosms is eliminated. Plotted in this way, the data suggest that

C/N ratio is determined by free CO concentration under nitrogen

2

limited conditions, and that the ratio remains essentially constant

except at relatively high free CO levels. But, in view of the error

2
made by Atherton and Garcia, the five points which constitute the

upward trend at high CO levels merit special attention.

2
All five of the points with C/N molar ratios in excess of 13
represent high alkalinity microcosms which experienced relatively_
small increases in pH. Because of this, small errors in pH measurement
would represent a relatively large proportion of the total pH change

and a large part of the total carbon fixed. All five of the micro-

cosms in question also had relatively small amounts of nitrate nitrogen

’43

 

 

 

 

26.» DATA APPROXIMATE
.JQLLBQL— ALBALLNJIL.
o- Atherton 4 meal!
24-) o- Garcia 4 meq/l
A- Garcia 3 meq/i
a - Garcia 2 meal!
22-» 0" Garcia i meal!
20-.
is...
$3 Ki- 0
g . O O . 0
HF 46
:3 A o a. 0
§ i2- A
D
Z '01 3 C
\
0 i o .
8- .
a
0
$1
4‘ Carbon 0
Limited
2‘ * _l
L‘— __ _. .... _. ._ .... ......
o ‘ ‘i I 1 r
8.5 9.0 9.5 l0.0 l0.5 i i. 0

Maximum Microcosm pH

Figure 11. Corrected carbon to nitrogen molar ratios at maximum.micro-
cosm pH levels as calculated from the data of Atherton
(l97h) and Garcia (197M).

26-

24-

cm Molar Raiio

Algal

4-.

2.

Mi

o- TrialI Low Alkalinity
o- TrialI l-llqn Alkalinlly

A- TrialIl: Law Alkalinity
A- TriOlK High Alkalinity

 

 

Figure 12.

I

- 8.5 b as l.‘o

free 002 ,(14 moles/l)

-l.'o
log
Variations in algal carbon to nitrogen molar ratios at

maximum.culture pH levels with media free carbon dioxide
concentration.

As
in the media (<O.350 mg N/l), therefore any errors in measurement of
initial inorganic nitrogen could constitute a significant fraction of
the total nitrogen ultimately taken by the algae. Measured initial
nitrogen differed substantially from that predicted in h of the S
microcosms (Table A2),adding more doubt to the validity of the initial
values. In addition to the possible errors in the uptake of inorganic
nutrients, each of the microcosms contained a small amount of organic
nitrogen. Kjeldahl digestions for dissolved organic nitrogen proved
too variable to eliminate the possibility that some organic was also
assimilated by the algae. Organic uptake was reported in similar
experiments by Sharp (1977).

The data collected from this investigation does not conclusively
illustrate a relationship between algal C/N ratio at culture maximum
pH and the free CO2 concentration of the media. However, it is clear
that if the C/H ratio of the nitrogen limited cultures at the extent
of growth does vary, there is a stronger relationship between this
ratio and media free CO2 concentration than between it and maximum
media pH. This, coupled with the difference in C/N ratios between
different alkalinity media at a given pH,.shows that the C/N ratio of
algal cells is not determined by the pH of the media; Therefore,

Atherton's and Garcia's conclusions concerning the use by algae of

excess nitrogen as an internal pH buffer are in error.

Cell Leakgge

The importance of leakage of organic material from algal cells
has been a topic of increasing discussion in recent years. Major
articles of interest on this topic have been presented by Fogg and

westlake (1955), Pavoni gt_§l3 (1971), Hellebust (l9Th), and Sharp

h6
(1977). From these articles it is apparent that algal contributions
to the dissolved organic matter (DOM) of aquatic systems is widely
accepted. However, questions still exist concerning the magnitude
and cause of the contribution.

Undoubtedly, the excretion of some organics such as vitamins,
enzymes, chelators, and toxins is functional (Aaronson, 1978; Sharp,
1978), but alarge part of the excretion reported may not be functional.
Sharp (1977) has questioned the magnitude of loss of organic matter
from algae and suggested that there is no good evidence for extensive
excretion by healthy, unstressed cells. Hellebust (197A) noted that
much of the extracellular products derived from cultures in the lag
and stationary phase of growth may be due to autolysis. The combina-
tion of these observations suggest that perhaps most of what is
reported as excreted organic matter may simply be the result of lysis
of cells under stress, but Fogg (1966) reported that as much as 35%
of the carbon fixed by marine phytoplankton may be immediately lost
from the cells. The fact that stress does seem to greatly affect
the extent of excretion has presented problems in determining how much
organic matter algae "normally" leak (Sharp, 1977). This is especially
. true in light of the great variety of factors which can impose a stress
upon natural populations of algae.

Throughout this study both dissolved organic nitrogen (DON) and
dissolved organic carbon (DOC) were monitored (Table A2). Because of
the nitrate interference discussed in the methods section, initial
values of DON were only attainable in microcosms with low initial
nitrate concentrations. For those microcosms in Trial I which had
nitrate concentrations too high to allow initial DON measurements, an

initial value of 0.0058 mmoles N/z was used. This value was an

MT
adjusted average of the initial values of microcosms 1 through h.
For Trial II a similarly derived initial value of 0.0020 mmoles N/z
was used for the high nitrogen microcosms. Using these initial values,
the maximum increase in DON found was about 0.0h mmoles N/z (microcosm
1:10). The percent leakage (increase in DON x lOO/total inorganic
nitrogen taken) was calculated for each microcosm in which nearly all
the nitrate present in the media was depleted. These values varied
between -38% and +125%. The average of those values between 0 and
100% (2h of 26 values) was about lh%. There was no apparent relation-

ship between maximum media pH or free CO concentration and the per-

2
cent of nitrogen leaked.

The time course of DON measurements also was inspected where pos-
sible. Five of the 12 cultures which showed initial DON (i.e., the

3
tial phase of algal growth and subsequent increases during the sta-

low initial NO -N cultures) had decreases in DON during the exponen-

tionary-phase. This is similar to the trend reported by Sharp (1977).
All other microcosms showed generally steady increases in DON through-
out the course of growth. Because of the NO3-N interference with the
Kjeldahl measurement of DON in high nitrogen microcosms, it is diffi-
cult to determine when the increase in DON begins. But, in some micro;
cosms, it appears to start before the beginning of the stationary
phase of growth.

Measurements of dissolved organic carbon (DOC) were only slightly
less variable than those of dissolved organic nitrogen. The DOC
values reported in Table A2 are measurements in excess of that found
in deionized water, which was estimated to be about 1.0 mg C/i. The

maximum increase in DOC found in any microcosm was 0.29 mmoles C/l

(microcosm 1:10). The percent leakage of carbon (increase in

h8
DOC x loo/carbon fixed) was calculated for each microcosm and ranged
from 0 to 25h%. If the high and low values are eliminated as being
unreasonable, the remainder of the values fall between 6% and 60%
with a mean of about 16%.

Inspection of individual microcosms reveal that there may be a
difference in the percent of carbon leaked from high and low alkali-
nity cultures. If only the microcosms which fixed more than 0.5
mmoles C/l are considered (in order to eliminate those which would
show high percentage errors with even small measurement errors), the
low alkalinity bottles of Trial I (microcosms 1:5, 1:7, 1:9) leaked
an average of 22.3% of the carbon they assimilated. In contrast, the
high alkalinity microcosms (1:6, 1:8, 1:10) leaked an average of only
9.5%. During Trial II, the only low alkalinity bottle which fixed
over 0.5 mmoles C/l (2:5) leaked about ll.h% of this carbon while the
high alkalinity microcosms (2:6, 2:7, 2:8, 2:9, 2:10, 2:11) leaked an
average of only 7.8%. When plotted against free CO2 concentrations in
the media, these leakage values create a smooth curve which may reflect
increased leaking as a result of carbon dioxide stress. However, the
microcosms fixing smaller amounts of carbon failed to reinforce this
trend (Figure 13).

Examining the time course of increase in DOC for the individual
microcosms of Trial I revealed that no leakage was detected until after
hOO hours had elapsed. Initially this was thought to indicate leaking
in the senescent phase of the growth curve, but more careful examina-
tion revealed that all the microcosms showed increases at the same time
regardless of the stage of growth. The reason for this is unclear but

it seems that it must be an artifact of the measurement methods used.

 

’49

 

° -Triai I High Alkalinity
60‘ 0 -—Trial I Low Alkalinity
' i a -Trioi II qun Alkalinity '
- -Triai JILow Alkalinity
x -ovor .5 mmoles/i 0 fixed,
5.0...
4.0-
1:
2
U D
0
.J
8 30.. ° °
.9
b
o
0 X
\
a? \ '
2.0.. W X
‘\ . I a a
\
\ I
\
\\ x O a
Lo. ii \ . 'x °
‘gll\?____x_x . .
X
o I f— U T I
-3.0 -2.0 - l. O O i. 0
log free 002 (“moles“)
Figure 13. Percentage of carbon fixed which leaked in experimen-

tal microcosms as a function of the media free carbon
dioxide concentration at maximum culture pH.

50
During Trial II a gradual increase in DOC occurred throughout the
- experiment. The data presented to this point reveal that the overall
percentage of carbon and nitrogen leaked by the experimental microcosms
were about the same (16% and 1h%, respectively). Because of this, the
N/C or C/N ratio of the leaked material is roughly the same as that
contained by the cells. This suggests that the dissolved organic
matter may simply be the product of cell 1ysis,<m'that at least cell
leakage is non-selective in terms of what is released. Therefore,
cellular excretion should not have significantly affected the algal
C/N ratios previously calculated. No relationship was found between

pH or the free CO concentration of the media and the C/N ratio of the

2
dissolved organic matter in the media.

The fact that no clear relationships between media pH or free CO2
concentration and leakage have been demonstrated here certainly does
not preclude their existance. Due to the large variations in sequen-
tial DOC or DON measurements and the fact that no effort was made to
measure excretion at a precise point on the growth curve of each micro-
cosm, significant information on organic leakage may have been masked.
Despite this, it seems apparent that excretion did not greatly exceed
15% of the nutrients fixed by the algae if microbial decomposition is
considered negligible. No information on microbial activity in the

microcosms was collected, but at the high pH levels attained in most

cultures such activity would be expected to be small.

Growth Kinetics
Throughout this study, growth has been measured by net carbon
uptake as previously discussed. In this section I will consider

specific growth rates (mg), or carbon uptake rates normalized for

51

algal density, as. described in the methods section. This parameter may
be important in determining species succession through competition.

Variations in specific growth rate with changes in the free CO2
concentration of the media are displayed by Figures 1h, 15, and 16.
These graphs show not only the rate of growth of individual microcosms
at various free CO2 levels, but also the extent of growth possible
under the conditions of each microcosm (i.e., the point where pg
reaches zero which is approximated by the baseline of each figure).
At the termination of growth all but four of the microcosms (1:5, 1:7,
1:9, and 1:10) had reduced the available NOB-N in the media to less
than 0.03 mg N/2 and were considered nitrogen limited.

The remaining four microcosms, containing 0.22, 1.27, 3.85, and
1.68 mg N/Q, respectively, at the termination of growth, were con-
sidered carbon limited. Each of these microcosms were expected to
cease growth at the same free CO2 level but instead varied over a
range of from 0.0010 umoles 002/2 (1:7) to 0.0026 umoles CO2/2 (1:10).
These differences can only be attributed to measurement errors and
experimental variance between individual microcosms, or perhaps to pH

differences as the microcosm which stopped at the highest free C0 con-

2
tained h meq/2 alkalinity and also reached the highest pH.

In Figures 1h through 19 the specific growth rate is indicated by
the vertical height of the curve presented. The curves in Figures 1h
through 16 resemble rectangular hyperbolas of the type usually
described by the Michaelis-Menten equation. This suggests that carbon
uptake follows the normal pattern of phytoplankton nutrient uptake
described by Caperon and Meyer (1972b). Closer examination of Figures

1h through 16 reveals that a great deviation exists between the shapes

of curves from high and low alkalinity microcosms, especially at lower

52

. .H Adana madame msofiaoausooaoa
oofixOHc confine moan mamas sud: maven zpnoum camwaooo Amman aw meouuaqaa> .aa oasmfla

2222:»... Non. a... 2.32.4 a...

 

 

 

 

 

 

 

 

0.. O 0.... O.Nl o.m l
— n - - ‘ . . on¢|
3
AV _
q .
m.
Au
4 . D C . ..OHI a”
. I
W.
O m
(33.3.2 0 J
noon .o— . . mW\ hm
...—on. .93....
o\\
O . . l\ i I
\i. . w o... -0 .... ..o o...

;:sss-mmwi . o...;. h.. in

9. l4 0... l4

¢.. lo n: lo

 

53

.HH downs no msmooonoaa hufiaaaoxaa 30H ca

acoapoApsoosoa omfixouc sophoo ovum mamas new: money apron» owmfioomo Human a“ escapofiao> .ma oasmfim
22.22.. 3 N00 00.. 332.4 co.
0.. ad m. 0.0.. 0.... 0..: 0a...
P _ [r _ - P 0.?!

 

 

 

 

(..moul blr 5°l

 

 

 

.oa-

 

 

 

 

5h

.3 2...... ..o 2.888.... 3.52.? and. a...

 

 

 

 

 

 

 

 

m:0waaausoo:oo oGHxOHm sonuao 00AM mamas saw: woven sauoum camaoogn Amman a“ m00wpowao> .md oasmwa
..\ao.o... 3.. «00 3.. «no.2...
0.. “.0 D40 C.Ni
0.7.
m.
6
> 6”
C
iO.MI \l
u.
m
. a ......
. \ at... D
o . 9...... O
. on- a
¢.Ni< TO.NI
g\\b.\ N..N to .
. leg
.4. .

 

55

Mlgrggggm I

 

 

 

 

o'-l:lO
a-|:9
A-l:8 ‘ég .° .
A-|:7 a O
‘2'04’ .—l:6 °
o-I=5 O o
A).2. o
o
4‘
O o
-25- o
A a
o
A:
‘0
o
T; ‘301. A.0
g C
o
2;
g a!»
a
I
-4.o.
a
-45-
‘ -l2 ‘ -l.o ' -63 ‘ -05 ‘

-i.4
log N/C Ratio

Figure 17. Variations in algal specific growth rates with cellular
nitrogen to carbon molar ratios for high nitrogen, Trial I

microcosms .

56

 

 

 

 

 

W a .
o- am 8 A
a- 2:9 .‘
A. 2:8 A
-20. 0- 2:7 ‘ %&
o-2:6 o .
“'2=5 .
l-Zzii ' O O
T0 o
o
a
4a 6 '2
A.
A
. O
-30d)
3 O
, o
b
g n
.0 o
v I’
O
q -3sl ' u
.130
0!
2
.A
o
4.0-4
' I
4.5- A
1 r I Hi I r I I I
- L4 --L2 ‘L0 -0.8 -O.6

log N/C‘ Ratio

Figure 18. Variations in algal specific growth rates with cellular
nitrogen to carbon molar ratios for high nitrogen, Trial II
microcosms.

57

 

 

 

 

Microcosm
o- I:|
O A . a- he
a O A Q A A- |:3
a 0 Q- |:4
-20.: a . . ‘ .- :32'
$ 0% ‘. .- '
i °° "3:2
I .- '
O o A X
I
J 9 .
-2.5- . I
a oI ‘
O
I
o.
-ao. ,0
A
.. A. A
Ib .
E o A ‘
V o
0
R as.
o ' .
3 A
O
-40..
4.5.. :
n a a. A
-l.4 ' -72 -I.O ’ -o.a ‘ -b.s ‘
log N/C Ratio
Figure 19. Variations in algal Specific growth rates with cellular

nitrogen to carbon molar ratios for low nitrogen microcosms.

58
free CO2 levels. One possible explanation for this is a difference
in light intensity reaching the algal cells. The high alkalinity
microcosms fixed much greater amounts of carbon (attained much higher
biomass) than did low alkalinity microcosms at any given free CO2
level. For example, microcosm 1:10 fixed three times the carbon
fixed by any low alkalinity culture, and it exceeded the total carbon
fixed by any low alkalinity microcosm after only 380 hours. In all
the microcosms, a major portion of the algal population settled to
the bottom of the flask due to the quiescent conditions present. In
those microcosms with very dense algal populations, the cells piled up
and shading maylunmeoccurred resulting in reduced specific growth rates.
Another possibility is that this build-up of algae on the bottom
causes a local zone of lower nutrient concentration or total nutrient
depletion which affected overall microcosm growth rate (Gavis and Ferguson,
1975; Gavis, 1976). If either of these phenomena occurred, the
use of quiescent batch cultures to determine specific growth rates may
be unwarranted.

Attempts to relate specific growth rate to external nitrogen con-
centrations with a simple expression have proven fruitless due to the
uptake of excess nitrogen. But, success has been achieved in defining
simple relationships between internal nitrogen and specific growth
rate by Caperon and Meyer (1972a, 1972b) and Droop (1973, l97h).

Figures 17 through 19 represent specific growth as a function of N/C
molar ratios of the algae (an expression of relative internal nitrogen
concentration). None of these figures clearly display the rectangular

hyperbolic relationship predicted by the Michaelis-Menten equation and

reported by Droop (1973) and Caperon and Meyer (1972a), although they

59

more nearly approach the expected shape than does the Thalassiosira

 

pseudonana data presented by Goldman and McCarthy (1978). These
authors explained the deviation of their data from the expected form

by suggesting that the maximum attainable growth rate for Thalassiosira
pseudonana is well below the maximum theoretical growth rate (Droop,
1973), and therefore the figure represented only the extreme left end
of a Michaelis-Menten type relationship. The current data for Chlorella
vulgaris display maximum specific growth rates which probably more
closely approximate the theoretical maximum growth, as curves of “8

as a function of N/C ratio tend to level off at higher nitrogen to
carbon ratios. This is most apparent for the higher density nitrogen
limited microcosms of Figure 17 (closed symbols) and Figure 18, and
could be made clearer by expanding the horizontal scale of these
figures. If only the data points to the right of the vertical lines
(N/C molar ratio of over 0.168) in Figures 17 and 18 are considered, it
does not appear that the specific growth rate varies with cellular N/C
ratio. This makes it possible to predict when external nitrogen may

be controlling algal growth rates.

Figure 20 illustrates how cellular N/C molar ratios vary in
response to the nitrogen concentration of the media for some Trial II
microcosms. The horizontal line in this figure represents the same
cellular N/C molar ratio as the vertical lines in Figures 17 and 18.
The curved line is an approximated best fit curve to the available
data. If this curve is accepted as the true relationship between the
two variables and it is assumed that maximum growth rate is achieved
when the cellular N/C ratio exceeds 0.168, Figure 20 implies that under

laboratory conditions whenever media nitrogen exceeds about 0.10 mg N/i

.nanooOAOAE dopooaom you saves as» ea

 

 

 

 

 

somoaafis owsamaosa was moupou Adaoa sowouaus ow sonaoo Adazaaoo avenues manuaouuoaom .om oAsMHa
52 9... 3.3.2.350 Suezz ...—.22
ON. GO. ON. C
I P h o
...m I
n:.N n.
.wnm n.
O «N 4 VD .o m
...~ 0 mi...
@«N C ..0-
ad 4
52.8.3.1 N
.00 2. w
w.
m.
D
4 ..
.... u.
o a m.
U 04 I. o
llllPlIlllllllllill) 4 Q ..ON.
o
O

 

61
the rate of algal growth is not controlled by nitrogen but must be
controlled by carbon or some other parameter. Below this level,
cellular nitrogen is reduced and nitrogen uptake ultimately limits
growth.

It should be noted that the first sets of data from each micro-
cosm often showed lower cellular N/C molar ratios than expected.

These data do not appear in Figure 20 as they occurred at external
nitrogen concentrations higher than the scale of the figure. Data
from microcosms which received little nitrogen showed much more varia-
bility than those presented, while those from the higher nitrogen
Trial I cultures displayed somewhat lower N/C ratios at a given exter-
nal nitrogen concentration. Cultures which were carbon limited pro-
duced cellular N/C molar ratios lower than those of nitrogen limited
cultures at a given external nitrogen concentration. This variability
in N/C molar ratio with respect to external nitrogen concentrations
makes any further effort to assess the role of nitrogen in controlling
specific growth rate unwarranted.

Overall, it appears that in the laboratory microcosms used for
this study, light intensity controls specific growth rate when all
nutrients are provided in excess. As either carbon or nitrogen
become scarce, specific growth rates decline. Carbon limitation is
characterized by'a Michaelis-Menten relationship between specific
growth rate and media free CO2 concentrations. The uptake of excess
nitrogen precludes such a simple relationship between specific growth
rate and media nitrogen concentration. However, it appears that the
rectangular hyperbolic curve does describe, to a limited extent, the

relationship between excess internal nitrogen and specific growth rate.

62
When external nitrogen concentrations are high enough to maintain
high internal nitrogen stores, growth is controlled by some other
factor. When internal N/C ratios drop below some critical level,
nitrogen uptake ultimately controls growth rate unless the decline

is the result of a shortage of some other nutrient.

Ammonia Volatilization
Ammonia loss from water in small plastic vessels was measured
over 89 time periods as described in the Materials and Methods
section. The conditions under which these trials were conducted,
and the result of each trial are presented in Table B1. In this

table and throughout the following discussion the terms "total

ammonia" and "ZNH3" refer to the summation of the NH3, NHhOH and _
NH“ forms, while the terms "ammonia" and "NH3" refer to the
unionized forms (NH3 and NHhOH) only.

Fick's Law

Passive flux of a compound across a given interface is a
function of the difference in concentration of the compound on the
two sides of the interface. This relationship can be described by

Fick's first law:

’1']
'll

KC(AC) (6)

where: F flux of compound C across the interface

(mass - length-2 time'l)

Kc = permeability or mass transfer coefficient of compound
C (length . time-1)
AC = concentration difference of compound C across the

interface (mass - length’3)

63

From this equation it is clear that a prediction of flux at a
given AC is dependent upon the mass transfer coefficient, Kc'
Knowing the flux for each of the experimental vessels and calcu-
lating ANH3 (as only the unionized form is transferred across
the air-water interface), it should be possible to determine the
ammonia mass transfer coefficient (KNH ) for the conditions
which existed during the experimental tiial.

When dealing with gas transfer across an interface between
two different solvents, it is necessary to consider the relative
solubility of the compound in each solvent during the calculation
of AC. If the ammonia concentration of the atmosphere is known,
the unionized equilibrium ammonia concentration at the surface

of the liquid can be determined from:

liH3_ew = KH NH3_a (7)

where: NH3_ew = the NH3 concentration of the water in equilibrium
with the air

NH

3-a the NH3 concentration of the air

KH 1159.73e-'O2h83 T, Henry's law constant derived
from Bunsen absorption coefficients (Lange,
1967) in which T is temperature in degrees C
(r2 = .999)

Due to difficulties with the apparatus, only one measurement of
background ammonia concentration in the air was successful. This
trial produced an estimate of 3.8 x 10"10 moles NH3/t, reasonably

close to the background concentration of 2.35 x 10"lo

moles NHB/E
reported by Rasmussen, Taheri, and Kabel (1975).
Unionized ammonia concentration of the bulk solution depends upon

the total ammonia concentration and the equilibrium constant for

Equation 8.

6h

NH3 + 3+ ._-—.—2 NHh+ (8)

Emerson gt_al, (1975) developed Equation 9 to predict the fraction of
total ammonia which is unionized (uf) under any temperature and pH

conditions.
1

uf = (9)
(1 + loPK'pH)

 

where: pH = 0.09018 + 2729.92/(T + 273)
pH = pH of bulk liquid
T = temperature of liquid in degrees C

This allows ionized, unionized, and total ammonia to be expressed in

terms of each other as follows:

 

NH3_w = uf 2NH3_w (10a)
NHh_w = 2NH3_w - NH3_W (10b)
NH,4 = — NH (10°)
‘V uf 3-W
NHh-w = (l-uf)(ZNH3_w) (10d)
where: NH3-w = the concentration of unionized ammonia in the water
NHh-w = the concentration of ionized ammonia in the water

Using the results of Equations 7 and 10 to calculate the difference

between NH and NH w (ANH ) for each experimental container, and

3-w 3-e

knowing the measured f1ux,it was possible to determine the relationship

3-w

between flux and ANH3_W.

ANH Mediated Flux
———.3 -w

If the mass transfer coefficient is a constant, as expected under
constant laboratory conditions, flux should be a linear function of

ANH3_W. igure 21 clearly demonstrates a relationship between flux

65

 

 

0 Trial A
.0 Trial 8 AA
-3- A. Trial C ¢ A,
A
on
.A o
95
A ,, A
2A ‘
A .°‘
-4. ° ..
0 , o
AA
'2 A"
.A .2
\ .
on mo
5 . ‘
Tn’ .
" A.
i; I
.A
5 A
O
a: '5‘ A
2 A ‘ -
LL .
A. o
A KNH3-w= l CHI/hf
0 ,. A
2 A
O
0
-e r J r T t
-7 -6 -5 -4 -3

log A N H3-.. (mmoles/ml)

Figure 21. Relationship between ammonia loss rate and the concen-
tration of unionized ammonia in the media in excess of
atmospheric equilibrium.

66

and ANH3_W. The line included in Figure 21 represents what would be
expected if the relationship between flux and ANH3_w was in the form
of Equation 6,and the ammonia mass transfer coefficient based on the
deviation of the water unionized ammonia concentration from atmospheric
equilibrium.(KNH ) was 1 cm/hr. Ekamining the distribution of the
data points with3;:spect to this line reveals that the slope of the
theoretical line fits the data well, indicating that the relationship
between flux and ANH3_w is linear. The fact that the majority of the
data points lie above the plotted line suggests that KNH for the
current study is somewhat greater than 1 cm/hr. 3-V

The relationship between the ammonia concentration difference
across the interface and ammonia loss rate was further analyzed using
linear regression techniques. The results of this analysis on each
of the three trials and on the combined data are presented in Table 2.

Other investigators have considered ammonia volatilization as a

linear function of ANH3_w in systems under constant physical conditions

(Blain, 1969; Folkman and Wachs, 1973; Bouldin 9; 31;, 197k; Vlek and
Stumpe, 1978; Weiler, 1979). Blain's recalculation of the data pre-
sented by Stratton (1968) resulted in KNH values ranging from 1.6
to 3.0 cm/hr (mean 2.1h cm/hr). Rough caiEZlations made from data
collected outdoors by Bouldin gt_§l, (197h) resulted in values of

from 0.65 to 3.15 cm/hr (mean 2.75 cm/hr). Folkman and Wachs (1973)
presented a field derived equation which yields a KNH of about

0.15 cm/hr under the conditions used for my studies. 3;:iler (1979)
presented an analogous equation derived from wind tunnel data which
yields a KNH3_W of about 1.23 cm/hr, close to those found in the cur-

rent study. Review of the literature reveals that only Weiler's data

was collected under well controlled conditions. The higher values

 

 

.. 31m
A; mmzav :22 + a u asac mama: *
.: mm.o ew.o aw.o mw.o a
6 m
sum
OH.H om.a FH.H :m.o Anz\80v mzz
muoa x m:.m muoa x :m.m muoa x :m.m muoa x m:.a Agn\m50\mmaossv a
mm a: as :m manned spam co pmpasz
spam omcanaoo o Hanna m Hanna a awake

 

*.:\mmaoav 35.5
1“:de owhmmmmoaud Sou.“ nova: map :a moflpmspcmocoo mwnosam Unicode: 93 SA, mcofifimfiroc

pmnwmwm Ahn\mao\moaoaav mead." GprdNHHfipmHocr mwcoaad mo mmoflmmmhmmh Hammad“ mo meafimmm .m manna

68

reported by Blain (1969) and Bouldin gt_al, (19TH) may be due to
greater water turbulence, while the extremely low value of Folkman
and Wachs (1973) is most likely due to a decrease in the pH of their
unbuffered test solution.

Despite the good agreement between Weiler's equation and the
current results, the low correlation coefficient of the equation from
the combined data (Table 2) suggests that a simple linear equation
may not adequately describe the relationship between ANH3_w and
ammonia volatilization rate. In an effort to determine the cause of
the variance in the data, the relationship between flux and ANH3_w
was examined at each pH. The result of linear regressions performed
on logarithmic transformations of the data from each pH is presented
in Figure 22. Those pH groups containing less than three data points
with positive flux values were not considered.

From Figure 22 it is evident that flux, and therefore KNH3 ,

V

at any given ANH increased with decreasing pH. The regression

3-w

lines presented have high regression coefficients and display slopes
which are all approximately equal to one. This indicates that at a

given pH, flux is a linear function of ANH w’ but that the coefficient

3..

of mass transfer varies between pH levels. Although the differences

in flux between pH levels at a given ANH may not appear to be great

3-w
on a log-log plot, they may be significant. For instance at 20 C
and a ANH3_w of 0.1 mmoles NH3/l, a change in pH from 11 to 9 would
result in roughly a doubling of ammonia loss.

Examination of Stratton's (1968) data recalculated by Blain
(1969), and Weiler's (1979) data reveals that this trend is present

in their work as well. At 20 C, the KN? values Blain calculated
a 1.3 -w

 

 

 

9 pH 9.5
'3‘ p" '0 H-I0.0
leL0
”18‘!
E c4-4
«3‘
E
Q
a
2
o
E
5
X
2
u.
o ’54 No. of
.9 1H r3 data points
8.0 .887 7
/ 9.0 .992 I?
9.5 .992 IS
I0.0 .977 20
”.0 .959 21
96 i 3 5‘
-6 -5 -4 -3

log ANN“ (mmoles/ml)

Figure 22. Linear regressions of the log of ammonia flux as a function
of the log of the unionized ammonia concentration of the

water in excess of atmospheric equilibrium.

70
at successively lower pH levels increased in h out of 5 cases. Care-
ful examination of weiler's plots of KNH versus wind speed reveals
that at any given wind speed, data colleitzd at pH 9.22 produced higher
mass transfer coefficients than data collected at pH 9.91. The con-

sistency of the relationship between pH and mass transfer coefficient

requires further analysis.

AZNH3_w Mediated Flux

The trends seen in Figure 22 indicate that at lower pH levels
something in the system enhances flux. One difference which exists

between systems with the same ANH but different pH levels is the

3-w

magnitude of ANHh , where ANHh is defined as the difference between

the ionized ammonia concentration of the bulk solution (Equation 10b)
and that in equilibrium with NH3_ew (Equations 7 and 10c). Systems

with lower pH levels have higher ANthw values at a given ANH To

3-W'
investigate the possibility of ANHh-w increasing flux, multiple linear
regressions were performed on each set of experimental data with flux

expressed as a function of ANH _ and ANHh-w' The results of these

3 w
calculations are presented in Table 3.

Comparison of Table 2 with Table 3 reveals that the rate of
ammonia loss is more closely correlated to the combination of ionized
and unionized ammonia concentrations than to the unionized ammonia
concentration alone. Because the physical conditions under which the
three trials were carried out were nearly the same and the constants
a, b, and c in the regression did not vary greatly between trials, it
seems Justifiable to combine the three trials and use the regression

based on the total data to generate an empiric loss rate equation. It

is obvious from theoretical considerations that the net transfer of

3| 3 3|
A ::z<v o +.A mmzav n + m u xdam muons

 

Tl

 

*

®®.o om.o ©>.o mm.o mm

ma.o ma.o ma.o wa.o Aan\aov o

NH.H mH.H mH.H mm.o Aan\sov A

Plea x m~.>l wloa x m>.m mica x :>.>I mica x aw.a Aam\mso\mmaosav m

am a: n as am wasted when mo amnesz
mama

umeabsoo 0 Hanna m amass 4 amass

*.Aa\moaoav asfianHHSUm oaaommmoEpm scam hops:
map a“ mmoHpmhpcmosoo mficoaam cmufimoflns and wouflnofl onp ca mcoHpmH>m© pmnflmmm AH: .
\N8o\moaoaav mopmh cofipmufidwpmao> mwsoaam mo mQmemonwmn amonwa mamfipasa mo mpasmom .m manta

72

ammonia across the interface should be zero when the concentration
of unionized ammonia in the liquid is in equilibrium with the over-
lying air. Because of this and the small size of "a" in the regres-
sion equation, this term can be attributed to experimental error and
omitted from the flux equation. The resulting equation represents
a first approximation to an empiric equation predicting ammonia flux.

As direct measurements of ammonia nitrogen in aquatic systems
usually represent total ammonia, it would be most useful to express
the flux equation in these terms. If it is assumed that the equili—
brium between free ammonia and ammonium ion is always maintained,
the difference in unionized ammonia and ionized ammonia from their
respective atmospheric equilibrium values can be related to the devia-
tion of the total ammonia concentration of the bulk solution from its

atmospheric equilibrium value by the relationship outlined in

Equation 10. If this new mass transfer coefficient is called K

ZNH ’
3-w
then from Table 3:
KZNH = 1.12 (uf) + 0.l8 (l - uf) (ll)
3-w
where: Flux = KENH (AZNH3_W)

Dual Layer Model

The mechanism by which NHh is involved in transport can most
easily be explained by the dual layer model outlined by Liss and
Slater (l97h). In this model there is assumed to be an unstirred
layer of air and an unstirred layer of water which lie on either
side of the gas-liquid interface (Figure 23). At the interface the
two phases are assumed to be in equilibrium. Within the boundary

layers, a concentration gradient exists which ranges from the

T3

.33....
2.23.3...

.223...
2.3:... .

 

.mommaoucw ofiszdlnmm a mo demos pohma Hos: .mm sham“:

2.3:... :oz. .333 ...:m

 

 

2.3.5:... $8.5 Toucaom

 

.23....
2.255

.223...
2.2.5.5. 1V

 

2.3.2.33 ._< Toccaom

2.2.... ...... .... ......

7h

equilibrium concentration at the interface to the concentration of the
well mixed bulk solution at the outer edge. Although gas movement
through the main body of both air and water is by turbulent mixing,
gas movement through the boundary layers is assumed to be limited to
molecular diffusion. Because this diffusion is very slow in compari—
son to the turbulent mixing of the main bodies, these boundary layers
provide the major resistance to gas movement and the rate of transfer
from one phase to the other is controlled here.

The flux through each of the two boundary layers can be defined

by the equation:

diffusivity
flux = , AC (12)
thickness

 

where AC refers to the change in concentration of the gas across the
layer, diffusivity to the rate of movement of the gas through this
layer, and thickness to the depth of the boundary layer (Dankwerts,
1970). Diffusivity is assumed to be constant for a given gas and pro-
portional to the square root of its molecular weight. The thickness
of each of the boundary layers is considered to be controlled by
physical parameters including wind speed and wave action. Therefore,
under constant physical conditions the terms for diffusivity and thick-
next can be combined into a constant(kJ,:reducing Equation 12 to the
form of Equation 6. Liss (1973) points out that the reciprocal of k
can be used as a measure of the resistance of the particular boundary
layer to the movement of the gas being considered, and that the total
resistance (i) to gas exchange across the interface will be the sum

of the resistances of the gas (fig) and liquid (%;) phases.

Under constant conditions, flux can be related to the rate of

transfer through either of the two layers:

75

F = ka (Cia - ca) = kw (Cw - Ciw) (13)

where Ca and Cw are the concentrations of the gas in the bulk air and
water phases, respectively, Cia and Ciw are the concentrations in these
two phases at the interface, and ka and kw are the mass transfer con-
stants of the gas in each of the phases. As it is not possible to
measure the concentrations at the interface, it is necessary to elimi-
nate the terms referring to these concentrations from Equation 13.
Knowing that at the interface the two phases are in equilibrium.and
that:

Ciw = KH Cia (in)
(Equation 7), this is possible. Substituting Equation 1h into Equa-

tion 13 and simplifying results in:

 

 

 

C
'w
kk (——C)
F kw (Cw - KHCa) = w a KH a (15)
k + k KH k
a w .43 .+ k
KH w
which can be written as
C
w
F - Kw(Cw - KHCa) - Ka (KH - Ca) (16)
.__1_-i EH. L-_l_. 1
where. I{ - k + k . and K - k + k (17)
‘w w a a a ‘w

In Equation 16, Kw and K3 are the mass transfer coefficients of
the system described in terms of deviation from equilibrium in the
water and air phases, respectively. They are also the reciprocals
of the total resistance to flux. If total resistance expressed in

terms of the liquid phase (fi%-or Rw) is considered, the resistance
'w

_l_

k
w

and the resistance due to the air boundary layer (E—) as ra where:
a

contributed by the water boundary layer ( ) can be written as rw

R = r + r (18)

In most studies, the resistances of the individual phases are not
determined; but for many of the common atmospheric gases, the
resistance of one of the phases tends to predominate. In these
cases, total resistance and the resistance of the controlling phase
are essentially identical. In a few cases, the resistance of both

phases may be important.

Enhancement

As shown by Equation 12, the transfer coefficient for either
boundary layer is the product of the compound's diffusivity and the
reciprocal of the layer's thickness. Under constant physical condi-
tions this relationship is adequate for unreactive gases such as
N2 and 02,
with other forms, the relationship proves too simplistic. Here the

but for gases which react in water to exist in equilibrium

water phase transfer coefficient not only depends upon the rate of
diffusion of the gas through the boundary layer, but also the rate
of diffusion of the dissolved ions of the gas and the rate at which
the ionic and gaseous molecules interchange. If flux calculations are
based on the difference in gas concentration of the liquid from.atmos¥
pheric equilibrium, the addition of an ionic form to augment the gas
transport would increase flux by effectively reducing the resistance
of the water phase. This increase has been termed enhancement. I
Enhancement of flux due to ionic Species has been most widely
studied in reference to carbon dioxide exchange (Kanwisher, 1963;
Hoover and Berkshire, 1969; Quinn and Otto, 1971; Liss, 1973; Broecker
and Peng, 197%),vfiifllenhancement resulting from the presence of car-

bonate and bicarbonate ions at higher pH levels (pH > 5). Most

77
investigators conclude that for CO2 exchange, enhancement is usually
not significant under natural conditions due to the relatively slow
rates of reaction involved, but for some gases such as $02, 803, and
N02, reactivity is enough to make the gas phase the controlling factor
(Hicks and Liss, 1976).
Ammonia reacts readily with water, creating a pH and temperature

dependent equilibrium.between NH NHHOH, and NHh (Equations 7 and 10).

3,
If the resistance of the water phase is important in controlling
ammonia loss, flux would be expected to be related to the mass
transfer coefficient of both the ionized and unionized forms. That
this is the case has been illustrated in Table 3 and by Equation 11.
Prior to this study, two other investigations considered the possi-
bility of enhancement influencing ammonia flux. Blain (1969) sug-
gested that an equation similar in form to Equation 11 may be necessary
to accurately predict ammonia loss from aquatic systems. Bouwmeester
and Vlek (1980) considered enhancement due to ionized ammonia in their
model and assumed that the diffusivity of the two forms were the same
and that conversion from one form to the other was instantaneous.
Reevaluation of the wind tunnel data presented by Weiler (1977)
would be expected to confirm the enhancement findings of the present
study. Weiler reported values at pH 9.22 and 9.90 under various wind
conditions. If for each pH, Weiler's KNH is expressed as a function
of wind, KXNH3_w can be recalculated frothZese fUnctions at the wind
speed of the current study, 0.66 m/s. Knowing the pH and temperature
allows calculation of "uf" and "l-uf" values for each of the generated
K2NH3.w values. With these data an equation can be generated at each
pH, in which KZNH is written as a function of the coefficients

3aw
of the ionized and unionized forms of ammonia (variables "b" and "c"

78
of Table 3). Solving the two equations simultaneously yields values
of "b" and "c" which range from 0.99 to 1.05 cm/hr and from 0.20 to
0.25 cm/hr, respectively, depending on whether linear or exponential
regressions are used to normalize the effect of wind speed. These
values are reasonably close to the 1.12 cm/hr and 0.18 cm/hr values

derived from.the current data (Table 3 and Equation ll).

Magnitude g£_Enhancement

Despite the apparent close agreement between the data sets, it
is difficult to assess the true magnitude of the coefficients "b" and
"c" which determine KZNH3_W' weiler presented very few data points,
and considerable variation exists in the data collected at any given
pH in the current study (Figure 2h). It is also evident from Table A
that the values obtained for coefficients "b" and "c" are dependent
upon the pH of the data used to derive them. When any of the lowest
pH data (pH 8, 7, 6) are omitted, the significance of NHh in the
transport process appears to be greater than when all the data are
considered. This is particularly emphasized when the lowest pH data
considered is pH 9.0. Figure 25 illustrates how the various regres-
sions of Table h fit the experimental data. This figure clearly
shows that at intermediate pH levels a simple linear regression
(regression number 9, Table h) underestimates KZNH and hence flux.
The two multiple linear regressions plotted represgn: the extremes of
Table h and provide for the higher KXNH3.w values found in the inter-
mediate pH ranges. Of course regression 3 (Table h) provides the best
fit for the upper four pH groups.

All of the multiple linear regressions greatly overestimate

KZNH at the lower pH levels. This indicates that either the
3-w

79

 

 

 

 

 

. T
Mm n 4} mean 3 lstandard ‘deviation for |
.L all (a) data points I
a mean! I standard deviation for I
'3' (a) data points when - '-
o<KzNH3-.<2 I IF
I
|.O-7 T | "i
: “+9“ 'lzo
‘ |
r
.. .8- I T l I
E I 15PM | |L
\ L ..
5 — I l r '
V .a- I ' J. I
s .7 .L
h I
z:
E 4. ‘
x - ta
.2 .. I
3117 |
2 z .I.
00" 2?. i
l n
‘2‘ l
6 ‘7 a 6 15 IT
pH

Figure 2’4. Mean mass ransfer coefficients at each pH level
tested in laboratory experiments.

80

3
.AN mHQmBV

momspmn mosao>

um

3um

$24 no :oflvocdm a mo add“ mo aoflmmoawoa adoofia mamaam a upcomoaaon m aofimmmawor ax

.AJN ohdwwhv h£\So o.m cum cams

.m manma ca coowmoo mm and 0 mod :9: Mom mmsam> .

M oopmHSUHmo a“ copadmoh Scam: moagaom mmomp haao so sun mm: m cmemoamom *

 

 

mm.o --- oa.a o.m .o.» .o.m .o.a .m.m .o.oa .o.HH m**
mm.o mm.o oa.H o.o .o.» .o.m .o.m .m.m .o.oa .o.HH . w*
Hm.o ms.o mm.o o.m .m.m .o.oa a
no.0 wa.o mo.H m.m .o.oa m
mm.o ma.o NH.H o.m .o.» .o.w .o.m .m.m .o.oa .o.HH m
mm.o ma.o ma.a o.m .o.m .m.m .o.oa .o.HH s
Hm.o os.o Ho.a o.m .m.m .o.oa .o.HH m
om.o om.o eo.a m.m .o.oa .o.HH m
No.0 Hm.o eo.a o.oa .o.HH H
mm +0 +9 mm nomwmwmwom

 

mmmoxo aw awoanm omufiao« com couwaowas mo mofipoasm a ma Mada ho maofimmoummh hmmawa oamwpasz

.mddOhw mm oopooamm you ssfihnfiawsdo cahmmmmoapd.mo

.smdaa

81

Table 4 '
l.4q Regression Na.

. 5

7-——

 

[‘2‘ 9......

 

 

04"

K [NHs-U (cm/hr)

 

 

0-0000000000000...

 

 

0'1
5

é

~8-

6
pH

Figure 25. A comparison of various linear regression equations
with experimental mean mas transfer coefficients a

different pH levels . '

81

Table4
L4. Regression No.

 

. 5

7---

|.2-1 9......

LO.

 

 

.6-

04"

K £NH3-w (cm/hr)

02“

 

 

0.0.0.0.....,..o'.

-02‘

 

 

04
8

a
pH

mil
N.

Figure 25. A comparison of various linear regression equations
with experimental mean mass transfer coefficients a

different pH levels . .

82
measurements made at these pH levels were in error or that the
proposed multiple linear relationships do not apply in this range.
Several flaws in the data exist which would be especially apparent
over this pH range. Few data points were collected at low pH levels
and those which were obtained represent small concentration changes
occurring over long periods of time. Any error due to incorrect
measurement of the concentration of ammonia in the air is magnified
over the low pH ranges as well because the driving force behind the
measured flux is the deviation of the ammonia concentration in solu-
tion from atmospheric equilibrium. Atta given ammonia concentration
in the water, a small error in the measured atmospheric concentration
is unimportant in determining AZNH3_wat high pH levels. But at lower
pH levels, a similar error in the atmospheric ammonia measurement
results in large changes in equilibrium total ammonia (Table 5) and
hence large changes in AZNH3_W.

The possibility of the multiple linear regressions not holding
over the entire pH range also is real. As noted previously, it is
possible to think of the mass transfer coefficient as the reciprocal
of the total resistance. This resistance is the sum of the resistance
in the gas phase and the resistance in the liquid phase. Enhancement,‘
as discussed here, would only effect transport through the liquid
phase. Therefore, the importance of enhancement would be expected to
vary along with the magnitude of liquid phase resistance. In order
to further investigate the role of enhancement it is necessary to look

at how the resistance is partitioned between the two phases.

83

3:0

 

 

 

.0 000 P 00009000m 00 00 0000900 000 MS 000 mm 000:3 omwv 00009009000000 00<v 00 u A :zwv *

P...00 x 00.0 0100 on P410 0:00 on 00.0 0.100 x 00.0 0.100 x m0.m ~.IO0 on him blo0 .0 No.0 0.00
0:00 a 00.0 0:00 x 00.0 0:00 0 00.0 0:00 0 00.0 0:00 0 00.0 0:00 0 00.0 0:00 a 00.0 0.00
0:I00 x 00.0 0:00 x 00.0 0:00 x 00.0 0:00 x 00.0 N....00 x 0:.m 0100 x 00.0 0:00 x 00.0 0.00
0:00 0 00.0 0:00 x 00.0 0:00 x 00.0 0:00 a 00.0 0:00 a 00.0 0:00 0 00.0 0:00 a 00.0 0.00
0:00 a 00.0 0:00 x 00.0 0:00 0 00.0 0:00 0 00.0 0:00 x 00.0 .0:00 0 00.0 0:00 x 00.0 0.0
0:00 0 00.0 0:00 0 00.0 0:00 a 00.0 0:00 a 00.0 0:00 0 00.0 0:00 x 00.0 0:00 0 00.0 0.0
0100 x 00.0 0100 x 00.0 0:00 x 00.0 0:00 x 00.0 0:00 x 00.0 0100 x 00.0 0:00 x 00.0 0.0
0100 N 00.0 0:00 x 00.0 0:00 x 00.0 0100 x 00.0 0:00 x 00.0 0100 x 00.0 0100 x 00.0 0.0
0:00 a 00.0 0:00 x 00.0 0:00 x 00.0 0:00 a 00.0 0:00 0 00.0 0:00 x 00.0 0:00 0 00.0 0.0
0:00 0 00.0 0:00 x 00.0 0:00 x 00.0 0:00 a 00.0 0:00 0 00.0 0:00 0 00.0 0:00 x 00.0 0.0
0:00 a 00.0 0:00 0 00.0 0:00 a 00.0 0:00 a 00.0 0:00 a 00.0 0:00 a 00.0 0:00 0 00.0 0.0
.100 x 00.0 00I00 x 0©.N 5.0:.00 .0 +5.: 01.00 N 00.0 Mlo0 on 00.0 Mlo0 N 00.0 mto0 N 00.0 0.0
0 00 0 00 0 00 0 00 0 00 0 0 0 0 :0

mégmaa
.00000000 00909 0003000590 000 9000: «\00008 0:00 x 00.0 mo 0009009000000
00002000890 00 0903 00099000000 90 90903 00 ana\000oa_3:mmew 0009009000000 0000080 00909 .0 00009

8h

Partitioning 2£_Resistance

Because the laboratory conditions used in this study were con-
trolled such that wind speed, temperature, and water turbulence
remained constant for all the samples, it can be assumed that the
thickness of the respective boundary layers was the same in every
case. Therefore, the differences in mass transfer coefficients which
appear between samples are the result of differences in the rate of
diffusion and not boundary layer thickness. As there were no dif-
ferences in the characteristics of the air boundary layer or the form
of ammonia traversing it, it is reasonable to assume that the gas
phase resistance is constant with respect to unionized ammonia for
all samples. In order to estimate this portion of the total resistance,
the data of Liss (1973) can be used.

Liss found that for a system receiving no direct wind agitation,
water vapor and oxygen displayed transfer coefficients of KHQO z

-&

lOOO cm/hr and K 2 h cm/hr, respectively, where it is assumed that

02-w

the resistance to the transport of H20 occurs entirely in the air

phase and the resistance to the transport of 02 occurs in the water

phase (i.e., KO = k0 ; KH 0 = RH O ). According to LlSS
2-w 2-w 2 -a 2 -a

(1973) the transfer coefficient for other compounds should be related
to those he measured by the square root of the ratio of their molecular

weights. For unionized ammonia this means:

 

k = K (VM.w. H2O/M.W. NH3)‘ . 1029 cm/hr (19)

 

kNH : KO [(VMJJ. O2/M.W. NH3)‘ (.80) +

3-w 2-w (20)

 

(vM.w. 02mm. NHhOH)‘ (.20)]~ 5.16 cm/hr

85
where M.W. refers to the molecular weight of the respective compound.
The two terms required to express the liquid phase transfer coefficient

(Equation 20) are due to the equilibrium which exists between NH and

3
NHhOH (Jolly, l96h). Failure to consider this equilibrium is not a
drastic error as the estimated value of kNH changes by less than
O.h0 cm/hr due to its inclusion. 3-W

When the results of Equations 19 and 20 are substituted into

Equation 17, the outcome is as follows:

1 K11

sz‘zgj‘
where KH is Henry's law constant calculated at 20 C as described for
Equation 7. The value obtained from Equation 21 closely approximates
that arrived at for high pH values using the multiple linear regression
of all the data collected in this study, 0.893 hr/cm (Equation 11,
Table h). If this total resistance is broken down in the same propor-
tions as that of Equation 21, 0.l9h hr/cm.(22.l%) can be attributed to
water phase resistance and 0.68% hr/cm (77.9%) to air phase resistance.
Of course, the use of this comparison is not strictly valid as the con—
ditions of the two studies were not identical and the total resistance
estimate of 0.893 hr/cm is probably somewhat low for the current data.'
But, close agreement between the estimates and the current data suggest
that the technique proposed by Liss (1973) and the partitioning of the
total resistance into water and air phase resistances as done above are
reasonable.

If total resistance and air phase resistance (estimated as
described above) are examined across the range of pH values used in

this study, it is seen that air phase resistance totally dominates the

system at pH levels lower than 9.5, while both phases are important at

86

high pH levels (Figure 26). It is unreasonable for the air phase
resistance to exceed the total resistance as it apparently does below
pH 9. This can be explained partially by recalling that Equations

3, 5, and 7 of Table h overestimate the mass transfer coefficients

at low pH levels when compared to the experimental data (Figure 25).

If it is assumed that at low pH levels all resistance is due to the air
phase, the mass transfer coefficients for these points can be calcu-
lated by finding the reciprocal of the high pH air phase resistance
(0.68h hr/cm) and multiplying it by the fraction of the total ammonia
which is unionized at that pH. This results in values of 0.0006 cm/
hr, 0.0058 cm/hr, and 0.0560 cm/hr for pH's 6, 7, and 8, respectively.
These values are somewhat below those indicated by the means of the
collected data at the respective pH levels, suggesting the estimate of
air phase resistance may be slightly high. Recalculating air phase
resistance to agree with the data obtained at pH 7 and 8 yields a value
of 0.326 hr/cm. Even with this drastic reduction in the air phase
resistance estimate, Figure 26 illustrates that it still dominates the
system at low pH levels. The precise pH at which water phase resistance
becomes insignificant depends upon the equation used to predict total
.resistance. Because those equations plotted in Figure 26 predict total
resistances which are too small at low pH levels when compared to the
laboratory results, water phase resistance is probably significant at
pH levels well below those indicated.

The shift from a volatilization system controlled by the rate of
transfer through both boundary layers at high pH levels to one completely
dominated by air phase resistance at lower pH levels explains the
lack of fit of the multiple linear regression equations of Table h. The

shift in control results from the reduction in unionized ammonia at

 

 

87

 

 

 

7‘ Total Resistance as
calculated from Table 4
Regression Na.
3 _ o _ e —-
5 — — — .—
6« 7 '
—— \ \Esflmated air
A phase resistance
\ based on pH
3 5- 7 and 8 data
t \
.15 Estimated air
I.) \ 6-— phase resistance
4 from Liss 0973)
2 4‘
O
0
c
O
L-
.2’. 3-
O
b
.5
0
2
c: 2‘
’5
'3
0
a:
I-l
O I Y I8 T I
6 8 9 IO ll
pH

Figure 26. Resistance to movement of ammonia across the air-water
interface as a function of pH for the current labora-
tory conditions. ‘

88
the lower pH levels. This reduction greatly impedes transfer through
the air boundary layer as only the unionized form exists here. The
water layer diffusion is affected to a lesser extent as NHh transport

compensates for the reduction in NH3.

Predicting ETNH
3-w

It is evident that the multiple linear regression equations pre—
sented in Table h are not accurate predictors of the total mass trans-
fer coefficient. However, by using the data collected and the estimates
made in the previous section to calculate total resistance, it should

be possible to create an expression predicting K It is reason-

ZNH3dw'
able to assume that the ANH3_w based air phase resistance is somewhere

between that calculated using the approach of Liss (0.68h hrs/cm) and
that calculated from the current data (0.326 hrs/cm). For the following
calculations these two values were averaged, resulting in an estimate
of 0.505 hrs/cm. To correct the air phase resistance value for changes

in pH, the ANH based air phase resistance is divided by the unionized

3-w

fraction of total ammonia at that pH. Air phase resistance (ra) for

the current laboratory conditions can then be expressed as:

'o.5g§ hrjcm _ ‘
uf - ra(lab) (22)

Knowing ra and the total resistance ( ), it becomes possible

KzN

H
-w
to calculate water phase resistance (rw) from Ehe laboratory data. If

regression 3 from Table h is used to calculate K the expression

ZNH3_W’
of water phase resistance (rw) becomes:
r = --;--- r
W(lab) K a(lab)
3-w

OI‘

89

 

 

 

= l _ 0.505 hr/cm
rw(lab) (l.01(uf) + 0.h0(1 - ur))cm/hr uf
OI‘
— 1 M2
rw(1ab) ' (0.61 uf + 0.h0 ‘ uf ) hr/Cm (23)

with the restriction that rw Z 0. Regression 3 was chosen for this
derivation because it best fits the data collected above pH 9.0. For
the data below pH 9.0 the resistance is dominated by the air phase and
therefore the errors in water phase resistance caused by use of regres-
sion 3 are small. After combining ra and rw to find the total resis-

tance (Equation l8),this value can be inverted to find K

2NH3_w

l
K = (2h)
ZNH ( l O. 05) + (O. 05)

3'“ 0.61 uf + 0.h0 ‘ uf uf
where the first half of the demonimator is restricted to being non-

negative.

Effects 31; m

Equations 22, 23, and 2h apply only under the conditions used in
this study. To make these equations more meaningful, it is necessary
to include terms to compensate for any environmental changes which
would alter resistance. The factor most likely to affect resistance
is wind speed, acting to agitate both the air and water phases and
reduce the thickness of the respective boundary layers. The relation-
ship between wind speed and mass transfer coefficient has been investi-
gated for several gases (Kanwisher, 1963; Hoover and Berkshire, 1969;
Liss, 1973; Weiler, 1979). These investigators appear to agree that

transfer through the air boundary layer is a linear function of wind

90
speed while transfer through the water boundary layer is an exponential
function.

Looking at the water boundary layer first, by definition:

1; 2.5.71 (25)

where kw is the water phase mass transfer coefficient, Dg-w is the
diffusivity of the gas in water, and 2V is the thickness of the
water phase boundary layer. When it appears that transfer through the
liquid phase depends upon more than one form of the compound, Ds-V
(DZNH

3-w

forms through the boundary layer and the subsequent rate of conversion

in this study) represents the total rate of movement of all

to the gaseous form.
Expressing the resistance of this phase as the reciprocal of its

mass transfer coefficient, Equation 25 becomes:

r = w (26)

 

Hoover and Berkshire (1969) present wind tunnel data which can be used
to show that resistance to C02 transport (assumed to be entirely water

phase) varies according to the expression:
r = 1.66 e'0-38U
w

where U is wind speed in m/s (r2 = 0.9h). Similar analysis of the
data presented by Liss (1973) for oxygen transportyieldsaniidentical
relationship with a coefficient of determination (r2) of 0.97 when
wind speeds (up to 8.2 m/s) were determined 10 cm above the water
surface. Equation 26 can be rearranged and combined with the above

relationship such that zw can be expressed as a function of wind as

follows:

91

zw = (1.66 e'0°38U)(DO ) (27)

2-w
From the current study, the diffusion rate of all forms of
ammonia through the boundary layer.can be expressed by rearranging
Equation 26.
DZNH = SEEP—L (28)
3-w w(lab)

where r is calculated from Equation 23 and z is the unknown

w(lab) w(lab)

water boundary layer thickness. Combining Equations 26, 27, and 28,

a general expression for rZNH can be formed.
3-w
z
r = w
ZNH D
3-w ZNH3_w
or
D0
rZNH = 1.66 e-'O°38U rw(lab) 2.223;) (29)
3-w w(lab)
DO2-w
where :——-—-—-is a constant. The value of this constant can be calcu-
“w(lab)
lated as it 18 known that rZNHB-w is equal to rw(lab) when U = 0.66 m/s
from the laboratory experiments. Equation 29 is then reduced to:
rZNH 1.29 e rw(lab) (30)

3-w
The variation in air phase resistance with wind speed can be
investigated using the data presented by Weiler (1977) if it is assumed
that wind speeds measured at approximately 9 cm (Weiler) and 10 cm
(Liss) are not significantly different. When the values weiler pre-
sented are converted to mass transfer coefficients based on AZNH3_w and
the appropriate total resistances are calculated, the water phase resis-

tance (Equation 30) can be subtracted from the total resistance to find

92

the air phase resistance (r ). Because Weiler's data were col-

ZNH3_a

lected at two pH levels, it is necessary to eliminate the effect of pH
by multiplying the air phase resistance based on the total ammonia

gradient (rZNH ) by uf to convert it to a resistance based on the
3-a

unionized gradient alone These r values can then be

(r ).
NH3_& NH3_a
plotted as a function of wind speed to determine the relationship

between the two. Figure 27 illustrates this relationship and the line
generated by the equation:

rNH = 0.583 - 0.0h50 _' (31)
3-a

This equation was derived from a linear regression of r against

NH3_a

‘U(r2== 0.90), omitting the low temperature data. To find a general
expression for air phase resistance in terms of the totad.ammonia.gradient,
Equation 31 must be divided by uf as explained for Figure 26. This
results in the expression:

= 0.583 - 0.0h50

ZNH3_a uf

(32)

when U is wind speed in meters/sec at a height of 10 cm above the
surface of the water. Because Weiler's data only covered wind speeds
up to 6.3 m/s, Equation 32 is strictly valid only over that range.
Even if this restriction is ignored, U must remain below 12.96 m/s

in order for air phase resistance to remain positive.

This equation predicts an air phase resistance of about 0.553 hr/
cm under the high pH conditions of my laboratory experiment. This is
close to the mean value used for the previous water phase resistance
calculations, but really only reflects the similarity in total resis-
tances of the two experiments and does not verify the correctness of

the estimate. It should be apparent from the way they were derived,

 

93

 

rm... - sea-.045 u for 20-2: C°Data (o)
r2=.90
.7-
.6-
\ A
\
k

5..

e \ . /.
g A\ etc [I
a. I
5 '4‘ \ |?\C a>'70c

\

'7 \ A
£5 A \A
3 '3‘ >

.2-

dd

0 l I ~ I f I 1—

O l 2 3. 4 5' 6
lJ rn/s

Figure 27. Calculated air phase resistance of data from weiler

(1977) as a function of wind speed. All data was
adJusted to pH 12 to eliminate pH as a cause of variance
and temperature was between 20 C and 21 C unless other-
wise noted.

9h
that in Equations 23, 30, and 32 the estimated air phase resistance
for the current laboratory data directly affects the magnitude of each
of the two phases of resistance. Thus, the accuracy of Equations 30

and 32 should be viewed appropriately.

Importance 9£_Enhancement

The magnitude of enhancement would be expected to be influenced
by both wind speed and pH, as wind speed controls the thickness of the
air and water boundary layers and pH determines the relative amount of
ionized ammonia present. Figure 28 illustrates the effect that
enhancement has on calculated mass transfer coefficients over a range
of pH values when U = 0.66 m/s and air phase resistance is assumed to
be 0.505/uf. The curve depicting mass transfer coefficients with
enhancement was calculated using Equation 2h. The other curve was
calculated assuming that all transport was as NH which had a mass

3

transfer coefficient of 1.01 cm/hr (i.e., K = 1.01 uf). From

ZNH3_w

Figure 28 it is evident that enhancement is essentially zero below

pH 7 as air phase resistance is so large that a reduction in water

phase resistance due to enhancement is not important. Above pH 11,

essentially all ammonia exists as NH3 so no enhancement occurs.

Midway between these two extremes enhancement is of maximum importance.
By selecting a pH which displays enhancement, it is possible to

observe how enhancement varies with increasing wind speeds. This has

been done for pH 9.5 and the results are displayed in Figure 29. The

air phase resistance (lowest line) was calculated using Equation 32.

Total resistance with enhancement was calculated by adding the air

phase resistance to the water phase resistance calculated from Equation

.94

e8“

07‘

06“

(cm/hr)

05"

KzNH3.w

 

 

95

U a .66 m/s
1'3 20’C

- . . . with enhancement
without enhancement

 

 

9'

an-l

pH

Figure 28. Variations in mass transfer coefficient with and without

enhancement as a function of media pH.

96

re ____

Total Resistance

l.6 J with enhancement - . - . . o
e 'without enhancement
' pH= 9.5
' T = 20° C
L4-

(hr/cm)
2?

Resistance

 

.2-i

 

O

 

‘ I [ Ii

0 2 4 6 8 lo
U (In/s)

Figure 29. Resistance as related to wind speed at 10 cm above
the water surface.

97

30. Total resistance without enhancement was calculated in a similar

. . l
manner except that rw(lab) in Equation 30 was replaced by (1.01 uf -
O. O )

uf '

In Figure 29 the vertical distance between the curve depicting
ra and the two total resistance curves represents the resistance
attributed to the water phase. The difference between the two total
resistance curves is due to the drop in resistance via enhancement.

It is apparent that based upon the available data, increases in wind
speed reduce the importance of water phase resistance and with it the
importance of enhancement. While enhancement increases the calculated
mass transfer coefficient by 35% at zero wind speed, the calculated
increase is only about 10% at 6 m/s and 5% at wind speeds of 9 and 10
m/s. Therefore, it is doubtful that enhancement makes a significant
contribution to ammonia transfer at wind speeds exceeding 10 m/s at

10 cm above the water surface.

Although Figure 29 accurately reflects the data available con-
cerning how the two components of total resistance vary with wind speed,
the results are surprising and seem physically unrealistic. If move-
ment in the bulk air phase is providing the energy required to reduce
boundary layer thicknesses, it seems unlikely that the prOportion of
total resistance attributable to the air boundary layer would increase
with wind speed as it does in Figure 29. This result suggests that
additional study in the area may be required. If it is found that the
contribution of the water phase has been underestimated, the importance
of enhancement would have also been underestimated as at this pH
enhancement decreases water phase resistance by about 50% at any wind

speed.

98

Effects 9£_Temperature 9§_Diffusivity

It is apparent from the solid squares in Figure 27 that tempera-
ture affects more than Just the equilibrium between the ionized and
unionized forms of ammonia. Broecker and Peng (197k) cite an empiric
equation derived by Himmelblau (l96h) which relates diffusivity of
gases through water to temperature:

log D = ~1010/(T + 273) + constant

What is needed here is a temperature function which yields a coeffi-
A cient to correct for temperature deviations from 20 C. Himmelblau's
equation can be rearranged to:

D = constant (lo-3'0“)“T + 273))

where the constant represents some multiple of the diffusivity at 20 C.

This equation can therefore be expressed as:

_ ~1010/(T + 273)
D = (D at 20 C)(constant)(lo )

-lOlO/T + (273)

where (constant)(10 ) must equal one at 20 C or:

-lOlO/(273 + T)) (33)

D = (D at 20 C)(2800)(10

Although this equation has been derived for application to diffu-
sion through the water phase, it is reasonable to assume that diffu-
sivity in the gas phase would be affected to at least as great a degree.
Based on this assumption, the low temperature data presented by Weiler
(1977) were adjusted to 20 C and are represented in Figure 27 as
squares. From this meager data Equation 33 appears adequate as a tem-
perature corrector, although it does not consider differences in boun-

dary thickness resulting from changes in the viscosity of the two

phases.

99
A_Predictive Equation for Ammonia Loss
Using the relationships outlined in the previous sections, it is
possible to derive an equationtx>predict ammonia losses from natural
waters. Recalling that total resistance is the sum of the resistances
in the water and air phases, respectively, Equations 30 and 32 can be

combined to yield:

-0.38U r + (0.583 - 0.0h50)

 

 

RZNH3-w = 1.29 e w(lab) uf hr/cm (3h)
- = 1 - 9:292.
where' rw(lab) (0.61 uf + 0.h0 uf' ) hr/cm a O
uf g 1/(1 + lo0.09018 + 2729.92/(T + 273) - pH)
T = Centigrade temperature

C.‘
II

wind speed in meters/second measured 10 cm above the
water surface .

The effects of temperature on diffusivity can be included by dividing
Equation 3h by the temperature correction term (Equation 33).

1010/(273 + T)

-O.38U + (0.583 - 0.0h5U)) (10 2800

rw(lab) uf ) (35)

R = (1.29 e
3-w

The result (Equation 35) represents the best estimate of total resis-

tance available from the current data. Flux can then be described by:

F = K (AZNH3_W)

where K = -*---.
ZNH3-w RZNH

Tables 6,'7,and 8 illustrate how the mass transfer coefficient
(KENH ) varies under differing physical and chemical conditions.
3-w
From Tables 6 and 7 it is evident that temperature strongly influences

K and hence flux through its effects on ionization and diffusion.
3-w

lOO

 

 

Table 6. Variations in mass transfer coefficient (KZNH ) with
temperature and pH at a wind speed of 3-w
0.66 meters/second.

pH 0 c 5 c 10 c 15 c 20 c 25 c 30 c
6.0 0.0001 .0001 .0002 0.000h 0.0007 0.0012 0.0019
6.5 0.0003 .0005 .0008 0.0013 0.0022 0.0037 0.0059
7.0 0.0008 .001u .0025 0.00h2 0.0071 0.0115 0.0185
7.5 0.0026 .00u6 .0079 0.0133 0.0221 0.0360 0.0576
8.0 0.0082 .01h3 .02h6 0.0h1u 0.0681 0.1097 0.1729
8.5 0.0253 .ouh1 .07h9 0.1238 0.1991 0.3112 0.h717
.0 0.0760 .1290 .2116 0.33u6 0.5085 0.6529 0.8109
9.5 0.2067 .3299 .h35u 0.5532 0.6921 0.8508 1.0273
10.0 0.3508 .hsoh .5661 0.6963 0.8395 0.99h9 1.162h
10.5 0.uh36 .5h66 .6598 0.7830 0.9167 1.0619 1.2196
11.0 0.5009 .5972 .7029 0.8190 0.9u65 1.086h 1.2397
11.5 0.5250 .6167 .7186 0.8316 0.9566 1.09h6 1.2h63

 

lOl

 

 

Table 7. Variations in mass transfer coefficient (KZNH ) with
temperature and wind speed at pH 9.0. 3-w

U m/s 0 c 5 c 10 C 15 c 20 c 25 C 30 c
0.0 0.0721 0.122h 0.2009 0.3175 0.h826 0.6037 0.73h1
1.0 0.0782 0.1327 0.2177 0.3hh1 0.5230 0.6793 0.8520
2.0 0.0853 0.1uu8 0.2375 0.3755 0.5707 0.7626 0.980u
3.0 0.0938 0.159u 0.261h 0.h132 0.6280 0.8571 1.123h
.0 0.10h3 0.1772 0.2906 0.h593 0.6982 0.9679 1.2876
5.0 0.117h 0.199h 0.3271 0.5171 0.7860 1.1023 1.h830
6.0 0.13u3 0.2281 0.37h1 0.591h 0.8989 1.2716 1.7253
7.0 0.1569 0.266h 0.h370 0.6907 1.0500 1.h9hu 2.0h03
.0 0.1885 0.3201 0.5251 0.8301 1.2617 1.8039 2.h7h0
9.0 0.2362 0.u011 0.6579 1.0h00 1.5807 2.2666 3.1178
10.0 0.3161 0.5368 0.8805 1.3918 2.1156 3.0380 h.185h
11.0 0.h778 0.8113 1.3307 2.1035 3.197h n.5891 6.3187

 

l02

 

 

Table 8. Variations in mass transfer coefficient (KZNH ) with pH
and wind at 20 C. 3-w
pH 0 l 2 1+ 6 8 10
(m/s) (m/s) (m/s) (In/S) (m/s) (m/s) (m/s)
6.0 0.0007 0.0007 0.0008 0.0010 0.0012 '0.0018 0.0029
6.5 0.0021 0.0023 0.0025 0.0031 0.00h0 0.0056 0.0093
7.0 0.0067 0.0072 0.0079 0.0097 0.0125 0.0175 0.0293
7.5 0.0210 0.0227 0.02u8 0.030u 0.0391 0.05u8 0.0920
8.0 0.06h6 0.0700 0.076h 0.0935 0.120h 0.1690 0.2833
8.5 0.1890 0.20h8 0.2235 0.273h 0.3520 0.h9h0 0.828h
.0 0.h826 0.5230 0.5707 0.6982 0.8989 1.2617 2.1156
.5 0.6152 0.7335 0.8632 1.1717 1.6020 2.3233 3.9u96
10.0 0.7299 0.899% 1.0901 1.5h93 2.1825 3.2211 5.5178
10.5 0.7915 0.9856 1.2063 1.7h20 2.h812 3.6862 6.3333
11.0 0.815h 1.0188 1.2508 1.8158 2.5958 3.8650 6.6h72
11.5 0.8235 1.0301 1.2659 1.8h08 2.63h5 3.9255 6.753u

 

103
The temperature effect should be treated with caution due to the
limited data available concerning the changes occurring in diffusivity.
Variations in mass transfer coefficient with pH (Table 6 and 8) result

from its effect on the equilibrium between NH and NHh' Increases in

3

wind speed increase K by reducing the boundary layer thickness

2NH3_w

of both the air and water phase (Tables 7 and 8).

Comparison of the K values generated using Equation 35 with

2NH3_w

the data collected in the experimental trials (Table 9) reveals that
the current equation tends to underestimate flux, particularly at the
lowest pH levels. This difference is probably due to the compromise
made in estimating the air phase resistance. The percent difference

between Weiler's buffered laboratory data and values predicted for

predicted - measured
measured

ranged from 1.5% to 16.h% with a mean difference of 6.0% (Table 9). No

those conditions from the current equations ( x 100)
trend appeared in the direction of the differences and the actual
average difference between the measured and predicted values was only
0.063 cm/hr. However, these data do not represent a true test of the
equation as the equation is not independent of the data. In addition
to his wind tunnel studies, Weiler (1977) collected field data from
small containers on a barge and from a floating mini-corral. Table

9 reveals that the barge data varies from that predicted by Equation 35
by a larger extent than the wind tunnel data, but the differences are,
not consistent in direction or magnitude. His mini-corral data (Weiler,
1979) displays a mass transfer coefficient much higher than predicted
by either his equation or the current one (Table 9). For these trials
Wéiler's equation underestimates the mass transfer coefficient by a
factor of 30 and the current equation underestimates it by a factor

of 22, the difference being due to consideration of enhancement. The

10h

A. . . cosoflpcoov

 

 

 

 

 

 

moo.o mm.m moemm><

mma.0 mm.ma N>:0.0 0:H®.0 0.>H :.m 0m.0 m
wm0.0 mm.m mmmP.0 0000.0 :.NH m.m mm.0 m
ma0.0 m:.m 0::m.0 0Hmm.0 m.® m.m mm.0 m
030.0 mm.m 00HH.N 00>0.m 0.00 0.0 mm.m m
mPH.0 No.0 0Nw>.H 00:0.a H.0N 0.m 00.0 m
:H0.0 0:.H 0000.0 00:0.0 0.Hm m.a 00.0 m
HOH.0 Ha.» 00am.a 00N:.H b.0N m.0 mm.0 m
0m0.0 :m.m 0HHH.H 00:H.H :.0N m.m mw.m m
>m0.0 0:.m 0~0H.H 00>0.H 0.00 H.m mm.m m
0:0.0 Hm.: 0H®>.0 0mmw.0 0.Hm :.m mm.m m
m00.0 No.0H 0PP0.0 0ma0.0 0.Hm m.H mm.0 m
Hm0.0 mm.>m m0¢mm><

0H0.0 0H.m m0:0.0 H>m0.0 0.0m 00.0 0.HH m
p0a.0 Hm.HH :mmw.0 mm:m.0 0.0m 00.0 0.0a <
m:0.0 Ha.m Hm00.0 H»Mh.0 0.0m 00.0 m.m <
ww0.0 mm.: mwom.0 m:mm.0 0.00 00.0 0.0 <
000.0 0:.0: H000.0 o:ma.0 0.0m 00.0 0.0 <
mm0.0 00.00 m:00.0 mbm0.0 0.0m 00.0 0.» ¢
000.0 00.00 ~000.0 m0©0.0 0.0m 00.0 0.0 <

sum 31m
Ah£\Eov monouommwm Hazpod monoquMwm R Ahn\aov mZNM Ahn\aov mzwz on B Am\av 0 mm condom
summzw mm eonoosom none none
mozaa> M mo domflhamsoo
aim .apac oHpmHfiw>n scum copaaaoaoo .

:2

anon» spa: 0m :ofluasvm sow nounHSOHmo A wxv mpoowowmmooo Mohammap mmsa mo moonwaamsoo .0 manna

105

A. . . cossflpooov .
.coomm new: comp sonpna mpwooao> sous: awonpm haa¢dpo< *

 

 

 

 

wm0.0 m0.mm m0<mm>¢

HH0.0 00.0 000:.0 200:.0 0.0m *0H.0 0.0 m
mm0.0 00.HH 000m.0 ma»m.0 0.0m *0H.0 0.0 0
000.0 H0.0HH 0omH.0 0300.0 0.0a *0H.0 0.0 0
400.0 00.0m mmm0.0 m000.0 0.00 *0H.0 0.0 m
0w0.0 0H.0H 000:.0 0~H:.0 0.0m *0a.0 0.0 m
hm0.0 00.0H m:0a.0 0amm.0 0.0m s0a.0 0.0 m
Hmo.o om.am mmoo.o omem.o 0.0m ema.o 0.0 m
ma0.0 :H.H: mamo.0 0000.0 0.00 *0H.0 m.b m
0mm.m w0.:0 m0<mm>¢

H0>.m 00.00 Noma.0 0000.0 H.0H m.H m:.0 0
mm0.: mm.00 0>0H.0 0000.: m.0H 0.0 00.0 0
000.m ma.m0 :H0m.0 0000.0 0.mm H.H 00.0 0
0N0.0H 0>.m0 mph:.0 000H.aH 0.:0 >.0 mh.0 0
00:.0 00.00 mu¢mm>¢

Nom.0 00.mma H000.0 m0mm.0 :.bH 0.0 m.0H 0
0ma.0 00.>N 000m.0 mpmz.0 H.0 0.0 0N.0 0
NOH.0 00.0w 000:.0 00am.0 0.0a v.0 mm.0 0
000.0 0».0N m>:0.0 00mH.H 0.:H 0.H H.0H 0
000.0 :0.>: 00:0.H 00H0.N 0.:0 0.0 H.0H 0
mm0.0 00.00H 0>00.H ozvm.0 H.0m v.0 0.0a 0

sum sum .
Ah:\aov monoyoMMHa adapo< monouommwm R Ap£\aov mZNM Ah£\aov 0202 A00 9 Am\av 0 mg meadow
Sum 00 :Oapasvm mama ogmn

mozad> 020% mo QOmfiaamaoo

 

.A.©.QOUV m mdfldB

106

 

 

 

.A000HV xma> 0cm popmmmazsom mo :oHumSUm m:# an umumamcmm mpmn u &

.A000Hv aOppmupm scum mpmc hpOpmhonmq u m

.Abw0av amawmz song mpmc Hmnaoolaafiz n 0

.ANPOHV hmawmz song mama mmpmm n 0

.APPOHV pmaflmz scam mama Hmaadp 00H: .umnmmmsm u m

.hcspm pamahso Bosh u:\so 0 0mm 0 cmmzpmn mmsam> mo smmz u <

000.0 H0.wwa mo<mm><
0N0.H 00.00H 00>0.0 0000.H 0.0: 0.0 0.0 m
H00.0 0:.»0H :H00.0 0000.0 0.0a 0.0 0.0 m
0HH.0 00.HHO 00HH.0 0000.H 0.00 0.0 0.0a a
000.0 00.00H 0HOH.0 0000.0 0.00 0.0 0.0 0
000.0 00.0»0 000H.: 000H.H 0.00 0.0a 0.0 0
000.0 00.000 :000.0 0000.0 0.00 0.0 0.0 0
0:0.0 . 0H.HOH 000>.H 000>.0 0.00 0.0 0.0 m
0:0.H 00.00H 0000.H 0000.0 0.00 0.: 0.0a b
000.0 00.0:H 00:0.0 0000.0 0.00 0.: 0.h 0
000.0 00.HHH 00H0.H 0000.0 0.0: 0.: 0.0 0
000.0 00.:0 00~H.0 00HH.0 0.0a 0.: 0.0 0
000.0 0:.0: 00~P.0 0000.0 0.00 0.0a 0.0 m
POH.0 00.00 0P0:.0 0000.0 0.00 0.: 0.0 0
0:0.0 00.0:0 00:0.0 000H.0 0.00 0.0 0.0 m
3580 8:98th 332 860823 u summzw almmzw
Aun\Sov a Aun\aov 0 A00 9 Am\av 0 mm meadow
3:0 00 :ofipmsdm muma mama
mmsHm> 0202 Mo QOmHnmmsoo

 

.A.0.coov 0 manme

107

circulating pump used in these studies may partially explain this
deviation, or the deviation may be due to incomplete mixing of the
water contained in the mini-corral as suggested by Weiler (1979).

Stratton's (1968) laboratory studies were made in a recirculating
stream and no wind data were provided. However, if the velocity of the
water is used for wind speed and the mass transfer coefficient is
determined as in Equation 36, the percent difference between the pre-
dicted and measured values ranges from 2.3% to 120% with a mean of
3h% (Table 9). Despite the large percent deviation no trend was ap-
parent from the data, and the average actual difference between the
predicted and measured values was only :_0.06h cm/hr.

Bouwmeester and Vlek (1980) generated an equation which predicts
flux as a function of total ammonia, pH, wind speed, temperature,

and fetch. Comparing the KZNH values predicted by that equation

with those derived from Equatiggw35 under the same conditions (fetch
omitted) revealed that the equation generated here produces estimates
of flux 1.5-ll.0 times larger than those of Bouwmeester and Vlek (1980).
This may be explained, to a small extent at least, by noting that
these investigators used what they termed 'free stream velocity' as

a measure of wind speed. This apparently refers to air speed above
the zone of decreasing velocities caused by friction with the water
(Liss, 1973). Examination of the values Bouwmeester and Vlek pre-
dicted and. those they obtained from laboratory studies reveal that

in 3 of the 9 trials they made, whenever the measured ammonia loss

5

rate exceeded 2.9 x 10- moles NH3/l/hr (5 out of 15 cases) they pre-
dicted only 97 to 55% of the loss which occurred. Comparisons of

Equation 36 with that proposed by Bouwmeester and Vlek (1980) also

108

reveal that when temperature, pH, and wind vary, the magnitude of the
change in flux that each equation predicts differs. Equation 36 shows
a slightly larger change in flux occurring per unit change in temper-
ature or pH than does the equation of Bouwmeester and Vlek (1980). A
variable change which would result in a doubling of flux by their equa-
tion results in a 2.5 fold increase using Equation 36. This discrep-
ancy is probably due to the different assumptions made in these equa-
tions concerning enhancement. A comparison of the effect of wind was
less clear as Bouwmeester and Vlek's equation tied the impact of wind
closely to fetch.

Champ §§_§l, (1973) report data which lack consideration of any
sources or sinks for ammonia which may be affecting the system.

Folkman and Wachs (1973) collected their data from an unbuffered system
raising questions about their pH measurements. Thus these two studies,
along with many others, display ammonia loss from aquatic systems but
do not provide data useful in testing Equation 36.

With the data available it is not possible to conclusively assess
the value of the generated flux equation (Equation 36). Although it
accurately reflects the data used to derive it, little independent
data exists to verify it. Possible shortcomings of the equation can
stem from several points. The dual layer theory itself is somewhat
suspect as it is difficult to imagine the two boundary layers remaining
physically intact and only being traversed via diffusion under the
broad variety of conditions occurring in nature. As mentioned ear-
lier, the temperature relationships used were based upon very little

data and did not consider different diffusion responses in the two

109
phases or different temperatures in the two phases. The estimates
made of laboratory air phase resistance for this study were subject
to a large degree of Judgement. Any errors made in these estimates
affect the estimate of water phase resistance and therefore the extent
of enhancement. The collection of data yielding both air phase and
total resistance under a variety of wind speeds and temperatures would
provide data allowing great refinement in the calculation of the coef-
ficients of the generated equation.

Other factors which were not considered at all in the formation
of Equation 36 may also be important. Relative humidity and possibly
sunlight appear to affect the rate of volatilization. Weiler (1977)
observed that there was no apparent loss of ammonia from his vessels at
night, but gave no reason for this phenomenon. Terry et_al, (1978)
report that ammonia volatilization is related to the evaporation of
water from wastewater sludges. In preliminary studies with C02,
Hoover and Berkshire (1969) found that condensation on the surface
of the water could effectively stop gaseous exchange. Quinn and Otto
(1971) suggest that these findings reflect a cessation in bouyancy-
driven convection within the boundary layer. They hypothesize that
this convection is the result of evaporative cooling of the surface
creating a density gradient. _0bviously, when the air is saturated,
evaporation stops and transport is inhibited.

Organic films at the surface of the water also probably would
impede transfer across the interface. The effects of turbulence
other than wind are largely untouched by research with the possible
exception of oxygen transfer. Even the effects of wind turbulence

probably vary depending upon the fetch of the water exposed

110
(Bouwmeester and Vlek, 1980) and its temperature. Although it is not
possible to include mathematical expressions for all of these factors
into an ammonia loss equation, those which have been included represent
a significant improvement over the most recent work in this area

(Weiler, 1979).

A_Predictive Equation for Concentration Change

For consideration of the impact that ammonia volatilization has
upon the ammonia concentration of a body of water,it is often useful
to rearrange Equation 36. If we consider the total amount of a sub-
stance in mmoles occurring in a body of water and call it A, then the
rate of loss from that body can be defined as:
%%-= -KC(AC)(surface area in cm2) (37)
where K and AC are as previously defined. Dividing both sides of

C

this equation by the volume of the system it becomes:

d(AC)= "KC (AC)

dt h (38)

where h is the depth of the system. Integrating both sides and defin-

ing AC at time zero to be ACo yields:

-KC t

< )
80 = ACOe h (39)

 

Based upon the variables previously defined this can be written for
ammonia as:

t
h»

3_w. (40)

) e-(KZNH

AZNH = (AZNH
3-w o

3-w
with t in hours and h in cm. The use of this equation is restricted

to situations where other sources and sinks of ammonia are negligible

111
and where the ammonia distribution in the bulk liquid remains com-

pletely homogeneous.

ECOLOGICAL IMPORTANCE

As with any study of this type, it is desirable to assess the
importance of the results with respect to the ecosystem. The results
of this and previous studies indicate that many species of algae are
capable of assimilating nitrogen very rapidly and in excess of their
immediate needs. This finding suggests that the link between the
carbon and nitrogen cycles through primary productivity is somewhat
flexible. However, the results of this study also indicate that at
the cessation of growth the cellular C/N ratio exhibited by the algae
under various nutrient limitations was much more uniform than expected.
0n the surface this would seem to imply that other than supplying more
nitrogen per unit carbon consumed to the herbivores consuming the
algae, the only ecological significance of nitrogen storage is to
allow those species which store nitrogen to grow after the nutrient is
depleted from the water column.

It must be remembered, however, that no data are available con-
cerning how nutrient limitations other than carbon and nitrogen affect
the cellular nitrogen content at cessation of growth. It is possible
that under phosphorus or some other limiting condition, cellular nitro-
gen levels may remain high until autolysis occurs. There is also a
lack of information concerning what degree of stress or what decline
in specific growth rate could occur before an algal population sinks
from the photic zone. Therefore, it is possible that although cellular

C/N molar ratios at the physiological extent of growth vary little,

112

113
larger amounts of nitrogen (with respect to carbon) may be lost from
the photic zone as stressed cells containing luxury nitrogen sink
(DiToro, 1980). This possibility is based upon the assumption that
the excess nitrogen taken by the cells is retained as they sink.

The information collected on the loss of dissolved organic
material from cells during this study suggests that the majority of
nutrients taken by algae are retained until the cell lyses, and there-
fore the rate of nitrogen cycling is directly tied to the life span of
individual cells. A portion of the material which is leaked and not
immediately recovered by the algae is undoubtably rapidly cycled
through the bacteria back to ammonia and may be retaken by the algae.
The nitrogen retained by the cells may be released in the upper strata
of the water column via autolysis but probably is retained until
after the cells sink and are ammonified by bacteria. In stratified
lakes, this may result in a net transfer of nitrogen from the tropho-
genic zone and a build-up of ammonia in the hypolimnion during the
summer.

The ammonia volatilization results presented earlier provide
additional information on a segment of the nitrogen cycle which essen-
tially has been ignored until recently. The identification of enhance-
ment may increase predicted loss rates at intermediate pH levels to
the extent that this mechanism is of major importance in some aquatic
systems.

Ammonia volatilization would be expected to have the greatest
impact in situations with high pH values and high ammonia concentra-
tions. A prime example of these conditions is Pond 1 of Michigan

State University's Water Quality Management Facility. During May,

11h
1977, the pH of this pond ranged from 8.65 to 10.30, the temperature
from 15 to 25 C, and the total ammonia in excess of equilibrium from
3.56 x 10‘7 to 3.37 x 10’1‘ moles NH3/9. (mean 1.21 x 10* moles 1013/2).
Mass balance calculations for this pond show a loss of about 16.3 mg
N/m2/hr during the month (King, personal communication). Predictions
of loss based upon Equation 36 result in an average rate of 15.6 mg
N/mg/hr. Thus, 96% of the total nitrogen lost may have been volatil-
ized as ammonia over this period.

Table 10 illustrates the half times for excess ammonia loss, or
time required for the excess ammonia concentration in a 5 meter deep
impoundment to decrease by 50% under a variety of conditions. In
this format, it is clear that at high temperatures dramatic losses
of ammonia can occur at pH levels as low as 8.5, and at higher pH
levels half times as short as 10 days are reasonable with winds of
only 2 m/s. Therefore, if excess ammonia concentrations are high,
significant nitrogen losses would be expected. .

The ammonia concentration Of the bulk water depends upon alloch-
thonous inputs, bacterial mediated ammonification, zooplankton and
phytoplankton release, sorption to the sediments, algal uptake, and
nitrification. Allochthonous inputs are probably on the increase due
to industrial operations but values for these inputs are currently
questionable.

Ammonia production via ammonification occurs wherever biodegrad—
able organic matter is available, but because most planktors sink
during senescence, it is most important in the tropholytic zone. The
ammonia produced in this way can be cycled immediately to the trOpho-
genic zone in shallow lakes or recirculated during overturn in strati-

fied lakes. In the latter case, ammonia probably builds up in the

Table 10.

115

Time (days) required to lose one-half of the ammonia in

the water in excess of atmospheric equilibrium from a
5 meter deep impoundment subjected to a 2 m/s wind.

 

 

pH 0 c 5 C 10 c 15 C 20 C 25 c 30 c
6.0 157,000 88,800 51,300 30,300 18,200 11,100 6920
6.5 h9,600 28,100 16,300 9580 5750 3520 2190
7.0 15,700 8880 5170 30h0 1830 1120 696
7.5 h960 2820 1630 967 583 358 223
8.0 1580 900 525 311 189 117 75
8.5 508 292 172 10h 65 hl 27
9.0 169 100 61 38 25 19 15
9.5 62 39 28 21 17 13 11
10.0 3h 26 20 16 13 11 9
10.5 25 20 17 1h 12 10 9
11.0 22 18 16 13 12 10 9
11.5 21 18 15 13 11 10 9

 

116
hypolimnion during summer stratification and comes to the surface as
a pulse during fall overturn. This pulse would occur at nearly the
same time as, and may be partly responsible for,a.fa11 algal bloom
which would raise the pHand increase the likelihood of ammonia loss.
Ammonification taking place under the ice creates a pool of ammonia
which is first exposed to the atmosphere as the ice melts. Shortly
thereafter, spring turnover occurs circulating the ammonia to the
surface. The following spring algal bloom would tend to raise the pH,
further increasing the likelihood of ammonia loss.

Recent research shows the plankton itself to be an important
source of’ammonia (Liao and Lean, 1978a). Zooplankton processing of
algae has been reported as causing the release of 10% to 50% of the
algal nitrogen, mostly as ammonia (Liao and Lean, 1978b). The same
investigators suggest that the phytoplankton also release a significant
amount of ammonia directly into the water column, but no evidence of
that was found in the current study.

In the ecosystem, the ammonia generated by the above pathways is
not necessarily available for volatilization, as other mechanisms
compete for the ammonia which is present. Under oxidizing conditions
nitrification is likely to occur unless the pH is low or extremely
high. Some portion of the ammonia may be bound temporarily to the
sediments but much of this would probably be released at turnover.
The third and probably most effective competitive pathway for ammonia
is uptake by algae. Most recent work points to ammonia being taken
preferentially over nitrate (Keeney, 1973). Liao and Lean (1978b)
report nitrate uptake is only 10-20% as great as ammonia uptake for
the systems they studied even though measurements of inorganic N

concentrations revealed relatively constant ammonia concentrations

117
and constantly decreasing nitrate concentrations. This type of evidence
suggests that ammonia is cycled through the algal community repeatedly
and that nitrate is used only when ambient ammonia is insufficient.
Therefore, whenever ammonia regeneration exceeds the rate of algal
uptake, nitrification, and sediment sorbtion; loss to the atmosphere is
likely.

It has been displayed that ammonia loss can be important in highly
enriched waters (Pond 1, water Quality Management Facility) and that
ammonia plays a vital role in the nitrogen cycle of all lakes. It
also can be shown that the ammonia volatilization pathway can play a
role in the evolution of nitrogen limited lakes. If a lake, phosphorus
limited and rich in micronutrients, is subjected to the activities of
man, the loading of phosphorus will probably exceed that of other
nutrients relative to algal requirements. Given this, one of two
things could happen. The lake could shift directly to a nitrogen limit
if the alkalinity was sufficient to provide carbon in excess of that
required by the algae to deplete the inorganic nitrogen reserves.
Alternatively, the lake could move to a carbon limit if the alkalinity
was low relative to the available nitrogen. In this situation, the pH
would be elevated due to the liberation of hydroxide ions as carbon
dioxide was assimilated from the carbonate and bicarbonate, and ammonia
loss would be likely, moving the system to an ultimate nitrogen limit.
The preceeding scenario appears to merit consideration, especially when
it is noted that many eutrophic lakes do display nitrogen limits
(Keeney, 1973).

Although the preceeding discussion has been built around ammonia
loss from aquatic systems, it must be realized that whenever the ammo-

nia concentration of the bulk solution falls below its atmospheric

118
equilibrium value, uptake from the atmosphere is predicted. Such
situations have been reported where atmospheric ammonia concentrations
were high (Hutchinson and Viets, 1969). Whether or not atmospheric
input of ammonia is likely under normal atmospheric concentrations
depends largely upon how efficient aquatic organisms are at removing
nitrogen from the water column. If the Michaelis-Menten half satura-
tion constants for ammonia uptake are near to or lower than atmospheric
equilibrium concentrations, a net flux into aquatic systems is possible.
Regardless of the direction of movement, it seems apparent that ammonia

flux should be considered in the calculation of nitrogen budgets.

CONCLUSIONS

Algal Microcosms

The following statements reflect the results obtained from the

laboratory algal microcosms:

1.

The extent of pH increase and carbon fixation is determined
by the nitrogen content of nitrogen limited microcosms.

Time of addition of nitrogen is unimportant as long as a
large proportion of the cells are viable when the nitrogen

is introduced.

Recarbonation of the media of nitrogen limited microcosms did
not result in extensive additional algal growth.

The rate at which carbon was accumulated by the algae in each
microcosm was the same until a nutrient limitation was
approached.

In nitrogen limited cultures, inorganic nitrogen uptake is

completed prior to the termination of carbon fixation.

, The addition of nitrogen to media containing nitrogen starved

cells results in an immediate increase in their N/C molar ratio.
The N/C molar ratio declines as media nitrogen is depleted and
to a less extent as media inorganic carbon is depleted.

The results reported by Atherton (197A) and Garcia (l97h)
concerning changes in N/C‘molar ratio with maximum culture

pH are in error.

119

10.

11.

12.

13.

19.

120
No conclusive evidence exists indicating that the cellular
C/N molar ratio changes with changes in pH or that internal
nitrogen is used as a buffer.
A better relationship exists between C/N molar ratio and
media free C02 than between C/N ratio and pH, but this
relationship is heavily weighted by data subject to large
measurement error and therefore is inconclusive.
The percentages of the total nitrogen and total carbon taken
by the algae which appeared as dissolved organic matter in
the media were approximately the same (about 15% if it is
assumed that microbial metabolism of the organic was negli-
gible).
Dissolved organic matter in the media was the result of non-
selective leakage or cell lysis. It was not possible to
determine the time course of leakage.
Some weak evidence suggests that there may be a relationship
between stress induced by high pH or low free CO2 and
increased leaking.
Carbon limited microcosms lowered the free C02 concentration
of the media to between 0.0010 and 0.0026 umoles C/l.
Carbon uptake curves resemble the rectangular hyperbola
suggested by the Monod or Droop equations except when some
other factor such as light intensity also is altered.
Algal growth rates appear to be directly related to cellular
N/C molar ratios when N/C ratios are below about 0.168.

Above this value, further increases in the ratio have no

effect on growth.

15.

16.

121
Throughout the current study, nitrogen uptake appeared to be
adequate to maintain optimal cellular N/C molar ratios when-
ever the media contained more than 0.10 mg inorganic N/l.
Below this level cellular N/C ratios and growth rate declined.
No clear evidence was found to suggest that media free 002

concentrations and cellular N/C molar ratios interact to

determine algal growth rate.

Ammonia Volatilization

The following statements reflect the current data gathered on ammonia

loss and information taken from literature on gas-liquid exchange:

1.

A linear relationship exists between flux and the difference
in the unionized ammonia concentration of the bulk liquid
from that which would be in equilibrium with the atmosphere,
but it does not adequately describe data from more than one
pH level at a time.

Ammonia loss from water is more closely correlated to a
combination of the differences in the unionized ammonia
concentration and the ionized ammonia concentration of the
bulk liquid from their respective atmospheric equilibrium
values than to the difference in the ionized fraction alone.
The dual layer model can be used to explain how ionized
ammonia increases the rate of transfer of ammonia across the
air-water interface.

The values obtained as coefficients of unionized ammonia
and ionized ammonia in the total mass transfer coefficient

based on the difference in the concentration of total ammonia

10.

11.

12.

122
from its atmospheric equilibrium, depend upon the pH range
of the data used.
Resistance to ammonia transport lies in both the water and
air phases. The relative importance of each phase depends
upon the pH of the water.
Air phase resistance dominates the system at low pH levels.
Mass transfer coefficients can be predicted for laboratory
vessels based on the effects of pH and temperature on the
ammonia mole fraction curve.
Variations in resistance to flux with wind speed can be
predicted using data provided by Liss (1973) and Weiler
(1977). Water phase resistance decreases exponentially
with wind speed. Air phase resistance decreases as a linear
function of wind speed.
Enhancement appears to have some effect on calculated mass
transfer coefficients between pH 7 and 11.5 and wind speeds
of 0 to 10 m/s.
Resistance decreases with an increase in temperature to a
greater extent than can be explained by changes in the mole .
fraction of ammonia alone. Himmelblau's (196A) data can
be used to develop a correction factor for this change.
Based upon enhancement estimates from laboratory data and
estimates of changes in resistance due to wind speed and
temperature, it is possible to predict a mass transfer
coefficient for a range of environmental conditions.
A variety of environmental factors which have not yet been
quantified are likely to affect ammonia transport across

the air-water interface.

123
13. Calculated mass transfer coefficients can be used to predict
ammonia concentration changes in well mixed bodies of water

of known dimensions.

LIST OF REFERENCES

LIST OF REFERENCES

Aaronson, S. 1978. Excretion of organic matter by phytoplankton in
vitro. Limnol. Oceanogr., 23:838.

Adamski, J.M. 1976. Simplified Kjeldahl nitrogen determination

for seawater by a semiautomated persulfate digestion method.

Allen, S.E. 1979. Chemical analysis of ecological materials..
Wiley and Sons.

Armstrong, F.A.J., and S. Tibbitts. 1968. Photochemical combustion
of organic matter in sea water, for nitrogen, phosphorus and
carbon determination. J. Mar. Biol. Assoc. U.K., h8:lh3-152.

Armstrong, F.A.J., P.M. Williams, J.D.H. Strickland. 1966. Photo-
oxidation of organic matter in sea water by ultra-violet radiation,
analytical and other applications. Nature, 211:981-983.

Atherton, M.J. 197A. The role of nitrate-nitrogen in algal growth.
Ph.D. thesis, Univ. Missouri-Columbia. 1A9 pp.

XBlain, W.A. 1969. Ammonia nitrogen losses from streams. J. Sanit.
Eng. Div. Proc. Amer. Soc. Civil Eng., 9711-978.

Bouldin, D.R., R.L. Johnson, C. Burda, and C. Kao. 197k. Losses of

inorganic nitrogen from aquatic systems. J. Environ. Qua1.,
3:107-11h.

’ Bouwmeester, R.J.B., and P.L.G. Vlek. 1980. Rate control of ammonia
volatilization from rice paddies. Submitted to: tmos. Envir.

Bradstreet, R.B. 1965. The Kjeldahl method for organic nitrogen.
Academic.

Brezonik, P.L. 1972. Nitrogen: sources and transformations in natural
waters, p. 1-50. In; H.E. Allen and J.R. Kramer (eds.), Nutrients
in natural waters. Wiley-Interscience.

Broecker, W.S., and T.H. Peng. l97h. Gas exchange rates between air
and sea. Tellus, 26:21-35.

Caperon, J., and J. Meyer. 1972a. Nitrogen-limited growth of marine
phytoplankton: 1. Changes in population characteristics with

steady-state growth rate. Deep-Sea Res., 19:601-618.

12%

125

Caperon, J., and J. Meyer. 1972b. Nitrogen-limited growth of marine
phytoplankton: II. Uptake kinetics and their role in nutrient
limited growth of phytoplankton. Deep-Sea Res., 19:619-632.

Champ, M.A., J.T. Lock, C.D. Bjork, W.G. Klussmann, and J.D. McCullough,
Jr. 1973. Effects of anhydrous ammonia on a central Texas pond,
and a review of previous research with ammonia in fisheries manage—
ment. Trans. Am. Fish. Soc., 102:73-82.

Danckwerts, P.V. 1970. Gas-liquid reactions. McGraw-Hill.
D'Elia, C.F., P.A. Steudler, and N. Corwin. 1977. Determination
of total nitrogen in aqueous samples using persulfate digestion.

Limnol. 0ceanogr., 22:760-769.

Dick, J.G. 1973. Analytical chemistry. McGraw-Hill.

'DiToro, D.M. 1980. Applicability of cellular equilibrium and Monod

theory to phytoplankton growth kinetics. Ecol. Modelling,

Donaghay, P.L., J.M. DeManche, and L.F. Small. 1978. On predicting
phytoplankton growth rates from carbon:nitrogen ratios. Limnol.
0ceanogr., 23:359-361.

Droop, M.R. 1973. Some thoughts on nutrient limitation in algae.

Droop, M.R. 1979. The nutrient status of algal cells in continuous
culture. J. Mar. Biol. Assoc. U.K., 59:825-855.

Dugdale, R.C., and J.J. Goering. 1967. Uptake of new and regenerated
forms of nitrogen in primary productivity. Limnol. 0ceanogr.,
12:196-206.

Emerson, K., R.C. Russo, H.E. Lund, and R.U. Thurston. 1975.
Aqueous ammonia equilibrium calculations: effects of pH and tem-
perature. J. Fish. Res. Ed. Can., 32:2379-2383. '

Eppley, R.W., E.H. Renger, and W.G. Harrison. 1979. Nitrate and
phytoplankton production in southern California coastal waters.
Limnol. Oceanogr., 29:983-999.

Eppley, R.W., E.H. Renger, W.G. Harrison, and J.J. Cullen. 1979.
Ammonium distribution in southern California coastal waters and
its role in the growth of phytoplankton. Limnol. OceanOgr., 29:
995-509-

Eppley, R.W., J.N. Rogers, and J.J. McCarthy. 1969. Half-saturation
constants for uptake of nitrate and ammonium by marine phyto-
plankton. Limnol. Oceanogr., 19:912-920.

126

Fogg, G.E. 1966. The extracellular products of algae. Oceanogr.
Mar. Biol. Annu. Rev., 9:195-212.

Fogg, G.E., and D.F. Westlake. 1955. The importance of extracellular
products of algae in freshwater. Verh. Internat. Verein. Limnol.,
12:219-232.

Folkman, Y., and A.M. Wachs. 1973. Nitrogen removal through ammonia
release from holding ponds, p. 503-520. In: S.H. Jenkins (ed.),
Advances in water pollution research, proceedings of the sixth
international conference held in Jerusalem, June 18-23, 1972.
Pergamon.

Garcia, F.C. 1979. Interacting carbon, nitrogen, and pH limits on
algal growth and development. M.S. thesis, Univ. Missouri-Columbia.
65 pp-

Gavis, J. 1976. Munk and Riley revisited: nutrient diffusion transport
and rates of phytoplnakton growth. J. Mar. Res., 39:161-179.

Gavis, J., and J.F. Ferguson. 1975. Kinetics of carbon dioxide
uptake by phytoplankton at high pH. Limnol. Oceanogr., 20:
211-221.

Gerloff, G.C.,zmuiFK Skoog. 1959. Cell contents of nitrogen and phos-
phorus as a measure of their availability for growth of Micro-
gystis aeruginosa. Ecology, 35:398-399.

Gerloff, G.C., and F. Skoog. 1957. Nitrogen as a limdting factor for
productivity for the growth of Microcystis aeruginosa in southern
Wisconsin lakes. Ecology, 38:556l56l.

 

Goldman, J.C., and J.J. McCarthy. 1978. Steady state growth and
ammonium uptake of a fast growing marine diatom. Limnol. Oceanogr.,
23:695-703.

Golterman, H.L. 1969. Methods for chemical analysis of fresh waters.
Blackwell Scientific.

Gordon, D.C., Jr., and W.H. Sutcliffe, Jr. 1973. A new dry combustion
method fortflxesimultaneous determination of total organic carbon
and nitrogen in seawater. Mar. Chem., 1:231-299.

Harris, K. Personal communication. A system for mass production
analysis for total phosphorus and total Kjeldahl nitrogen.
U.S. EPA, Corvallis, Oregon.

Harvey, H.W. 1957. The chemistry and fertility of a sea water,
2nd Ed. Cambridge Univ.

Hellebust, J.A. 1979. Extracellular products, p. 832-863. In;
W.E.P. Stewart (ed.), Algal.physiology and biochemistry
Univ. California.

127

Hicks, B.B., and P.S. Liss. 1976. Transfer of SO and other reactive
gases across the air-sea interface. Tellus, 26:398-359.

.______Hill, M.T. 1977. Interacting carbon and light limitation of the
kinetic sinking rate of Chlorella vulgaris. M.S. thesis,
Michigan State Univ. 71 pp.

Himmelblau, D.M. 1969. Diffusion of dissolved gases in liquids.

Hoover, T.E., and D.C. Berkshire. 1969. Effects of hydration on
carbon dioxide exchange across an air-water interface. J.
Geophys. Res., 79:956-969.

Hutchinson, G.L., and F.G. Viets, Jr. 1969. Nitrogen enrichment
of surface water by absorption of ammonia volatilized from cattle
feedlots. Science, 166:519-515.

Jolly, W.L. 1969. The inorganic chemistry of nitrogen. W.A. Benjamin,
Inc. Cited by National Research Council Subcommittee on Ammonia.
1979. Ammonia. University Park Press.

Kanwisher, J. 1963. On the exchange of gases between the atmosphere
and the sea. Deep-Sea Res., 10:195-207.

.1... Keeney, D.R. 1973. The nitrogen cycle in sediment-water systems. ‘_x
J. Environ. Qual., 2:15-28.

Kessler, E. 1969. Nitrate assimilation by plants. Annu. Rev. Plant

Ketchum, B .H. 1939. The absorption of phosphate and nitrate by
illuminated cultures of Nitzschia closterium. Am. J. Bot.,

Kevern, N.R., and R.C. Ball. 1965. Primary productivity and energy
relationships in artificial streams. Limnol. Oceanogr., 10:

7h.87 0

King, D.L. 1970. The role of carbon in eutrophication. J. Water Pol.

King, D.L. 1972. Carbon limitations in sewage lagoons. Nutrients and
Eutrophication,special symposia. Limnol. Oceanogr., 1:98-110.

King, D.L. Personal communication. Institute of water Research,
Michigan State University, East Lansing, Michigan.

King, D.L., and H.E. King. 1979. Interaction between light and carbon
dioxide availabilities as a control of algal species succession.
A report to the Missouri Water Resources Res. Center, Univ. of
Missouri-Columbia.

128

King, D.L., and J.T. Novak. 1979. The kinetics of inorganic carbon-
1imited algal growth. J. water Pol. Contr. Fed., 96:1812-1816.

Lange, N.A. (ed.). 1967. Handbook of chemistry. McGraw-Hill.

Liao, C.F.-H., and D.R.S. Lean. 1978a. Seasonal changes in nitrogen
compartments of lakes under different loading conditions. J.

Liao, C.F.-H., and D.R.S. Lean. 1978b. Nitrogen transformations within ~¢x
the trophogenic zone of lakes. J. Fish. Res. Ed. Can., 35:1102-
1108.

Liss, P.S. 1973. Processes of gas exchange across an air-water
interface. Deep-Sea Res., 20:221-238.

Liss. P.S., and P.G. Slater. 1979. Flux of gases across the air-sea
interface. Nature, 297:181-189.

Luebbers, E.H., and D.N. Parikh. 1966. The effect of algal concentra-
tion, luminous intensity, temperature and diurnal cycle or periodi-
city upon growth of mixed algal cultures from waste stabilization
lagoons as determined on the Warburg appartus. Proc. Purdue Ind.
Waste Conf. (2lst). Eg. Ext. Ser. No. 121, Part 1:398-367.

Lund, J.W. G. 1965. The ecolOgy of the freshwater phytoplankton.
Biol. Rev., 90:231-293.

McCarthy, J.J., and J.C. Goldman. 1979. Nitrogenous nutrition of marine
phytoplankton in nutrient-depleted waters. Science, 203:670-672.

McElroy, M.B., J.W. Elkins, S.C. Wofsy, C.E. Kolb, A.P. Duran, and W.A.
Kaplan. 1978. Production and release of N20 from the Potomac
Estuary. Limnol. Oceanogr., 23:1168-1182.

MacIssac, J.J., and R.C. Dugdale. 1969. The kinetics of nitrate and
ammonia uptake by natural populations of marine phytoplankton.
Deep—Sea Res., 16:95-57.

-Manny, B.A., M.C. Miller, and R.G. Wetzel. 1971. Ultraviolet combus-
tion of dissolved organic nitrogen compounds in lake waters.
Limnol. Oceanogr., 16:71-85.

Morris, 1. 1979. Nitrogen assimilation and protein synthesis:
p. 583-609. In; W.D.P. Stewart (ed.), Algal physiology and bio-
chemistry. Univ. California.

Nicholls, K.H. 1975. A single digestion procedure for rapid manual
determinations of Kjeldahl nitrogen and total phosphorus in natural
waters. Anal. Chim Acta, 76:208-212.

129
Okita, T., and S. Kanamori. 1971. Determination of trace concentration

of ammonia in the atmosphere using pyridine-pyrazolone reagent.
Atmos. Environ., 5:621-627.

Park, P.K. 1969. Oceanic CO system: evaluation of ten methods of
-investigation. Limnol. Oceanogr., 19:179-186.

Pavoni, J.L., M.W. Tenney, and W.F. Echelberger, Jr. 1971. The rela-
tionship of algal exocellular polymers to biological flocculation.
Proc. Purdue Ind. Waste Conf. (26th). Eng. Ext. Ser. No. 190,
Part 11:997-979.

Quinn, J.A., and N.C. Otto. 1971. Carbon dioxide exchange at the
air-sea interface: flux augmentation by chemical reaction. J.
Geophys. Res., 76:1539-1599.

Rasmussen, K.H., M. Taheri, R.L. Kabel. 1975. Global emissions and
natural processes for removal of gaseous pollutants. Water,
Air, and Soi1.Pollut.,9c33-69.

Rhee, G-Y. 1978. Effects of N:P atomic ratios and nitrate limitation
on algal growth, cell composition, and nitrate uptake. Limnol.
Oceanogr., 23:10-25.

Robinson, R.A., and R.H. Stokes. 1978. Solutions giving round values
of pH at 25 C, p. D-l39. In; R.C. Weast (ed.), CRC handbook
of chemistry and physics, 58th edition. CRC Press.

Rosenfeld, J.K. 1979. Ammonium adsorption in nearshore anoxic
sediments. Limnol. Oceanogr., 29:356-369.

Scheiner, D. 1976. Determination of ammonia and Kjeldahl nitrogen by
indophenol methods. Water Res., 10:31-36.

Semenov, A.D., V.G. Soier, V.A. Bryzgalo, and L.S. Kosmenko. 1976.
Photochemical mineralization of organic materials during the

determination of carbon, phosphorus, and nitrogen in natural water.

Sharp, J. H. 1977. Excretion of organic matter by marine phytoplankton:
Do healthy do it? Limnol. Oceanogr., 22: 381- -399.

Sharp, J.H. 1978. Reply to comment by S. Aaronson. Limnol. Oceanogr.,

Sievers, D.M. 1971. Carbon-phosphorus availability and eutrophication.
Unpublished Ph.D. thesis, Univ. Missouri-Columbia. 87 pp.

Soeder, C.J. 1979. Algal cultures, p. 178-181. 12; R.A. Vollenweider
(ed.), A manual for measuring primary production in aquatic envi-
ronments. Blackwell Scientific.

Solorzano, L. 1969. Determination of ammonia in natural waters by
the phenolhypochlorite method. Limnol. Oceanogr., 19:799-801.

130

Stratton, F.E. 1969. Nitrogen losses from alkaline water impoundments.
J. Sanit. Eng. Div. Proc. Amer. Soc. Civil Eng. 223-231.

Stratton, F.E. 1968. Ammonia nitrogen losses from streams. J. Sanit.
Eng. Div. Proc. Amer. Soc. Civil Eng. 1085-1092.

Strickland, J.D.H., and T.H. Parsons. 1968. A practical handbook of
seawater analysis. Bull. Fish. Res. Bd. Can. 167.

Syrett, P.J. 1953. The assimilation of ammonia by nitrogen-starved
cells of Chlorella vulgaris. Ann. Bot., 17:1-37.

.illllSyrett, P.J. 1956. The assimilation of ammonia and nitrate by nitrogen-
starved cells of Chlorella vulgaris. Physiol. Plant., 9:1-37.

Syrett, P.J.‘ 1962. Nitrogen assimilation, p. 171-188. 12; R.A. Lewin
(ed.), Physiology and biochemistry of algae. Academic Press.

Terry, H.E., D.W. Nelson, L.E. Sommers, and G.J. Meyer. 1978. Ammonia
volatilization from wastewater sludge applied to soils. J. Water

U.S. Environmental Protection Agency. 1979. Methods for chemical
analysis of water and wastes. EPA-625/6-79-003.

Vitula, 1.1. 1971 . Problemy stanoveni dusiku pri ruznych zpusobech
spalovani. Agrochemia 11:150-151.

Vlek, P.L.G., and J.M. Stumpe. 1978. Effects of solution chemistry
and environmental conditions on ammonia volatilization losses
from aqueous systems. Soil Sci. Soc. Am. J., 92:916-921.

weiler, R.B. 1977 (ms.). Loss of ammonia from lakes to the atmosphere.

weiler, R.B. 1979. Rate of loss of ammonia from water to the atmos-
phere. J. Fish. Res. Bd. Can., 36:685-689.

Yentsch, C.M., C.S. Yentsch, and L.R. Strube. 1977. Variations in
ammonium enhancement, an indication of nitrogen deficiency in New
England coastal phytoplankton populations. J. Mar. Res., 35:
537-555. '

Young, T.C. 1972. Relationships of carbon, phosphorus, and light to
the growth of Anacystis nidulans. M.S. thesis, Univ. Missouri-
Columbia. 68 pp. ‘

Young, T.C., and D.L. King. 1973. A rapid method of quantifying algal
carbon uptake kinetics. Limnol. Oceanogr., 18:978-981.

Ziesemer, C. 1979. The influence of carbon and light variation on
Chlorella vulgaris and Anacystis nidulans ability to maintain
planktonic populations. M.S. thesis, Univ. lissouri-Columbia,

76 pp.

 

APPENDICES

APPENDIX A

ALGAL MICROCOSMS - RAW DATA

131

132

Table A1. Composition of inorganic nutrient medium used for algal

microcosms.

 

 

Nutrient Concentration
NaHCO3 varied

KNO3 varied
CaC12-2H2O 28.5 mg/2
FeCl3 9.0 mg/l
MgSOh'7H20 20.0 mg/l
Nae-EDTA 1.2 mg/z
KH2POh 5.0 mg/l
Micronutrient Solution 1.0 mz/l

Composition of Micronutrient Solution:

H3BO3
MnC12-9H20
ZnSOh~7H2O
Na2MOOh-2H2O
CuSOh-5H20

Cu(NO3)2o6H20

2.86 g/l
1.81 g/t
0.22 g/l
0.25 g/l
0.0782 g/£

0.99 g/l

 

Table A2 .

Microcosm 1:1 and 1:1'

Total N0 -N added: 0.075 mg/9.

133

Raw data collected from algal microcosms.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3
0
es .9
e: .4 m
2'; \ .2 a r
‘\ a) 5 'o m '3
A d V A Hm 43A 0)"
U 0 A to" H at a) o: a: z:
.2 V 5 z. A ,2 3... ‘5... ‘5‘ Q\ s.
m 3 ON a 2 sen. 2.5.53.5 .8
39 5’ z 2 3° sanswssstsss
.a H '3 .s v v 0 '3 S? 0 m m a. m m 0
~- h -a a a a a) a a) I» h a:
m at z: 2: 2: r4 0 m o .4 o aic>lg o m o
m a. m I I I m.o -p.0 m u .9 A A -p h
E E 33 «a a: «l -p h .4 3 43+: r4+a.p.p .4-9
H O a: O O I: C d H Owl or-l-r-t Cor-t «4-H
Ev 54 a. z: z: 2: Eat: a.c> 5+2: a.== sczz a.z:
O 2300 10031 70820 0093 0000 000 0000 0080 0029 011 0280 --
65.0 26.0 --- 8.110 .030 .000 .00 0.00 0.00 0.23 .13 .030 .050
16105 2600 --- 80910 0003 0020 002 1035 0025 00119 00h 01.76 0100
215.0 26.0 --- 8.970 -- -- -- -- -- -- -- -- --
233.5 25.5 --- 8.990 .000 .021 .03 0.85 0.00 0.16 .09 .160 .050
279.5 25.0 --- 8.920 -- -- -- -- -- -- -- -- --
307 05 2505 ..- 80865 0000 0000 00h 1010 0000 0015 008 elhs 0100
382.5 25.0 1.116A 8.010 .056 .000 .02 0.00 0.00 0.21 .12 .265 .192
906.5 26.0 --- 8.0108 -- -- -- -- -- -- -- -- --
978J+25.5 --- 8.310 .009 .000 .06 1570 0.86 0.29- .11 .195 .116
55005 2600 -..- 80730 0000 0000 001 2025 0095 0008 011 0137 0068
618.5 26.0 --- 8.560 .009 .000 .01 1.25 1.20 0.16 .10 .063 .030
688.5 26.0 --- 8.575 .006 .000 .01 0.98 1.52 0.09 .12 .180 .193
779.5 26.0 --- 8.990 -- -- -- -- -— -- -- -- --
812.5 26.0 --- 8.6102 .016 .000 .12 0.95 0.85 0.29 .19 .115 .000
868.5 26.0 --- 8.560 -- -- -- -- -- -- -- -- --
912.0 26.0 --- 8.590 -- -- -- -- -- - -- -- --
A - Denotes that a new microcosm with fresh media was initiated at this
point.
S - Denotes seeding of new microcosm.
E - Reconditioned pH electrode used from this point on.
N - Denotes N03-N was added to microcosm after sample was taken.
0 - Denotes microcosm was bubbled with lab air after sample was taken.

 

Table A2.

(con't.).

Microcosm 1:2 and 1:2'

139

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Total NOB-N added: 0.075 Ins/2

... 3

3 Q o g «S

3 3 g 'o o '3
A V A HA gr‘ was
0 0 to FIG 0°! 6°! $4.!
,‘ ~’ 8 2: 0A (3" 55“ ;?\~¢H‘\ (”\s
u a " '01 :3 33 '§4:. :1 to 2’ 59'; 3’ ‘L 3’
a g 3? g E) a) bag 33 '3" .97" 2" .91"
.d a d .a v v ‘5 '8 £27 5 '8 5 ‘1’ 53' '3 5
v H 01-! :3 H G H) 34 w m In M
H z 2 2 Fl 0 O O H O O O I: O 0 O
a E . 'm a. a. 3" :se 3:; 22...: 2:;
-H m 33 :3 C) C) :n o 8 -H 8 God ‘vivi O-H od-fi
H B Q: a z 2 BO FAD 5'02 [AZ E4 2 m2
0 23.0 288 8.590 .093 .000 .00 0.00 0.00 H021 .11 .260 .170
65026.0 --- 8.590 .060 .002 .00 0.00 0.00 0.22 .11 .050 .030
.161. 26.0 --- .030 .003 020 .02 1.35 0.30 0.18 .05 .230 080
215.0 26.0 --.. .050 .. .. -_ .. .. .. .. -- ..
233.5 25.5 --- .030 .000 .021 01 2.75 0.00 0.15 .09 .190 060
279.5 25.0 --- .980 -- -- -- -- .. -- .. -- -—
307.525.5 -..- .955 .000 .000 .05 1.85 0.00 0.19 .12 .180 .095
382.525.o 9.33911 .690 .055 .000 .02 0.00 0.00 0.07 .08 .150 .100
906.5 26.0 --.. .6253 -- -- .- .- -- .. -- -- --
978.925.5 --- .710 .000 .000 .09 1.25 0.77 0.23 .12 .190 .105
550.5 26.0 --- .900 .000 .000 .00 .36 -- -- .09 .150 .080
618.5 26.0 --- .880 .005 .000 .01 .85 0.07 3.18 .09 .125 .090
88526.0 --- .875 .005 .000 .00 .60 0.35 3.16 .07 .172 .132
779.5 26.0 -- .850 -- -- -- -- -- -- -- -- --
812.526.0 --- .9501: .015 .000 .11 18 0.97 3.26 .12 .165 .073
868. 26.0 «- .880 -- -- -- -- -- -- -- -- -
12.0 26.0 --- .890 -- -- -- -- -- -- -- -- --

 

 

Table A2.

Microcosm.1:3 and 1:3'

(con't.).

135

 

 

 

Total 1703-17 added: 0.32? Ins/9.

0

" 7»

°* .4 d

2? \ .2 45 ‘H

\ 2 5 .5 ., 9
A U. V A p-{A 43A was
0 0 to He! we! use! not
A v .5 5 A A .2: a: 5\ 2““ W
a: 0 N at 03 §\ \ '6 g g g 3 a. 3
a 9 5’ S 3» go .9ng wrav 3v
2 *8 2 .. A A o 2 2 .5 a .373: .5 . 8 8
v in 0H I: in I: m h m In no
r3 2 z z .2 O 0 O H o 0 O H O 0 O
a 8 'm a. 'm .5 :2 3.5 :53: 2.5
v! o 3 :3 O o a: O 21 or! 3 00H «4*! 0..-I H on
El E-i D: = z z E-‘O tho s-«z Ina Bz L'Az
0 23.0 1.031 7 820 .350 .000 .00 0.00 0.35 0.22 .07 .550 .h20
65026.0 -- 8180 .285 .002 .00 0.00 0.00 0.23 .07 .150 .110
61.5 26.0 - 9 1:90 005 020 .02 2.h8 0.10 0.h3 07 .380 .120
15.0-26.0 -- 9.770 -- .— .. .. .... AA .. .. ..
233.5255 -- 9.880 005 .019 .01 11.25 0.00 0.h1 .10 .1420 .070
27h.5 25.0 -- 9.870 -- -- -— -— .. .. .. -- --
307.525.5 -- 9.875 .000 .000 .05 3.55 0.00 0.h6 .11 .395 .110
382.5 25.0 1.051A8.020 .290 .009 .02 0.00 70.00 0.07 .05 .h10 .3h0
h06 .5 26.0 -- 8.0708 -- -- ..- —— .. .. -- -- -_
h78.h25.5 -- 8.h80 .207 .007 .03 0.9h 0.h2 0.26 .07 .h39 .300
550.5 26.0 -- 9.h50 000 .000 .00 2.65 0.7!. 0.36 .08 .3h0 .090
18526.0 -- 9.700 oou .000 .01 .00 0.20 0.1.2 .12 .292 .015
688. 26.0 -- 9.720 .006 .000 .00 .88 0.30 0.10:. .08 .370 .110
77h .5 26.0 -- 9-700 -- -- -- -- -- -- -- -- --
812.5 26.0 -- 9.7901; .003 000 .11 h.06 0.1+7 5.1.1. .10 .105 .065
868.5 26.0 -- 9.710 -- -- -—- - -- -- -- -- ~-
12 .0 26.0 -- 9.700 -- -- -- -- -- -- -- -- --

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table A2 .

Microcosm 1:3 and 1:3'

Total NO -N added: 0.327 Ins/9.

(con't.).

135

 

 

 

3

A 3

or H d

A 3 g g A 3 .. g
o u m Ha'?3§: 2:
A "’ 3 “'7‘ A A .3: ‘5’: fig. 2; 33. £3,
a g >5 cm i i a? ME 3" my my ”V
O +3 :4; z a g Av 2V 0 :2 g r: 33 c: S I:
.c d a .a A, ~a c> 6 52’0 d o 0.4:- d 0
A. u -n a u a a) a ufl u) u u:
g. '3 5 5 5 2333 523252 ”a

“E 0 is :: cr‘ c9] 3;" '3 g '4 3 -u.p rc+:-p-p 35.9
a 24 n. a: z 2 so Eu 32' E212 3'5 EE
0 23.0 1.031 7.820 .350 .000 .00 0.00 0.35 0.22 .07 .550 .h20
65026.0 -- 8.180 .285 .002 .00 0.00 0.00 0.23 .07 .150 .110
161.526.0 - 9.h9o .005 020 .02 2.h8 0.10 0.143 .07 .380 .120
21507260 -- 9.770 -- - -- -- ... -- .. .. ..
233.525.5 -- 9.880 .005 .019 01 11.25 0.00 0.h1 .10 .h20 .070
27h.5 25.0 -- 9.870 -- -- -- -- -- -- .- -- -—
307.525.5 -- 9.875 .000 .000 .05 3.55 0.00 0.16 .11 .395 .110
382.525.0 1.051A8.02o .290 009 .02 0.00 0.00 0.07 .05 .h10 3140
h06 .5 26.0 -- 8.0705 -- -- -- -- -- -- -- -- .-
h78.h25.5 -- 8.1480 .207 .007 .03 0.9h 0.h2 0.26 .07 .h39 .300
550.526.0 - 9.h50 .000 .000 .00 2.65 0.7h 0.36 .08 .3140 .090
18526.0 -- 9.700 .00h .000 .01 3.00 0.20 0.12 .12 .292 .015
533, 25,0 -- 9.720 .006. .000 .00 11.88 0.30 5.1.5 .08 .370 .1h0
77h .5 26.0 -- 9.700 -- -- -- -- -- —- .. -- ..
312,5 25,0 -- 9.79OE.OO3 .000 11 h.06 0.h7 5.171. .10 .1705 065
868. 26.0 -- 9.710 -- -- -- - -- -— .. .. ..
12.0 26.0 -- 9.700 -- -- -- —- -- -- -- .. ..

 

 

 

 

 

 

 

 

 

 

 

 

 

 

136

Table A2. (con't.).

Microcosm l:h and l:h'

 

 

 

Total NOB-N added: 0.327 mg/£
A 0
A z o g 3?.
g 5 g ... .. 5
A V A HA .pA [0A
0 0 to H o: a or d or 3.. a
v E 2 0A Lo" .5\ Q\ t...\ m\
"‘ V ' " " "‘ °' ° °' 5’ 2.? '3 5 °‘ 2
'” " o‘" i i 5‘ 5‘ 3A 5A 5A 5A
a g 5’ z a) E) SE Hg 0 H 3.. H
.5 5 5 .. A A o 5 25 5 5 a: 5 5 5
v In H c: In :3 no 34 m no Lo m
'3 z. ’7 z. 8.2 3 B '5‘ 2 3 2 '5' 8 .8 8
g E a m N m 49 a H a +3 «P H +9 +3 +3 H 4-?
o a: O O :1: o 0H o H 0H 0H o or! on or:
E4 E! Dc 2 z z E-4 U L's. U 54 2 It. 2 E4 z ta. 2
0 23.0 5.288 8.520 .350 .000 .00 0.00 0.00 0.37 .13 .530 .hzo
65.0 26.0 -- 8.610 .285 .002 .00 0.00 0.00 0.21 .09 .160 .730
161. 26.0 -- 9.200 .000 .020 02 2.15 0.10 0.h0 .05 350 .060
215.0 26.0 -- 9.365 -- -- -- -- -- -- -- - --
233.5 25.5 -- £105 000 .016 .01 17.90 0.00 0 19 .09 .370 .070
2714.5 25.0 -- .1100 -- -- -- -- -- -- -- -- --
307.5 25.5 -- 9.th .000 .000 .oh 11.90 0.00 0.115 .12 .1100 .070
382.5 25,0 h.320A 8.660 .271 .009 .01 0.00 p.00 0.15 .06 .162 .363
h06.5 26.0 -- 8.6508 -- -- -- -- -- -- -- -- --
h78.h 25.5 -- 8.735 .213 .006 .03 0.85 0.65 0.211 .06 .390 .320
550.5 25.0 -- 9.135 .000 7.000 .00 3.08 0.7h 0.39 .06 .337 .078
515.5 25.0 -- 9.285 .005 .000 .01 .00 1.00 0.h1 .08 .338 .052
88.5 25.0 -- 9.300 .000 .001 00 .55 L60 0.1.2 .12 .3h2 .173
77h.5 26.0 -- 9-207 -- -- - u "' "' " -- '-
812.5 25.0 -- 9.285E .003 .000 11 3.1+8 1.25 D.h2 .21 .380 .025
868.5 26.0 -- 9-265 -- -- -- -- -~ - -- -- --
12.0 26.0 -- 9-270 -- -- -- -- -- -- -- -— --

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table A2.

(con't.).

Microcosm.l:5

137

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Total N03—N added: l.h5 mg/z

A 0’

or H “g

'3 \ .3 5 m

I; a, 5 'U o '3
A V A HA +3" (DA

0 ¢’ no Hot on! do: 8...!
A v .5 5 A A .52: 5:: 5‘ :2\ M 5‘
a o a: .1 ea 5~A ~A :u E, 2’23}? p‘fi?
é 3 >. C) ~A - g) o a) r4~¥ m~a rose o~a
O +3 :3 z 3 2 3V 3" a) c: g c: 3 I: S :3
.n d a .a ~a ~r c> 8 53'0 a o 0.:: d o
v k 0H I: h c: an A to to h m
5 5 5 5 55.55 525255 55

E g» 3 5 0" .9 ..m 55:15 55 555:: :5
e: 24 a. z: a: :z 24:: &.c> 612: E32: 21:: 5.5!
0 23.0 1.031 7.820 1.117 .000 .00 0.00 0.00 0.00 00 1.65 1.h7
65.0 26.0 -- 8.110 1.11 .002 .00 0.00 0 00 0.017 00 1.53 1.1m
161.5 26.0 - 9.390 1.03 .019 .02 1.75 0 20 0.50 00 .52 1.25
215 .0 26 .o -- 9.810 -- ..- -- _- -- .. .-.. A- --
233.5 25.5 -- 10.030 7h0 022 .01 3.65 0.00 0.95 .03 1.116 .860
2724.5 25.0 - 10.325 -- -- -- -- -- -- -- -- --
3075 25.5 -- 10.635 .370 .032 .03 5.60 0.00 1.38 .16 1.12 .560
332.5 25.0 -- 10.760 .260 016 .01 10.0 0.00 1 35 .20 1.33 .125
1106 .5 26.0 -- 10.815 -- - -- -- -- -- - -- --
2.73.1255 -- 10.730 .2h5 .008 .03 10.0 1.12 1.60 .05 .170 .h60
550,525.73 -- 10.685 .2111 .013 .01 10.11 2.18 1.140 .26 .317 .130
18.5 26.0 -- 10.685 .225 .010 .01 9.80 2.20 1.35 .31 .29 .h7o
33525.0 - -- .235 011 .00 10.8 1.95 1.1714 .35 .27 .h80
1.. 25.0 -- 10.520 -- -- -- -— -- -- -- -- --
812; 26.0 -- 10.550131 .220 .015 .12 8.50 2.25 1.12 .51 55 .600
868.5 26.0 -- 10.1130 -- -- -- -- -- - -- -- --
12 .0 26.0 -- 10-h10 -- -- -- -- -- -- —- -- ~-

 

 

Table A2.

Microcosm.1:6

(con‘t.).

138

 

 

 

Total NO3-N added: l.h5 mg/£
A 3
a H d
A \ O “4
a g ”I. 34 a) '3
8 g V 5 A H pm (“A
m H o: 0 01 a! a :4 at
A v .5 5 A A .2: a: 5\ 2V“ 4"
no 0 N or 01 §\ \ '6 g g g E a. g
g g 5. g \ \ m g) .94 g HV 0V lav 0v
3 g q-c 5’ g) luv .OV £7 1:: S t: 3 c: :3 :3
2. .. .‘3 " V " ° 2 2 c: 8’. 2 8% °‘ 88 d ‘”
$4 an
0 .2 z z z I; O o O H O m 0 a; O 0 O
8 8' 'm 'm 'm «9'0 :3” 3.5 3.52.5 3“
a Eu: 3 =5. 0 o a o 21 *1 S! o «4 «4 «4 o H on fa
z z z E! 0 tn. 0 E4 z (a. .2 E4 2 In. a
0 23.0 h 288 8.510 1.57 000 .00 0.00 0.39 0.00 .00 1.57 1.67
65.0 26.0 -- 8.610 1.145 002 .00 0.00 0.00 0.03 .00 1.29 1 28
161. 26.0 - 9.060 1.07 019 02 1.h0 0.28 0.h3 .00 1.56 1.29
215 .0726.0 -- 9 .330 ..- .. .. .. .. .. .. .. --
233.5 25.5 -- 9.130 .6110 .020 01 17.70 0.00 1.13 .06 1A3 .790
2711.5 25.0 -- 9.595 - -- -- -- -- -- -- -- .-
307.5 25.5 -- 9.780 .009 .000 .03 8.95 0.00 1.57 .09 1.30 .220
382.5 25.0 -- 9.995 .000 .000 .02 13.7 0.00 1.517 .114 1.13 .110
h06 .5 26.0 -- 10.070 -- -- -- -- -- -- -- -- _-
h78.h 25.5 - 10 080 .000 .003 .03 11+.6 1.15 1.67 .12 1.32 .180
550.5 26.0 -- 0.0 090 .000 .000 .01 16.7 1.50 1.50 .19 1.25 .152
618.5 26.0 -- 10 1h0 .005 .000 .01 16.1 1.75 1.12 .117 1.38 .100
688. 26.0 -- 10 120 .005 .000 00 13.7 0.70 1.55 .16 1.25 .260
77h .5 26.0 -- 10 090 -- -- -- -- -- - -- -- -
812.5 26.0 -- 10 15073} .006 .000 .12 12.2 1.33 1.52 .22 1.18 .270
868.5 26.0 -- 10 100 -- - -- -- -- -- -- - --
12.0 26.0 -- 10 100 -- -- -- -- -- -- - -- --

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table A2. (con't.).

Microcosm.l:7

139

 

 

 

 

 

 

 

 

 

 

 

 

 

Total NOB-N added: 2.85 m3/£

A 3

" vi a

A \ O '94
o o m HQ'Jafi: 2:
V a z 0" HA \ \ §4\ \

A v | A A ...-q 61 o .3 '5 Q ad:
a d) (\l or at 5\ \ 'U E) g g g) g
g Q s. g \ \ m 0 g HV my my 0V

0 v - 3’ 3’ 2’5 2.. 0 ,3 g 2 a ...
.c: a: a .a v v o a Q o a a) 2‘3 5 '3 5
v 34 H :1 $4 1: no 34 to to h m
0 '3‘ z z z H O o O O 0 o O a: O
O Q: I l I d .0 +3 .0 '3 $4 45 $4 '3 Lo +3 H
.5 5 3 2 .5 ow 2°" 85:38 8" ...... 5»
a E-c o. z z z a o III. 0 a 2 E: '2 a '3 E "z"
0 23.0 1.031 7.920 2.92 .000 .00 0.00 0.15 0 00 00 2.77 3.00
65.d 26.0 -- 8.210 2.90 .002 .00 0.00 0 00 0 00 .00 2.63 2.50
161.5 26.0 -- 9.1415 2.177 .021 .02 1.80 0.35 0 36 .00 2.88 2.55
215 .0 26.0 -- 9.905 -- -- -- -- -- -— —— -.. . .-
233.5 25.5 - 10.160 1.96 .030 01 5.23 0.00 1.17 .02 2.92 2.0h
2714.5 25.0 -- 10.1470 -- -- -- -- -- -- .- -- --
307.5 25.5 - 10.750 1.53 .oho .01 8.58 0.00 1.88 .00 2.65 1.6!.
1.06 .5 26.0 -- 10.890 - -- -- -- -- -- -- -- --
1731 25,5 -- 10.850 1.31: .026 .02 8.70 2.70 2.17 .29 2.78 1.62
550,5 25,0 -- 10.825 1.21 002 .00 12.2 2.82 2.19 .33 2.67 1.65
l618.5 25,0 -- 10.805 1.38 .016 .01 12.0 2.90 2.03 .37 2.75 1.63
688.5 26.0 -- 10.760 1.27 01h 01 6.00 1.50 2.117 .176 2.61; 1.53
77h .5 26.0 -- 10.665 -- -- - -- -- -- - -- -—
812,5 26,0 -- 10.750812? 018 12 8.75 2.82 2.117 .hT 3.08 1.96
868.5 26.0 - 10.690 -- -- - -- -- -- -- -- --
912.0 26.0 -- 10.6110 -- -- -- -- -- -- -- -- -

 

 

 

 

 

1110
Table A2. (con't.).

Microcosm 1:8
Total NO -N added: 2.85 mg/i

 

3
0
A +3
Q r-i d
A \ O .5 $0
°' 3’ "* '3
\ 5 "U Q)
A 0‘ V A Hap/5 [an
O 0) b0 HO! 0°? “0! ‘40!
v E z 0" $4" ,5\ fi\9—0\ d)\
co 0 N 03 02 §\ \ "d
g 3 >5 0 \ \ m a)? g—{V 0v av my
0 4: 3:3 z 8’ 2" 39533 ”2328: S:
.c d c .a 5’ 2, «o «3 £24, «301:; 0 a 0
v H 0H ‘3 INC. 80 $4M m MM
0 H z z 2 H0 00 H0 UOHO 0°
0 B: d I I I “.0 Pro “‘4 4934”“ PM
8 :fi «1 a: to ‘9 a .4 a .p.p r1+= p.» :4.9
w-I a.) m o O :3: 0 H6 O-u-i «4-H 00H «4-H
E1 E0 < Q! 2 z 2 EU [AU [-12 (2425423 :52.

 

O

23.0 b.228 8.520 2.92 .000 .00 0.00
65.0 26.0 .. 8.6102.88 .002 .00 0.00
161.5 26.0 .- 9.195 2.32 .021 .02 2.35
215 .0 26.0 -- 9.17.85 -- -- -- --
233.5 25.5 -- 9.625 1.60 .032 .01 7.18
27h.5 25.0 -- 9
307.5 25.5 -- 9.980 .690 .05h .01 10.2
382.5 25.0 .. 10.285 .0h2 .020 .00 17.8
1706 .5 26.0 .. 10.1.1.0 .. .. .. --
h78.1+ 25.5 -- 10-620 .026 .008 .03 23.3
550.5 26.0 ..- 10.795 .012 .006 .01 27.7
618.5 26.0 .. 10.915 .010 .006 .01 22.5

88.5 26.0 -- 10.890 .0h1 .007 .00 17.17
77h.5 26.0 .. 10.860 -- -- _- -_
812.5 26.0 .. 10.930E .012 .010 .12 6.0
868 .5 26.0 _- 10.865 .. .. __ _-

12.0 26.0 .. 10.860 -- -- -- ..

 

 

 

 

 

 

 

 

I000
I000

co. 0 o
NNNEHOIO
F'CDm \I'I\OIU'|

 

 

 

 

 

 

 

Table A2.

(con't.).

Microcosm.l:9

lhl

 

 

 

Total NO -N added: 5.65 mg/l

A 3

or H a)

A \ '0 ‘H

A g. g '15 1 3 . 2
U 0 m H or '3: 43: £27
A v 3 z. A A .2: s: 5\ 2* M a.“
g g A o“ a : 83,.3. 3§6§§§ .E’
O +3 13 z 2 g I?" 2V u G g c: 33 c: S. r:
:3 a .5 .3 ~u .. <3 : g : 55’3, g 0 cu m a o
0 a: z z z .3 O 0 0 Fl 0 o 3) 3° 3 8°
3 g: l l I .0 4P .O a! 34 «P 8-4 '3 h 43 H
.4 6 3 a: 6‘" 6‘“ a" 624.2133 ‘5’" ”4"” H”
E-c a n. z z z E0 0 cs. o 2-4 It." i: 2' 59 I": E E
o 23.0 031 7.830 5.77 .000 .00 0.00 0.00 0.00 00 5.57 5.72
65.0 26.0 - 8.100 5.90 .002 .00 0.00 0.00 0.06 00 5.10 5.19
61.5 26.0 -- 9.355 5.13 .020 .02 2.10 0.05 0.00 00 6.00 5.50

15 .0726 .o -- 9.705 -- --' -- -- -- -- -- -- --
233.5 25.5 -- 10.050 h.8h .027 .01 h.05 0.00 0.67 01 5.73 h.95
27h .5 25.0 -- 10.355 -- -- -- -- -- -- -- -- -- '
307.5 25.5 -- 10.630 0.31; .001 .01 8.1.0 0.00 1.69 .00 -- h 20
382.5 25.0 -- 10.830 h.22 .0143 .00 10.1 0.00 1.86 00 5.1+2 0.27
1406.5 26.0 -- 10.870 -- -- -— -- -- -- - -- -—
h78.h 25.5 -- 10.850 14.00 .000 .01 8.70 1.68 2.18 .00 5.1.0 0.30
550.5 25.0 -- 10.850 3.85 .032 .01 10.7 3.13 2.25 .00 5.23 h.21
513,5 25,0 -- 10.855 h.20 .031 .01 10.0 3.15 2.16 .09 5.27 14.27
88.5 26.0 -- 10.790 3.85 .030 .12 9.25 1.23 2.00 .10 5.27 3.97
77h.5 26.0 -- 10.730 -- -- -- -- -- -- -- -- —-
812.5 26.0 -- 10.77011 3.85 .018 .13 9.00 2.30 2.32 25 1.61 3.11;
868.5 26.0 -- 10.750 -- -- -- -- -- -- -- -- --
12 .0 26.0 -- 10.720 -- -- -- -- -— -- -- -- --

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table A2.

(con't.).

Microcosm 1:10

 

 

 

Total N0 ~N added: 5.65 mg/z
A 3
’3 i .3 lg .53
A .3 2 . ... . 9
o 0 V S": H: 32? ‘3’: 22?
A v 5 7 A A 2:: so 53. 25.33. a;
0) d) N oi oi \ \ Id
g s :2 9. g. r. 5.2.9.2 ravav 3v
.6 *3 2 .. A 5 6*”:8" 2835835 '88
V 8 z: z z z A 8 :3 8 A 8" 0 8" A 8° :3 3"
a) Q. «I I I I 0 .D +3 .0 Cd I-o +1 In 03 In 4D In
.5 8 3 A o" 69' A” 882:: 83:38:: 2::
E-I [-4 O: z z z E-I D It. 0 8 2 In 2 [-4 2 It. 2
0 23.0 1.288 8.510 5.80 .000 00 0.00 0.00 0.00 .00 5.65 5.80
65.0 26.0 -_ 8.590 5.87 .002 00 0.00 0.00 0.00 .00 5.28 5.28
7161.5 26.0 .... 9.070 5.08 .020 02 1.50 0.20 0.00 .00 5.90 5.05
215 .0 26.0 .. 9.320 -- -- - -- -- -- -- -- --
233.5 25.5 -- 9J415 14.71 .021 01 3.70 0.00 0.61 .01 5.60 17.83
2724.5 25.0 -- 9.570 —- -- - -- -- -- -- - --
307.5 25.5 -- 9.720 3.95 .oh8 .00 8.65 0.00 1.83 .00 17.92 3.80
382.5 25.0 -- 9.990 3.35 .056 .00 1h.5 0.00 2.58 .00 5.0h 3.16
1:06 .5 26.0 -- 10.090 -- ~- -- -- -- -- -- -- --
h78.h 25.5 .- 10.330 2.57 .065 .03 15.7 1.12 h.07 .00 5.06 2.87
550.5 26.0 -- 10.635 2.07 .053 .01 23.6 1.95 h.50 .76 h.87 2.66
618.5 26.0 A. 10.900 2.00 .oh7 .01 22.0 2.05 5.07 .22 h.61 2.18
688. 26.0 -- 11.050 1.78 0m: .01 2h.3 1.1.0 5.00 .39 0.71. 2.13
7714.5 26.0 -- 11.130 -- -- -- -- -- -- -- -- --
812.5 26.0 -- 11.225.81.68 .052 .12 30.3 3.1.2 5.00 .78 h.67 2.56
868. 26.0 -- 11.210 -- -- -- -- -- - -- -- --
912.0 26.0 -- 11.225 -- -- -- - -- -- -- -- --

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table A2. (con't. ).

Microcosm 2 :l

1173

 

 

 

 

A E O .—| A
A 3. 8° '3 1 3,. 5
0 w 00 c-I or '3 or g)
1; I 5 ”I 3 A .3: 8: at. :252 g
g g h c3N \ i 53 a} 33 0v 3
o 4" 3 z 3 E, {.95 35 a) c: "' r8
.: d c: a v v o 0 Q 0 '3 5
V I" 'H c: x. e no a no
a: H z z z A o 0 o H o o o z
I
§ ‘5" “’ 'm '6 'm 3g 2?, :2: :1: Am
2 2'3 3 3. 2 2 2 2.. 2:. 22 2:: a
0 27.0 015 8.115 .076 010 .00 0.h2 0.62 -- .10 .075
6805 2700 "'"' 8.390 .011 010 (.02 "" 003s -- 002
13h.5 26. -- 8.835 .005 000 <.03 -- 0.75 -- .10
206.8 27.0 -- 8.950 .001. 000 <.02 -- 0.75 -- .02
236.5 27.0 -- 8.950 -- -- -- -- -- -- --
27h.0 26.5 -- 9.015 .002 .000 .02 -- 0.62 - .02
31.72 26.0 -- 8.985 00317 .00 .oh -- 0.57 -- .06
h19.0 26.5 -- 9.135 .010 000 00 2.02 0.70 0.2!. 06 .201
169.0 26.0 -- 9.365 -- -- -- 3.60 -- -- --
h91.0 26.5 -- 9.h10 .000 .000 <.01 3.60 0.60 0.26 .0h
562.5 26.0 -- 9.th .000 .000 .00 3.57 0.50 0.21 .07
617.0 2605 -- 9.1415 -- -- -- -- -- 4"- ""-
632.0 26.5 -- 9.1+15 OOON .000 .02 2.92 0.69 0.21. .06
710.5 26.0 -- 9.570 -- -- -- -- -- -- -- .327
732.0 26.5 -- 9.695 .000 .000 .01 h.h0 0.58 0.1.0 .09
30,0 27.0 -- 9.730 .000 <.01 <.01 2.81 0.82 0.52 .08
99.0 26.0 -- 9.695 -- -- -- - -- -- --
72.0 26.5 -- 9.670 .001. .000 .01 -- 0.70 0.56 .06
64.0 26.0 -- 9.590 .000 .000 .00 -- 0.75 0.1.6 .01.

 

 

 

 

 

 

 

 

 

 

 

 

 

Table A2. (con't.).

Microcosm 2:2

lhh

 

 

 

A E ., A A
A E. 2’ '2 A 3,. 5
o u no He: 0).: E
3:25 >~ o“: 222.221.13
° ‘9 :3 g: 3’ 3’ §K§_:3~5 ” g 53,: g
.a d c .a ~a ~a <3 «I §?<v d o
\I h °H z: z: z: .4 5 h 5 u: “ “’ F:
2 E g a. '2. 'm 22 22 2'3 £33 .22
2 2 3 2 2 2 2 2.. 2., 22 22 2
0 27.01.3017 H8.7h5 .078 .010 .00 0.50 0.17 -- .02 .075
68.5 27.0 -- 8.775 .017 .011. <.02 —- 0.35 -- .00
13h.5 26.5 -- 8.915 .006 .000 <.03 -- 0.79 -- .02
206.8 27.0 -- 8.935 .001. .000 <.02 -- 0.32 -- .00
236.5 27.0 -- 8.970 -- -- -- - -- -- -
2711.07 26.5 -- 9.010 .000 .000 .01 - 0.56 -- .02
31:72 26.0 -- 8.980 .OO3N .000 .oh -- 0.67 -- .03
h19.0 26.5 -- 9.020 .010 .000 .00 2.03 0.38 0.26 .05 .201
169.0 26.0 - 9.110 - -- -- h.10 -- -- --
h91.0 26.5 - 9.170 .000 .000 <.01 3.90 0.57 .022 .06
552.5 26.0 - 9.180 .000 .000 02 1.62 0.1.0 0.26 .08
17.0 26.5 -- 9-175 -- - -- - -- -- --
32.0 26.5 -- 9.165 .OOON .000 .02 2.85 0.61; 0.28 .07
710. 26.0 -- 9-255 -- -- -- -- -- -- -- .327
732.0 26.5 -- 9.300 .000 .000 .09 3.10 0.70 0.1.0 .03
30,0 27.0 -- 9.330 5.01 <.01 <.01 2.87 0.81 0.1.1. .08
99. 26.0 -- 9-310 -- - -- -- -- -- --
72.8 26.5 -- 9.285 .007 .000 .01 -- 1.17 0.h8 .10
61+.o 26.0 -- 9.250 .000 .000 .00 -- 0.55 0.62 .10

 

 

 

 

 

 

 

 

 

 

 

 

 

Table A2. (con't.).

Microcosm,2:3

th

 

 

 

A 3 o A 1

i 2 - g i

c) o In .47a‘ «In \,

1; I 3 z‘... 3 2: 3333 53233:...

2 a 2 2 \ \ 22 22 5.. 2v 3

2 :3 2 .. 3’ 3’ 2v 2" :22 2 2 3

V 3 ”" z z z 3 A 5 no u co 2

D: g l I I '39 «g .0 g 2 3 2 lm

-§ 5 53 :n cf, cfi‘ =§n '3 3 'd 5 '3'" '4'" C)

a 54 o. z z z so ED 93 2‘2 5'3

0 27.0 1.015 8.110 .330 .016 .00 0 16 0.07 -- .07 .327
68.5 27.0 .— 8.h10 .250 .035 <.02 -- 0.17 - .03
131.5 26.5 -- 9.11:0 .063“ .010 <.03 -- 0.57 -- .10
206.8 27.0 - 9.510 .001: .000 <.02 -- 0.85 -- .03
236.5 27.0 -- 9.6h0 -- _- -- .. .. .. ..
2717.07 26.5 -- 9.705 .000 .000 .02 -- 0.6h - .0h
31.7.2 26.0 -- 9.700 .OO3N .000 .0h -- 0.u7 -- .03

1.19.0 26.5 -- 9.820 .015 .000 .00 2.99 0.71 0.68 .06 .607
169.0 26.0 -- 10.070 -- -- -- 6.70 -- A- A-
191.0 26.5 - 10.160 .000 .000 <.01 6.80 0.68 0.58 .05
552,5 25,0 —- 10.210 .000 .000 .02 3.08 0.69 0.75 .05
617.0 26.5 -- 10-150 -- - -- -- -- -- -
632.0 26.5 - 10.180 .oocw .000 .01 h.25 0.72 0.70 .11

710.5 26.0 -- 10-185 -- -- - -- -- -- -- .887
732.0 26.5 - 10.260 .095 .010 .01 6.20 0.60 0.93 .16
30,0 27,0 ..- 10.280 .062 <.01 <.01 6.60 1.17 0.8h .13
99 .0 26.0 -- 10.230 -- -- -- -- - -- --
72.0 26.5 -- 10.210 .0145 .011: .01 -- 1.21: 0.95 .1h
u.0 26.0 -- 10.1h0 .020 .000 .02 -- 1.37 1.05 .11

 

 

 

 

 

 

 

 

 

 

 

 

 

Table A2.

Microcosm 2zh

(con't.).

lh6

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A 3 .... A

.. go .3 +3 3

A \ 5 ”A 3’

U 8. v 00 H: 3.: v

A V 3 ’7 A A .32? 8: 5‘ 2‘ rd

m o a: at ca 5- - '6 g) E” 0

9 9 5' 2 ‘ \ 63° 33° 8" .9." "d

0 +3, H 3’ 3’ I-Iv .Dv :: .O I: '8

.r: d r: .a v v o a Q o «I a: ,

" 3 33 z: z: z: 8 3 5 .4 3’ H u: ‘7
ID 0

0 g: d I I I '3.0 .p.0 0 u .p a ‘“

-§ 0 33 2: cf“ cf" :3“ '3 3 13 3 '3'” '4'" :2

E4 54 p. -z: z: z: E4LJ 2.02 PIE; 3353 N

0 27.0 lL301; 8.750 .330 .018 .00 0 17 0.25 -- 6 .327
68.5 2700 -- 80795 0252 .039 <.02 "- 0032 "' 01
13h.5 26.5 -- 9.035 .013 .002 <.03 -- 0.h5 .. 07
27.0 -- 9.225 .00h .000 <.02 -- 0.72 -- 02
27.0 -- 9.300 -- -- -- -- -- -- --
26.5 -- 9.330 .000 <.01 .oh - 0.66 -- .01.
26.0 - 9.325 .0010: .000 .03 -- 0.55 -- .01;

26.5 -- 9.380 .010 .000 .00 3.27 0.80 0.66 .oh .607
26.0 -- 9.500 -- -- -- 8.00 - .. ..
26.5 -- 9.530 .000 .000 .01 h.50 0.80 0.58 .08
26.0 -- 9.570 .000 .000 .02 h.h5 0.56 0.72 .11
26 .5 -- 9.570 -- -- -- -- -- -- --
26.5 -- 9.550 .OOON .000 .02 3.22 0.h8 0.68 .06

26.0 -- 9.655 -- -- -- -- -- -- -- .887
26.5 -- 9.710 .000 .000 .01 5.70 0.88 0.92 .1h
27.0 -- 9.790 <.01 <.01 <.01 11.5 1.21. 0.96 .17
26.0 -- 9.765 -- -- -- -- -- -- --
26.5 -- 9.755 .001; .000 .01 -- 1.00 1.06.09
26.0 -- 9.710 .000 .000 .00 -- 0.96 1.0h .10

 

 

Table A2.

Microcosm 2:5

(con't.).

11.7

 

 

 

A E H A

a 0 a

,. 2. 3° 3 5 g.

0 m to a: '33...

’3 .. V 'm 2': 2: ...... 3°! 53093”:

g g :0 o \ \ 5?) 0}) 3V ”V 8

o .p 0H :z 5’ N’ NIB .4 a a F. '3
:5 2 .5 .. - 5 8" 2" :28 2 3

o 8. '3 2| 2' =7 g§ 33' 5 .-I 3° :3 3’ z:

3 E +3 D d h .p h m

-fl 0 33 :2 cf) cf“ :Q“ '3 Q g: Q -g-p ad-p g;

B a ‘3‘ z z z 540 no s-c'é' E; N '~

0 27.01.015 8.115 .900 .018 .00 0.00 0.12 -- . 0 .887

6805 27.0 -'- 80510 0791‘ 00,45 <.02 -" 0.22 -" .00
13h. 26. -- 9.225 .602 .050 <.03 -- 0.60 - .01
206.8 27.0 -- 9.7h0 .220 .035 <.02 -- 0.67 -- .01.
236.5 27.0 -- 10.0145 -- -- .. .. .. .. __
27h.0 26.5 -- 10.390 .032 .000 .03 -- 0.80 -- .05
31.7.2 26.0 -- 10.555 .006 .000 .oh -- 0.67 -— .11
h19.0 26.5 -- 10.555 .010 .000 .00 9.65 0.92 0.88 .09
169.0 26.0 -- 10.5600 -- -- -- 7.70 -- -- .-
1:910 26.5 -- 8.770 .000 .000 <.01 7.1+0 0.90 0.81. .08
552,5 25,0 -- 8.910 .000 .000 .02 9.35 0.76 0.9». .07
617.0 26.5 -- 8.91m -— - -- ~ -- -- -- —-_
632,0 26,5 -- 8.8h0 .000 .000 .02 9.h0 0.75 0.92 .10
710.5 26.0 - 8.860 -- -- -- -- -- -- --
732,0 25,5 -- 8.830 .000 .000 .01 8.50 0.82 0.89 .07
30,0 27,0 -- 8.735 <.01 <.01 <.01 7.60 1.0h 0.95 .06
99.0 26.0 -- 8-725 -- -- -- -- -- -— --
72,0 25,5 -- 8.560 .000 .000 .01 -- 1.07 1.02 .09
11.0 25.0 -- 8.070 .000 .000 .02 -- 1.17 1.02 .08

 

 

 

 

 

 

 

 

 

 

 

 

 

Table A2.

(con't.).

Microcosm.2:6

1h8

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A E H A
g g: .3 '5 i
c: 0 :0 F403 0).: ~«
,3. T. 3 2. ... A .22? 8:? fig. SE 8
S 9 >3 cf“ :1 :1 575 I015 :31. Iu~u '8
° ‘9 :3 g: a) E? §?J§ 5§J§ 1%.: Egs: '3
.n a a .3 ~a 1. <3 8 :2 0 a a
~a u -H a u c u) u my 23
.. v 2 55 53.333.32.08
.9 u
.5 5' g a: a“ a" =9 88:88“”*’%
E4 54 a. z: z: z: 24:: h.C> 2‘5! 5:53 ‘5
0 27.0 h.30h 8.750 890 .018 .00 0.00 0.10 -- .00 .887
68.5 27.0 - 8.775 .810 .0h5 <302 -- 0.28 -- .00
71321.5 26. - 9 075 .525 .019 <.o3 -- 0.67 -- .05
. 27.0 -- 9 360 007 .000 <.02 -- 0.71 -- .01
27.0 -- 9.530 -- - -- —- - . —- --
26.5 -- 9.665 .013*§01. .02 -- 0.57 -- .03
26.0 -- 9.7h0 .00h .000 .Oh -- 0.80 -- .06
26.5 -- 9.730 .010 .000 .00 9.85 0.79 0.93 .08
26.0 -- 9.7733 -- -- -- 11.9 -- -- --
26.5 -- 9.290 .000 .000 .01 9.70 1.06 0.83 .21
26.0 -— 9.320 000 .000 01 6.60 0.66 0.99 .0h
26.5 - 9.360 -- -- -- -- -- -- --
26.5 -- 9.325 .000 .000 .02 13.3 0.69 0.90 .08
26.0 -- 9.360 -- —- -- -- -— -- -
26.5 -- 9.355 .000 <.01 .01 11.9 0.66 0.90 .13
27.0 -- 9.385 <.01 <.01 <.01 12.2 1.16 0.95 .06
26.0 -- 9.350 -- -- -- -- -- -- -
26.5 -- 9.310 .003 .000 .00 -- 0.96 1.02 .17
26.0 -- 9.210 .000 .000 -- 1.12 1.08 .06

 

 

Table A2.

(con't.).

M1crocosm.2:7

1h9

 

 

 

 

1 E . .. ,1
1 E 8° g ... 3,. a
(J 0 ~’ to F101 ?:-1 g?
V a z 0" QA \ \ V
’3 o " '01: 2: 7:0: 323:...
g a >. c: - - 5 u) m u! r4~v u~a 9
o .9 3: :2 3’ 3, Eng 2313 3L:;:2 : :g
in d :3 Ca v V o d g a, d 0
" “ '” a H c u) a um z
u '3 z: z: 2: :3 o o o .4 o m o ,
g a. 'm 'N 'm +3.3 fig 3:3 fl}; Om
a a: 5 =3. 2 g a .90 :0 .92 :2 a
o 27.0 h.3ou 8.705 1.h5 020 .00 0.00 0.00 -- .00 1.05
68.5 27.0 -- 80735 10"? 00,46 <.02 -" 0032 .- .00
13h.5 26.5 -- 9.015 1.12 05h 5.03 - 0.77 -- .00
206.8 2700 -- 90285 0580 .050 <.02 -- 0.75 -- .01}
236.5 27.0 -- 9.1180 -- -- .. -- -- .. --
2717.07 26. 5 -- 9.690 .009 < .01 .02 -- 0 .72 -- .05
31.7.2 26.0 -- 9.890 .001. .000 .oh -- 0.75 -- .oh
1.19.0 26.5 -— 9.9170 .000 .000 .00 12.1 0.58 1.3h .07
1769.0 26.0 -- 10.000 -- -- -- 16.1 -.. -- ..
1.91.0 26.5 -- 10.050 .000 .000 .03 16.1 1.00 1.11 .09
552,5 25,0 -- 10.070 .000 .000 .02 11.0 0.6h 1.61 .13
617.0 26.5 - 10-090 -- - -- -- -- -- --
532,0 25,5 -- 10.050 .000 .ooh 01 16.5 0.69 1.h8 .06
, 25,0 -- 10.050 - -- .. -- .. .. --
7823 26.5 -- 10.055 .000<.o1 .00 15.7 0.1h 1.52 .10
830:0 27,0 -- 10.080 <.01 <.01 < 01 16 2 0.90 1.52 .13
a , 25,0 -- 10.020 -- ..- -- .. -- .. ..
997:3 26.5 -- 9.975 .007 .030 .02 -- 0.97 1.52 .11.
160.0 26.0 - 9.880 .000 .000 .00 -- 1.12 1.57 .00

 

 

 

 

 

 

 

 

 

 

 

 

 

Table A2.

(con't.).

Microcosm.2:8

150

 

 

 

A E H A
Q a: .3 "g i
23 3' v 5 1 .... E’
w H Cl 0 .3 v
g g 5. o \ i 5?.) 0?: 3V .... .8
° 4’ 7‘ 2: a, a) ELE':3~5 u a :3 c E
:3 g .5 .a 1, 1, <3 a §?¢u a o
I: 34 c: 00 in to z
o '3 a: a: if .3 o o o .4 o m 0 Im
a E. «1 n7 «3 .p'0 35'0 .3.3 :5.3 o
«4 on a x O o :1: O 3 .... 3 o -1 «4 or! z
a E4 n. z z 2 so 72.0 E-az In: N
o 27.0 h.30h 8.750 1.h6 .020 .00 0.20 0.27 -- .00 1.05
68.5 27.0 -- 8.780 1.175 .0176 <.02 -- 0.30 -- .00
131.5 26.5 -- 9.025 1.12 .05h <.o3 -- 0.52 -- .02
206.8 27.0 -- 9.325 .sho .050 <.02 -- 0.72 -- .01»
236.5 27.0 -- 9.515 -- -- -- -- -- -- --
27h.o 26.5 -- 9.715 .000 <.01 .02 -- 0.60 -- .08
3147.2 26.0 - 9.890 .0017 .000 .01. -- 0.75 -- .oh
1719.0 26.5 - 9.965 .000 .030 .00 16.7 o 60 1.17 .07
1.69.0 26.0 - 10.050 -- -- -- 16.2 -- -- --
1791.0 26.5 -— 10.080 .000 .000 <.01 -- 0.77 1.23 .07
562.5 26.0 - 10.0850 .000 .000 .00 13.8 0.77 1.58 .11
17.0 26.5 -- 9.020 -- -- - -- -- - --
632.0 26.5 -- 9.035 .000 .006 .01 17.5 0.76 1.61 .10
710.5 26.0 -- 9.105 - -- -- -- -- -- --
732,0 25,5 -- 9.155 .000 <.01 .00 1h.8 0.99 1.140 .13
30,0 27.0 -- 9.205 <.01 <.01 <.01 18.h 1.23 1.61 .15
99.0 26.0 -- 9.230 -- -- -- -- -- -- ~-
72.0 26.5 -- 9.200 .000 .011. .01 -- 1.178 1.62 .11
1617.0 26.0 -- 9.130 .000 .000 .00 - 1.13 1.5h .15

 

 

 

 

 

 

 

 

 

 

 

 

 

Table A2.

(con't.).

Microcosm.2:9

lSl

 

 

 

1 3 .. 1
3 2° 3 3 a
U 0 an H a! 0 oz
" V 5 z: A a .22: 8:? is, as, j;
a) I! N d at \ \ d
g g g 9. 3 3. a: 3 1° 5.?“ 2v 8
5 g g .a v v 3: 13" 25 ’35 E
8. .2 z: z: a: .3 o 3 8 .4 3’ 3 5’ 5f
3 a Im IN Im ‘p.o ;§.o a u .p A a,
-H 0 53 :n <3 <3 :3 o 8 .H 8 '313 1:13 o
64 £4 a. z: z: 2: 64C) &.c> 512: &.2: £3
0 27.0 0.300 8.750 2.87 .020 .00 .00 0.55 - .00 2.85
68.5 27.0 -- 8.785 2.87 .050 <.02 -- 0.30 -- .00
13h. 26.5 - 9.050 2.514 058 <.o3 -- 0.57 -- .00
206.8 27.0 -- 9.31.5 2.07 .062 <.02 -- 0.53 -- .02
236.5 27.0 -- 9.520 -- -- --. -- -- -- --
27h.0 26.5 -- 9.720 1.21. .061 .02 -- 0.52 -- .01.
3177.2 26.0 -- 9.990 .1730 .071 03 -- 0.67 -- .11
h19.o 26.5 - 10.200 .020 .030 .00 19.2 0.60 2.57 .12
1.69.0 26.0 - 10.1710 - -- -- 20.0 -- -- --
h91.0 26.5 -- 10.560 .010 .000 <.01 -- 1.08 2.78 .13
562.5 26.0 -- 10.710 .010 .000 .02 25.8 1.21 3.12 .19
517.0 26.5 - 10.790 -- -- -- -- -- -- -
632.0 26.5 -- 10.780 .011 011 .02 23.6 1.18 -- .22
710.5 26.0 - 10.800 -- -- -- -- -- -- -
732.0 26.5 - 10.835 .01h .015 .01 17.1 1.77 2.75 .30
30.0 27.0 -- 10.830 .021» .015 <.01 25.2 2.76 3.02 .31
99 .0 26.0 -- 10.765 -- -- -- - -- -- --
72.0 26.5 -- 10.690 .020 .0317 .01 -- 2.57 3.07 .31.
16h.o 26.0 -- 10.510 .020 .000 .00 -- 2.70 3.10 .31

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table A2.

(con't.).

Microcosm 2:10

152

 

 

 

 

 

3 Q o H A
1 3 .5”? "5:“ 1 3.. 5
3 g z m Hal 0.: g
2 2 .5 9‘“: 550.633.1538
o .. .. a 2 2.5 53 .. .. 5
:5 a .5 .. v v o a 25 as 3
.. 2. 2 5. 5. 5. 2:25 §§ 249'? 3 3" 7
5 5 3 5 a" 65 =5 30.21: 55 533 om
5‘ 5* 9* 2: 2: z: 57L) &.L> €35: E:E§ E3
0 27.0 0.300 8.750 2.85 .020 .00 0.00 0.00 - .00 2.85
68.5 27.0 -- 8.775 2.86 .007 <.02 -- 0.30 -- .00
130. 26.5 -- 9.025 2.57 .056 <.o3 - 0.83 -- .00
206.8 27.0 - 9.300 2.08 .060 <.02 -- 0.57 .- .00
236.5 27.0 - 9.080 -— —- -— .. 2- -- ..
270.0 26.5 -- 9.670 1.35 .061 .01 -- 0.50 -- .00
307.2 26.0 -- 9.965 .580 .080 .02 -- 0.52 -- .08
019.0 26.5 -- 10.135 .050 .000 .00 15.1 0.59 2.00 .13
h69JD 26.0 -- 10 380 -- -- -- 20,0 -_ -_ -_
091.0 26.5 - 10 510 .020 .000 <.01 22.0 0.87 2.56 .12
562.5 26.0 - 10 690 .020 .000 .02 27.5 9.50 3.07 .21
617 .0 26.5 -- 10 780 - -- -- -- -- .- ..
632.0 26.5 - 10 775 .017 .012 .02 20.8 1.26 -- .19
710.5 26.0 -- 10.830 -- -- -- -- -— .. ..
732.0 26.5 -- 10.865 .015 .015 .01 -- 1.02 2.57 .20
30.0 27.0 -- 10.560 .020 .010 .01 20.6 2.00 -- .20
99 .o 26.0 -- 9.035 -- -- -.. -- .. .. __
72.0 26.5 -- 9.000 .000 .010 .01 -- 1.88 3.22 .21
160.0 26.0 - 9.010 .000 .000 .00 -- 0.60 3.32 .07

 

 

 

 

 

 

 

 

 

 

 

 

Table A2.

(con't.).

Microcosm 2:11

153

 

 

 

 

 

 

 

 

 

 

 

 

 

A E u g E

S 5’ E .. 2

A V A HA v
<9 0 to .4cu 07a

V E z 0" “A ‘5\ “\7 'U

A V l A A ”4.3 0.; g G)

5‘” >. o~::§\mg§.§5

3 3 *’ 5 2 2 23° 5.5 . .. 5‘

.2 d '3 .a .3 .3 <3 '3 52’5 '3 5 z:

" h 'H a n c u: n u) l

a E» '3 ‘7‘... ‘7‘... 7.. Egfig 5:35.53 9;“
a: 2 5 5. 2 2 .5. 26 2:6 22 22

o 27.0 .300 8.7501.02 -- .00 - -- -- .00 1.05
68.5 27.0 - 8.7251.02 .006 <.02 -- 0.37 -- .00
130.5 26. -- 8.8851.33 .052 <.03 -- 0.00 -- .10
206.8 27.0 -- 9.015 1.12 .050 <.02 -- 0.50 -- .03
236.5 27.0 -- 9.195 -- -- -- .- -— -- --
270.0 26.5 -- 9.355 .512 .000 .00 ..- 0.00 - .00
307.2 26.0 -- 9.610 .012 .000 .02 -- 0.55 -- .00
019.0 26.5 —- 9.755 .010 .000 .00 13.0 0.02 1.03 .09
069.0 26.0 a 9.900 -- -- -- -- -.- -- --
091.0 26.5 -- 9.980 .000 .000 <.01 8.50 1.10 1.33 .07
562.5 26.0 -- 10.055 .000 .000 .00 7.15 0.95 1.07 .26
17.0 26.5 -- 10.100 -- -- -- -- -- .- --
632. 26.5 - 10.080 .000 .000 .02 7.80 0.67 1.39 .08
710.5 26.0 ' -- 10.100 -- .- .. .. .. .. ..
732.0 26.5 -- 10.105 .006 .013 .01 13.0 0.80 -- --
30.0 27.0 -- 1o.105<.01 <.01 <.01 15.3 1.00 1.51 .12
99 .o 26.0 -- 10.110: -- -- —- -- -- -- --
72.0 26.5 - 10.070 .000 .030 .01 -- 1.05 1.56 .12
67.0 26.0 -- 10.010<.01 .000 00 -- 0.77 1.58 .09

 

 

 

APPENDIX B

AMMONIA VOLATILIZATION - RAW DATA

150

US

Ammonia volatilization data from experimental laboratory

vessels.

Table Bl.

 

Flux
(Loss Rate)

(mmole/cmz/hr)

Average Total
Ammonia Concentration

Temperature (C )

pH

(mmoles/m2)

 

TRIAL A

923337956395935290197131
82066 11“.. 730896 502.“.05914071

66229914314000 1414899308081
5251499667790 99672731016

72721516138571393362141142

00000000000000000000.0000
222222222222222222222222
222222222222222222222222

OOOOOOOOOOOOOOOOOOOOOOOO.

11111111 1 l

TRIAL B

162207220
307302.414‘4 2614

81488883991460
279336 9506 38

112718137512

000000000000

000000000000
222222222222

000005555000
899999999000
1.1.1.

(continued . .

155

ry

-

Ammonia volatilization data from experimental laborato

Table Bl .

vessels.

 

Flux
(Loss Rate)

(mmole/cmg/hr)

Average Total
Ammonia Concentration

Temperature (C)

pH

(mmoles/m2)

 

TRIAL A

37956395935290197131
661147308965021405914071

66229914314000 1414899308081
5251499667790 99672731016

72721516138571393362hlh2

00000000000000000000.0000

222222222222222222222222
222222222222222222222222

OOOOOOOOOOOOOOOOOOOOOOOO

/o/onlqlnuguRVo/olo/o/o/nvnvnvnvnvnv1.1.nu1.na1.
1.1.1.1.1.1.1.1. 1. 1.

TRIAL B

361162207220
30 7130 2114141426114

1&111496125612

8148888 399h60
279336950638

112718137512

000000000000
000000000000
222222222222

000005555000
899999999000
1.1.1.

(continued . .

Flux
(Loss Rate)

(mmole/cmz/hr)

(mmoles/m2)

Average Total
Ammonia Concentration

156

TRIAL C

(con't.).
Temperature (C )

 

 

Table Bl.
pH

558659997663010269293212298509523
552238881105376h216llh922h1721397

OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO

271298h1288h lid. 995h116538h6 16218‘43

. 20h085381hh92961713\4h850066h35
\u.1‘421502282h5352509h9252.“..014553085

ail/09.118.431.1821118h1187h17185211.5/‘4

00000000000000000OOOOOOOOOOOOOOOO

0000000000000OOOOOOOOOOOOOOOOOOOO
222222222222222222222222222222222

O0000000OOOOOOSSSSSSSSSSSOOOOOO00

888889999999999999999999900000000.
llllllll

(continued . . .

157

(con't.).

Table Bl .

 

Am
a
m.n
1i
3
Fm
n
o
”u
a
an
.t n
o e
mi c
n
e o
zzu
a
r a
mm
.A.m
mw
m
m
r
m
m
e
mi
"n
p

(mmole/cmz/hr)

(mmoles/m1)

 

01659 14.4058318
Dana/...“ 0.4630630

h73l63219h51h6

9399881918h26h
093182.“..2689‘48‘4

35196u218h3252

00000000000000

00000000000000
22222222222222

00000000000000

mmmnunnuunnunu

 

157

(con't.).

Table Bl.

 

Flux
(Loss Rate)
(mmole/ch/hr)

Average Total
Ammonia Concentration

(mmoles/m2)

Temperature (C)

pH

 

01659 Mano—38318
Danna—‘4 0‘4630630

\473163219h51h6

9399881918u26‘4
093182.4268914814

35196x4218h3252

00000000000000

00000000000000
22222222222222

00000000000000
0..... ........

mmmununuunnuuu