Title Page
 Table of Contents
 List of Tables
 List of Figures
 Mass balance models for nitrog...
 Significance of nitrogen mineralization...
 Estimation of denitrification in...
 A nitrogen mass balance for peninsular...
 Veracity and utility of nitrogen...
 Summary and conclusions
 Biographical sketch

Title: Nitrogen mass balances in Florida ecosystems
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00090219/00001
 Material Information
Title: Nitrogen mass balances in Florida ecosystems
Series Title: Nitrogen mass balances in Florida ecosystems
Physical Description: Book
Creator: Messer, Jay James,
 Record Information
Bibliographic ID: UF00090219
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000081137
oclc - 05086256

Table of Contents
    Title Page
        Page i
        Page ii
        Page iii
        Page iv
        Page v
    Table of Contents
        Page vi
        Page vii
        Page viii
    List of Tables
        Page ix
        Page x
        Page xi
        Page xii
    List of Figures
        Page xiii
        Page xiv
        Page xv
        Page xvi
        Page xvii
        Page xviii
        Page xix
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
    Mass balance models for nitrogen
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
    Significance of nitrogen mineralization and denitrification in a sugar plantation
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
        Page 152
        Page 153
        Page 154
        Page 155
        Page 156
        Page 157
        Page 158
        Page 159
        Page 160
        Page 161
    Estimation of denitrification in Lake Okeechobee using nitrogen mass balance calculations
        Page 162
        Page 163
        Page 164
        Page 165
        Page 166
        Page 167
        Page 168
        Page 169
        Page 170
        Page 171
        Page 172
        Page 173
        Page 174
        Page 175
        Page 176
        Page 177
        Page 178
        Page 179
        Page 180
        Page 181
        Page 182
        Page 183
        Page 184
        Page 185
        Page 186
        Page 187
        Page 188
        Page 189
        Page 190
        Page 191
        Page 192
        Page 193
        Page 194
        Page 195
        Page 196
        Page 197
        Page 198
        Page 199
        Page 200
        Page 201
        Page 202
        Page 203
        Page 204
        Page 205
        Page 206
        Page 207
        Page 208
        Page 209
        Page 210
        Page 211
        Page 212
        Page 213
        Page 214
        Page 215
        Page 216
        Page 217
        Page 218
        Page 219
        Page 220
        Page 221
        Page 222
    A nitrogen mass balance for peninsular Florida
        Page 223
        Page 224
        Page 225
        Page 226
        Page 227
        Page 228
        Page 229
        Page 230
        Page 231
        Page 232
        Page 233
        Page 234
        Page 235
        Page 236
        Page 237
        Page 238
        Page 239
        Page 240
        Page 241
        Page 242
        Page 243
        Page 244
        Page 245
        Page 246
        Page 247
        Page 248
        Page 249
        Page 250
        Page 251
        Page 252
        Page 253
        Page 254
        Page 255
        Page 256
        Page 257
        Page 258
        Page 259
        Page 260
        Page 261
        Page 262
        Page 263
        Page 264
        Page 265
        Page 266
        Page 267
        Page 268
        Page 269
        Page 270
        Page 271
        Page 272
        Page 273
        Page 274
        Page 275
        Page 276
        Page 277
        Page 278
        Page 279
        Page 280
        Page 281
        Page 282
        Page 283
        Page 284
        Page 285
        Page 286
        Page 287
        Page 288
        Page 289
        Page 290
        Page 291
        Page 292
        Page 293
        Page 294
        Page 295
        Page 296
        Page 297
        Page 298
        Page 299
        Page 300
        Page 301
        Page 302
        Page 303
        Page 304
        Page 305
        Page 306
        Page 307
        Page 308
        Page 309
        Page 310
        Page 311
    Veracity and utility of nitrogen mass balance models
        Page 312
        Page 313
        Page 314
        Page 315
        Page 316
        Page 317
        Page 318
        Page 319
        Page 320
        Page 321
        Page 322
        Page 323
        Page 324
        Page 325
        Page 326
        Page 327
        Page 328
        Page 329
        Page 330
        Page 331
        Page 332
        Page 333
        Page 334
        Page 335
        Page 336
        Page 337
        Page 338
        Page 339
        Page 340
        Page 341
        Page 342
        Page 343
        Page 344
        Page 345
        Page 346
        Page 347
        Page 348
        Page 349
        Page 350
        Page 351
        Page 352
        Page 353
        Page 354
        Page 355
        Page 356
        Page 357
        Page 358
        Page 359
        Page 360
    Summary and conclusions
        Page 361
        Page 362
        Page 363
        Page 364
        Page 365
        Page 366
        Page 367
        Page 368
        Page 369
        Page 370
        Page 371
        Page 372
        Page 373
        Page 374
        Page 375
        Page 376
        Page 377
        Page 378
        Page 379
        Page 380
        Page 381
        Page 382
        Page 383
        Page 384
        Page 385
        Page 386
        Page 387
        Page 388
        Page 389
        Page 390
        Page 391
        Page 392
        Page 393
        Page 394
        Page 395
        Page 396
        Page 397
        Page 398
        Page 399
        Page 400
        Page 401
        Page 402
        Page 403
        Page 404
        Page 405
        Page 406
        Page 407
        Page 408
        Page 409
        Page 410
        Page 411
        Page 412
        Page 413
        Page 414
        Page 415
    Biographical sketch
        Page 416
        Page 417
        Page 418
Full Text







Dedicated to


Terrence O. Weitzel
Jesse S. Robertson
Patrick L. Brezonik

exemplars and friends


It is impossible to acknowledge all of the scientists,

technicians, field helpers, and government officials who

offered their time, expertise, and advice, without whom

this project would have been almost impossible. Thus I

will not attempt to acknowledge the scores of people to

whom thanks should go for protracted telephone conversations

and personal interviews with an overwrought and insistent

graduate student. However, I would like to single out

several people for special thanks.

Professor Patrick L. Brezonik, my advisor, spent many

evening and weekend hours keeping me honest and improving

my prose, often only with the aid of an almost undecipher-

able manuscript. He also has supported me faithfully, both

financially and professionally, for over four years, for

which I am sincerely grateful. My remaining committee mem-

bers, Professors Wayne C. Huber, Gabriel Bitton, and

Edward S. Deevey, Jr., deserve thanks for their intelligent

questions, welcome advice, and for enduring my interminable

verbal and written presentations.

Many of my associates in the department have kindly

allowed me to use their unpublished data for construction

of the mass balance models. Dr. Earl Shannon of Black,


Crow, and Eidsness deserves particular mention for his

permission to use unpublished data on nitrogen inputs and

outputs from the sugarcane field. Fred Davis, of the

South Florida Water Management District, furnished the

most recent (unpublished) nutrient balance data available

for Lake Okeechobee, for which I am also grateful. Thanks

for other data go to Charles Hendry and Eric Edgerton for

their precipitation data; Forrest Dierberg for data on

cypress wetlands; and Dean Marbury for data on Lake

Okeechobee macrophyte composition.

Special thanks go to Bruce Snyder, who has collaborated

with me on the Lake Okeechobee field work and worked tire-

lessly in the laboratory on both his work and my own. I

am grateful for permission to use his unpublished nitrogen

fixation estimates for Lake Okeechobee. Also assisting

with field work were Vernon Meyers, Roger Conley, Donald

Thompson, and Patrick Brezonik.

Barbara Smerage was responsible for the very profes-

sional production job on the dissertation. Her eleventh

hour efforts are greatly appreciated. Deborah Tuschall

spent long hours editing copy and checking references.

Adelle Koehler typed the reference section with speed and


Finally, I would particularly like to thank my wife,

Janice, for spending many sleepless nights with our infant

son, John, so that I could work refreshed on.the following


Work on this project was supported by a contract from a

Black, Crow, and Eidsness CH2M Hill (Chapter 4), the

South Florida Water Management District (Chapter 5), and

the National Research Council (Chapters 2, 6, and 7).






LIST OF TABLES . . . . .


ABSTRACT . . . . . .


. . ix



Nitrogen as a Pollutant . . . 3
The Biogeochemical Cycle of
Nitrogen .. . . . ... 5
Environmental Modeling . . . 5


Historical Development of the
Nitrogen Mass Balance Model . 14
Local Terrestrial Systems . . .. 16
Lakes . . . . . . . . 19
Watershed Models. . . . . . 25
Global Mass Balances for Nitrogen . 38
Summary . . . . . . . 42


Sampling and Field Methods. . .
Analytical. . . . . . .
Microbiology. . . . . .
Mineralization Experiments. . .
Nitrate Disappearance Experiments
Denitrification by Acetylene
Blockage . . . . .
Sediment Cores .... . . .
Sediment Traps. . . . . .

S 44
S 46
S 46
S 47

S 48
S 49


PLANTATION. . . . . . ... 52

Site Description. . . . . . 55
Nitrogen Transformations in
Histosols .......... . 58
General Characteristics of the
Sugarcane Plantation Histosols. 72
Ammonification . . . . ... 83
Nitrification . . . . .. .94
Denitrification . . . . ... 105
Summary . . . . . . ... 160


Denitrification in Lake Waters and
Sediments . . . . . .. 164
Limnology of Lake Okeechobee ... 171
Denitrification by Mass Balance . 175
Denitrification in Lake Okeechobee
Sediments . . . . . .. 205
Summary . . . . . . .. 221

FLORIDA . . . . . . . 223

Selection of System Boundaries. . 224
Hydrologic Submodel . . . . 229
Agricultural Submodel . . ... 259
Atmospheric Submodel. . . . .. 283
The Peninsula Mass Balance Model. . 304

BALANCE MODELS. . . . . .. 312

Data Quality. . . . . . ... 313
Nitrogen Mass Balance Models for
Florida Systems. . . . .. 332
A Summary Statement on the Utility of
Nitrogen Mass Balance Models. . 358




FIELD .................. .366


REFERENCES ................. .388

BIOGRAPHICAL SKETCH . . . . ... .416



Table Page

2-1 Summary of nitrogen balances on several
lakes. . . . . . . . . 26

2-2 Estimated nitrogen balances for some
bioclimatic types. . . . . ... 29

2-3 Soil nitrogen balance for the United
States . . . . . . ... 37

2-4 Estimates of nitrogen inputs and returns
to the total land area of the United
States, 1970 . . . . . . 39

4-1 Mean concentration of nitrogen forms, total
organic carbon, and physical composi-
tion of soil samples for the entire
study period . . . . . . 73

4-2 Bulk density of sugarcane plantation
soils. . . . . . . . ... 75

4-3 Total soil carbon and Kjeldahl nitrogen
in samples pooled from five stations
on 9-28-77 . . . . . . 86

4-4 Anaerobic production of ammonium ion in
shaken flasks at 230C with and
without glucose. . . . . .. 92

4-5 Relative nitrifier activity in the soil
column and in drainage system water
and sediments. . . . . . . 96

4-6 Logarithm of the most probable number (MPN)
of denitrifying bacteria in the soil
column . . . . . . . . 109

4-7 Percent nitrate remaining in water over-
lying flooded or totally anaerobic
soils after 26 days. . . . . ... 121

4-8 Reduction of nitrate and nitrite under
anaerobic conditions at 30C ... . 123

Table Page

4-9 Reduction of nitrite and nitrate from
anaerobic rhizosphere soil with and
without glucose amendment . . ... .126

4-10 The effect of glucose on the reduction of
nitrite and nitrate in soils from the
saturated zone. . . . . . ... 128

4-11 Logarithm of the most probable number of
denitrifiers in ditch water and
sediments, and in water from wells
piercing the limestone. . . . ... 140

4-12 Reduction of nitrate and nitrite in ditch
water under anaerobic conditions . 142

4-13 Concentration of nitrogen forms in
hydrologic fluxes across sugarcane
field boundaries. . . . . ... 147

4-14 A soil column balance sheet for the sugar-
cane plantation . . . . . . 148

4-15 Mineralization and nitrification rate
calculations. . . . . . ... 149

5-1 Representative rates of denitrification
in various limnetic systems . . .. .172

5-2 Summary of water quality in L. Okeechobee. 176

5-3 Nitrogen and phosphorus balances for
Lake Okeechobee . . . . . . 179

5-4 Total phosphorus and total Kjeldahl nitro-
gen concentrations in sapropel cores
from the center of Lake Okeechobee on
June 1, 1977. . . . . . ... 181

5-5 Total phosphorus and total Kjeldahl
nitrogen concentrations in core L08
taken on February 4, 1978 . . ... .184

5-6 Mass balance estimates of denitrification
in Lake Okeechobee, Florida . . .. .192

5-7 Calculation of denitrification in sediments
using aFickian diffusion equation.- . 206

Table Page

6-1 Rainfall recorded at selected stations . 232

6-2 Nitrogen concentrations in Florida
precipitation . . . . . ... 234

6-3 Nitrogen loadings from precipitation to
hydrologic units in the Florida
peninsula . . . . . . ... 237

6-4 Gauged drainage area, mean annual dis-
charge for the period of record and
calendar years 1970 through 1975, and
quality of record for gauging stations
in the U.S.G.S. hydrologic subregions
of the Florida peninsula. . . .. .242

6-5 Fraction of area represented by drainage
area and fraction of precipitation
discharged during 1970-1975 . . .. .244

6-6 Nitrogen concentrations in Florida streams
from U.S.G.S. and stored data
collected from 1970-1975. . . .. .247

6-7 Nitrogen concentrations in Florida streams
calculated by Slack and Goolsby
(1976). . . . . . . . . 250

6-8 Nitrogen exported in surface water
discharges during 1970-1975 . . . 253

6-9 Hydrologic submodel for the Florida
peninsula . . . . . . ... 258

6-10 Mean annual citrus production in Florida
during 1969-1970 through 1974-1975
crop years. . . . . . . ... 267

6-11 Vegetable production and outshipments
during 1970-71 through 1975-76
seasons . . . . . . ... 270

6-12 Mean annual field crop production during
1970-1975 . . . . . . . 272

6-13 Forest products retention, shipments, and
production for the peninsular system. 275

6-14 Average nitrogen fertilizer sales in
Florida during fiscal years 1969-1975 279

Table Page

6-15 Data for calculating nitrogen fixation .. 286

6-16 Production of nitrogen oxides by
combustion of fossil fuels in
peninsular Florida... . . . .. 297

6-17 A mass balance sheet for nitrogen in the
Florida peninsula . . . . ... 305

7-1 Quality of data used in construction of
nitrogen mass balance models for
Florida systems . . . . ... 315

7-2 Percent contribution and removal by sources
and sinks for nitrogen in three Florida
systems . . . . . . ... 325



Figure Page

2-1 Nitrogen pools and'fluxes within (A), the
Upper Santa Ana River Basin, and
(B), the Lower San Joaquin River
Basin, in California. . . . . ... 33

2-2 The flow of nitrogen in Wisconsin
agriculture in 1974 . . .. . .. 35

4-1 Sugarcane plantation soils map. . . .. 56

4-2 Map of the sugarcane field showing the
drainage and irrigation system,
sampling wells, and sampling points
for hydrologic data . . . . ... 57

4-3 Mean concentrations of nitrogen species and
fraction of water over all depth in
soil over the study period. . . . 78

4-4 Mean distribution of NO2+NO3-N, soluble
NH4-N, total NH4-N, TON, and TOC in
all cores with depth on two dates . .. 79

4-5 Mean moisture content of all soil samples
taken during the study versus depth in
the sugarcane plantation. . . ... 81

4-6 The relationship between soluble ammonium
ion and fraction water in field samples 84

4-7 Arrhenius plot of mineralization rate vs.
reciprocal Kelvin temperature in
experiments on surface muck from
sugarcane field . . . . . ... 88

4-8 Production of NH4-N over 62 days at 370C
in leaching tube experiments on surface
muck from sugarcane field . . ... 89

4-9 Typical distribution of oxidized and re-
duced inorganic N species, and auto-
trophic nitrifiers in a soil profile. . 99


Figure Page

4-10 Microstratification of NO2+NO3 N and
NH+-N concentrations in the surface
layer of muck . . . . . . .. 101

4-11 Inorganic nitrogen production and percent
inorganic nitrogen nitrified in sugar-
cane soils incubated in leaching
tubes . . . . . . . . 103

4-12 EH profiles at site A on September 28,
1976, and at sites A and B (Figure
4-1) on May 5, 1977 . . . . ... 107

4-13 Nitrate concentration profiles at stations
A through J (see Figure 4-1) on
February 6, 1977. . . . . . ... 113

4-14 Nitrate concentration profiles at stations
A through H (see Figure 4-1) on
September 28, 1977. . . . . ... 114

4-15 (Upper) Comparison of percent nitrate
removal after 27 days of incubation
with denitrifier population in soils
from various depth in a soil core.
(Lower) Changes in nitrate concentra-
tions versus time in the same soil
samples . . . . . . . ... 118

4-16 Increase in N20 concentration versus time
in the head space of a laboratory soil
column in which reduction of N20 to N2
was blocked by acetylene. . . . ... 130

4-17 N20-N pumped from headspace of a laboratory
soil column followed a "rain event" . 132

4-18 N20 concentration in soil column air. ... .136

4-19 Profiles of various nitrogen species in
drainage ditch sediments. . . . . 138

4-20 Reduction of N03 in water overlying ditch
sediment under anaerobic conditions . 144

5-1 Bathymetric map of Lake Okeechobee showing
bottom composition. . . . . ... 174

5-2 Temperature profiles and sedimentation rates
determined using sediment traps over one-
and two-day periods . . . . .. .186


Figure Page

5-3 Variations in percent water of sediments
with depth in several cores from the
depositional zone in the center of
Lake Okeechobee . . . . . .. 187

5-4 Nitrate concentration profiles in inter-
stitial water of three sediment cores:
LO10; LOll; and L012. . . . . . 201

5-5 Nitrate, nitrite, and exchangeable ammonium
concentrations in the interstitial
water of core LO8 . . . . ... 203

5-6 Concentrations of nitrite, nitrate, total
(exchangeable) and soluble ammonium,
and soluble organic nitrogen in core
LO015 . . . . . . . . ... 204

5-7 Most probable number of denitrifying
bacteria in the top 10 cm of core LO10. 208

5-8 Most probable number of different microbial
populations in core LO8 . . . .. .209

5-9 Most probable number of microbial popula-
tions in core L015. . . . . ... 211

5-10 Changes in NON and N03-N in time in the
supernatant water of a stirred,
anaerobic culture-filled flask with
ditch sediment and spiked with NH4NO3 . 214

5-11 Nitrifying activity in core L08 expressed
as disintegrations per hour per gram
of sediment of 14C02 taken up by auto-
trophs in excess of background
anapleutrotic 14CO2 uptake. . . .. .215

5-12 Denitrification rate as measured by the
acetylene blockage technique in core L08. 217

5-13 Variation in sediment N:P ratio, N03-N
concentration (pg N/mL) in inter-
stitial water, and MPN of potential
denitrifiers with depth (log MPL/mL). . 220

6-1 Sources and sinks for nitrogen in
peninsular Florida. . . . . ... 225

6-2 Hydrologic accounting units in the Florida
peninsula (modified from Conover and
Leach 1975) . . . . . . ... 226

Figure Page

6-3 Meterological stations used to estimate
precipitation in the peninsula ... . 231

6-4 Rivers and gauging stations in the Florida
peninsula. . . . . . . ... 239

6-5 Flows of nitrogen in the livestock, fish
and dairy industries in peninsular
Florida. . . . . . . . ... 260

6-6 Details of compartment construction for
the livestock minimodel. . . . ... 262

6-7 The agricultural submodel. . . . ... 281

6-8 Atmospheric nitrogen fluxes in the Florida
peninsula. . . . . . . ... 302

6-9 A mass balance model for nitrogen in the
Florida peninsula. .. . . . .. 307


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



Jay J. Messer

December 1978

Chairman: Patrick L. Brezonik
Major Department: Environmental Engineering Sciences

Nitrogen mass balance models were investigated as

instruments for the assessment and management of nitrogen

as an environmental pollutant. Nitrogen balances were

constructed for three Florida ecosystems, and field and

laboratory data were collected to estimate or verify the

magnitude of nitrogen fluxes in the systems. A sugarcane

field in the Everglades Agricultural Area (EAA), a large

lake (L. Okeechobee), and the entire Florida peninsula

were chosen as prototype systems. The peninsula model

was constructed by combining the results of separate

hydrological, agricultural, and atmospheric submodels.

Mineralization of soil organic nitrogen stored in

Everglades peat accounted for the majority of nitrogen

input to the sugarcane field, and input from surface streams

draining the Kissimmee River Watershed and the EAA

supplied most of the nitrogen to the lake. Nitrogen inputs

to the Florida peninsula were dominated in the model by

large atmospheric fluxes of nitrogen delivered by gas


absorption and bulk precipitation on environmental sur-

faces. The nitrogen mass balance in all of the systems was

characterized by the failure of calculated outputs to

remove between 60 and 65 percent of the nitrogen inputs to

the system. Laboratory experiments and field data indicated

that denitrification was likely to account for most of the .

discrepancy, but nitrogen accretion in the peninsula model

could not be ruled out.

A semiquantitative comparison of overall quality of the

data used to construct the models indicated that reliability

decreased as the size of the prototype ecosystem increased.

Veracity was good in the sugarcane model, and laboratory

work and denitrification studies in the laboratory were in

good agreement with denitrification predicted by mass

balance calculations. Veracity was only fair in the lake

model and fair to poor for the peninsular balance.

Subsystem models of the peninsular mass balance model

were more reliable and suggested insights into the nitrogen

economy of the system. The hydrologic submodel indicated

that areal nitrogen export coefficients were similar in five

hydrological accounting units, despite differences in land

use and climate. The atmospheric submodel indicated that

fixation of nitrogen gases during fossil fuel combustion,

.together with ammonia volatilization, closely balanced the

return of fixed nitrogen to the system by precipitation and

gas deposition. A minimodel of the Florida citrus groves


indicated the potential for nitrate pollution of the

groundwater aquifer under grove sites.

The role of nitrogen mass balance models in environ-

mental monitoring and management was found to be dependent

on the size and heterogeneity of the prototype ecosystem.

Small system models and subsystems may have some predictive

value. At this time the primary benefits of large scale

nitrogen mass balance models are the direction of research

priorities suggested by preliminary model results and the

heuristic value gained by model'construction.



If ecosystems can truly be said to have strategies for

development (Odum 1969), one of the most remarkable of these

is their ability to assimilate assemblages of chemical ele-

ments into biomass against the disruptive force of entropy.

Death and the normal physiological maintenance of proto-

plasm dictates, however, that some of the complex molecules

which have been organized into biomass must be released to

the environment as soluble waste products or detritus.

These wastes are subsequently mineralized and transported

through frequently complex biogeochemical cycles until they

are reassimilated into biomass.

Biogeochemical cycles have historically been driven by

the diffuse solar energy flux acting through phototrophic

processes and the hydrologic cycle. Human technology has

accelerated these cycles by augmenting the diffuse solar

flux with the concentrated power of fossil fuels used in

agriculture, mining, and transportation. These anthropogenic

perturbations have increased environmental concentrations

or fluxes of many elements to the extent that they exert

either present or potential deleterious effects on the en-

vironment. These elements thus become pollutants, rather

than resources.

Frequently, these pollutants make their presence known

through wholesale destruction or alteration of habitats

(e.g. the New York Bight) or through sudden outbreaks of

mortality or morbidity (e.g. Minimata Bay, Japan). When

environmental degradation has reached such an extent, the

costs in terms of both irreparable damage and rehabilita-

tion are usually staggering. The ability to understand the

biogeochemical cycles of potential pollutants and to pre-

dict the effects of certain perturbations on the fluxes and

storage within the cycles has become essential to en-

vironmental management. Such predictive capacity would al-

low monitoring programs to be initiated that could spotlight

increasing fluxes or concentrations of pollutants before

they could do significant or irreparable damage. Predictive

capacity in science is often realized through the con-

struction of mathematical models. In simplified terms,

models are abstractions that replace the complexity of

nature with a simpler, more tractible structure (Rosenblueth

and Wiener 1945).

The purpose of this research is to examine the use of

a particular type of formal model, the mass-balance model,

to study nitrogen biogeochemistry in large systems. Nitro-

gen compounds have been implicated in both air and water

pollution, causing aesthetic and commercial damage to the

environment and in some cases endangering human health. The

biogeochemistry of nitrogen is complex, and quantitatively

not well understood. Therefore, a broad-brush modeling

technique seems to be indicated.

Nitrogen mass balances on three geological scales and

representing a variety of Florida ecosystems are investi-

gated. The systems include a sugarcane field in the Ever-

glades Agricultural Area; Lake Okeechobee in south-central

Florida; and the Florida peninsula. The validity of the

data upon which the various models are based is examined,

together with a consideration of the utility of the models

in environmental research and management. Finally, the

effect of system scale on the utility and validity of the

models is discussed, and some data needs revealed by the

construction of the three nitrogen mass balances are


Nitrogen as a Pollutant

Concern about the environmental impact of culturally

induced dislocations in the nitrogen cycle is evidenced by

the number of recent symposia on nitrogen as a pollutant

(SIDA 1972; IAEA 1973; IAWPR 1975; NRC 1978). Nitrite and

nitrate are responsible for adverse environmental effects

associated with food, water, and atmospheric quality.

Nitrate concentrations in excess of 10 mg N/L in drinking

water may cause methemoglobinemia, a disease which reduces

the oxygen carrying capacity of red cells in animals and

humans (NRC 1978). Although nitrate poses no direct health

problems to adult humans, the pH in the stomachs of infants

or in the rumen of cattle is sufficiently high to allow

nitrate reducing bacteria to produce nitrite which enters

the blood stream and oxidizes hemoglobin to methemoglobin,

a form that does not transport oxygen (Shuval and Gruener

1974). In persons eating meats preserved with nitrites,

e.g. bacon, or using drugs which contain secondary amines,

nitrite may interact with these drugs to form nitrosamines,

which may have carcinogenic or hepatoxic action at low con-

centration (Wintringham 1973; Mirvish 1975). Nitrite or

nitrate may also accumulate in food and crop plants ir-

rigated with water high in nitrate and be consumed by

livestock or humans in toxic amounts (Viets and Hageman

1971). Hydroxylamine, an intermediate in many of the

microbially mediated conversions of inorganic nitrogen, is

a potent mutagen (Alexander 1974); fortunately, it is un-

likely to accumulate in the environment. Nitrous oxide, a

product of denitrification, has been implicated in the

destruction of stratospheric ozone (Crutzen 1971). Finally,

nitrogen is responsible for the eutrophication of coastal

aquatic systems (Ryther and Dunstan 1971) and of lakes

situated in watersheds whose parent material is rich in

phosphorus. Nitrogen may limit algal growth in lakes in

the late stages of hypereutrophy (Vollenweider 1968). The

role of nitrogen as a pollutant has been thoroughly reviewed

in a recent report of the National Research Council (NRC


The Biogeochemical Cycle of Nitrogen

Although the qualitative machinations of the nitrogen

cycle have been understood for decades (cf. Lotka 1924), the

rates of the transformations of nitrogen species are ac-

curately known for only a few pathways under restricted

conditions. A generalized nitrogen cycle, cf. Brezonik

(1972), demonstrates the reason for this phenomenon.

Nitrogen is stored and transformed in solid and gas phases

as well as in solution, with oxidation numbers ranging from

plus 5 to minus 3. Many of the transformations occur at

phase interfaces, andmanyprocesses occur in soil and

sediment environments. Tight loops of mineralization, trans-

formation, and reassimilation in these systems make it

difficult to measure the rate of a single transformation

in situ. Most investigations yield information on net rates

of transformation or so alter the environment that

extrapolation to field conditions is tenuous. The trans-

formations of nitrogen in terrestrial and aquatic habitats

have been reviewed by Painter (1970), Brezonik (1972, 1977),

Keeney (1973), Burns and Hardy (1975), Alexander (1977),

Focht and Verstraete (1977), and by the National Research

Council (NRC 1978).

Environmental Modeling

Types of Models

Rosenblueth and Wiener (1945) and Hempel (1963) appear

to offer an instructive framework to categorize the various

models. Models may first be categorized as formal (mathe-

matical) or material (iconic). In order for a material

model to be useful, it must replace an unfamiliar phenomenon

with a familiar one and also enable experiments to be car-

ried out under more favorable conditions than is possible

in the prototype. The material model is thus of minimal

use in large scale biogeochemical modeling, since the model

becomes useful only as it approaches one to one correspon-

dence with the system (Rosenbleuth and Wiener 1945). While

an analog computer simulation constitutes an iconic model,

in practice this model shares a formal model in common

with the prototype (the real system). Since the formal

model could as easily be converted to a digital computer

algorithm, the analog model needs little further considera-

tion per se. This is not to deny its utility once a formal

model is constructed, however.

Formal models are divisible into inductive-probabilistic

models and deductive-nomological ones (Hempel 1963). Both

types of models account for a particular event by showing

that it resulted from general laws or theoretical principles.

Deductive-nomological models produce invariant output given

a set of inputs, but inductive-probabilistic models include

some relevant laws of a statistical nature. The construc-

tion of probabilistic models is generally undertaken by a

statistical analysis of the effect of one (or more) inde-

pendent variables on one (or more) dependent variables.

Such an analysis may confirm certain deterministic laws by

by induction within bounds of acceptable error.

Inductive-Probabilistic Models

The probabilistic approach has found considerable

utility in laboratory and small field plot experiments.

The results of probabilistic models frequently provide the

quantitative relationships upon which deterministic models

are based. Large scale probabilistic models have found con-

siderable utility in modeling various aspects of nitrogen

biogeochemistry. Multivariate analysis has been used to

determine the rate of nitrogen loading which may cause

eutrophication in lakes (Brezonik and Shannon 1971) and to

relate land and fertilizer usage to nitrate levels in

streams (Klepper 1978) and groundwater (Ludwick et al.1976;

Rajagopal 1978).

While inductive-probabilistic models are useful and

necessary, they suffer an important disadvantage when used

as predictive environmental management tools. The first

problem is one of applicability beyond the data set used to

derive the model. The results of an inductive-probabilistic

model which relates the concentration of nitrate in a stream

to nitrogen fertilizer application in the watershed must be

used with caution on an adjacent watershed. Soil type,

slope, depth to water table, and other edaphalogical factors

may differ between adjacent watersheds,and these differences

may be expected to increase with increasing distance from

the original model watershed. Inductive-probabilistic

models cannot be adjusted for these changing parameters,

since they imply no causal relationship between dependent

and independent variables. Multivariate models of eco-

systems which include terms for a variety of environmental

independent variables are likely to require a very large

(and expensive) data base if they are to prove useful. This

is particularly true of large system studies which require

a long data base so that short-term climatological and

other effects may be separated from the long-term trend of


Deterministic-Nominalistic Models

In the absence of an extensive, longer-term data base

for a particular ecosystem, deterministic-nominalistic

models are frequently selected for use in large system

studies. These models of course also depend on a large

data base, but some of the data can frequently be gathered

from a much larger literature pertaining to a variety of

different ecosystems and laboratory studies. For example,

the temperature dependence of a biological process such as

nitrification may be assumed to be relatively independent

of even the hemisphere in which it occurs. The simplest

form of deterministic-nominalistic model, which has found

frequent use in the environmental sciences, is the mass

balance (or materials-balance) model.

The potential utility of mass balance models in the

study of complex aquatic biogeochemical cycles was suggested

as early as 1715 by Edmund Halley, who suggested that the

age of the world ocean could be estimated from the rate of

salt transport by rivers (MacIntyre 1970). John Jolly (1901)

was the first to actually construct such a model. It was

not until 1952, however, that Barth proposed that the

residence time of an element in an aquatic system could be

calculated using the annual input and steady state concen-

tration of an element (Goldberg 1963). More recently mass-

balance models have been found useful for studies of the

ionic composition of seawater (Garrels et al. 1975),

chlorides in .the Great Lakes (O'Connor and Mueller 1970),

biochemical oxygen demand in streams (Whipple 1970), and

phosphorus in lakes (cf. Rast and Lee 1978). These models

are of the general form

A storage = Z inputs Z outputs (1.1)

A system is defined by choosing boundaries which offer con-

venient points at which to measure the inputs and outputs

to the system. The algebraic sum of the individual inputs

(forcing functions) and outputs (responses) are determined,

together with the change in material stored in the system.

If the latter value is nonzero, the mass balance model is

termed "dynamic." If there is no change in storage, how-

ever, "steady state" model results. A sufficiently long

time step is frequently chosen so that the sum of transient

changes in storage equals zero over time, and the model be-

haves in a quasi-steady state manner. In a steady state

mode, missing input or output values can be calculated as

the difference between the measured inputs and outputs.

Alternatively, if all system inputs and outputs can be meas-

ured and if the flows can be assumed to be constant over a

given time step, changes in storage can be calculated from

Equation (1.1).

Recently, the use of dynamic mass balance models based

on time-domain simulation techniques has been advocated for

environmental studies (cf. Odum 1971; Rich 1973; Canale

1976). Such models have been attempted for large-scale

nitrogen cycle models (Endelman et al. 1972; Duffy et al.

1975; Fried et al. 1976; Tanji et al. 1977) and several

smaller scale soil column studies (Frere 1975; Rao et al.

1976; Ardakani et al. 1974a). However, in a mass balance

model which represents the different nitrogen species as

compartments, the difficulties outlined in the previous

section generally make valid simulation difficult. The

transfer coefficients from one compartment to another

usually involve either high order or Michaelis-Menton

functions. Furthermore, the individual, in situ coeffi-

cients are virtually impossible to measure. Finally, many

of the nitrogen cycle processes are highly dependent on

carbon substrates, EH, and moisture content. Such a dynamic

model soon begins to make inordinate demands of its data


Simple algebraic mass balance models avoid intrac-

tability by failing to attain a high level of sophistication.

Their simplicity, however, may engender considerable power

at the expense of elegance. Specifically, mass balance

models yield information concerning (1) boundary fluxes

that are difficult or impossible to measure in situ;

(2) potential changes in steady state system storage or

stresses leading to transient behavior; and (3) general

information regarding data gaps and other research needs.

Whereas (1) and (2) may provide more quantitative informa-

tion, (3) may offer qualitative data of significant value.

Boundary fluxes may be calculated by difference when

all but one of the fluxes can be measured, and the system

is known to be at steady state. Frequent modifications of

this technique are the assumptions that one or more of the

fluxes is negligible, or that the system is in a quasi-

steady state. The first assumption may range from easily

justifiable to completely unwarranted. The assumption of

steady state or quasi-steady state dynamics is least tenu-

ous when storage is small relative to flux, i.e., when the

residence time is short. When residence times are long and

storage large, however, a small digression (relative to

storage) from steady state may lead to a rather large

relative change in response to a constant input. This could,

of course, lead to concommitantly large errors in estimating

the value of a given flux, if steady-state kinetics were

supposed to occur. When more than one flux is unknown, only

a net flux may be calculated, but this can be useful for

some purposes.

Mass balance models may also be of utility in sensi-

tivity analysis. If fluxes are known with some certainty,

changes in system storage or responses may be calculated

under different input scenarios. Frequently long-term

extensive environmental monitoring will fail to detect

gradual increases in the concentration of a pollutant be-

cause of patchiness or environmental "noise" caused by

long- or short-term'variation in one or more confounding

input variables. Since monitoring programs are expensive,

they will generally not be undertaken unless they can be

directed toward specific objectives. If mass balance models

can predict steady-state increases in pollutant concentra-

tions or increasing fluxes caused by dynamic behavior, the

need for monitoring programs may be demonstrated. Such an

approach may also be valuable for examining the effects of

subtractions (mining) or additions (waste disposal) of

materials to systems with large storage, where the meas-

urement of short-term changes in the storage themselves is


If a mass balance model fails to meet the needs of the

first two purposes mentioned above, it may still offer in-

valuable semiquantitative information to the environmental

manager or scientist. A review of the literature necessary

to determine forcing functions, storage, and responses

frequently turns up much information regarding the variabil-

ity of process rates or storage. This analysis may lead

to further understanding of the variables controlling a

process, or it may call into question the techniques pre-

viously used to measure the rate of a particular process.

It may be found that a little-known process whose measured

range of rates is wide may be quantitatively much more sig-

nificant than a familiar process with well-known kinetics.

Occasionally, mass balance models may reveal that transfers

of potential environmental significance have been completely

ignored by the scientific community. Some examples will be

shown in the following chapters. Thus mass balance model-

ing may indicate how research time or funding may be allo-

cated to produce the greatest likelihood of benefit.


Historical Development of the Nitrogen
Mass Balance Model

The earliest nitrogen mass balance probably dates from

1804 when DeSaussure used enclosed pot experiments to demon-

strate that plants absorb nitrogen from the soil rather than

the atmosphere (Russel 1961). In 1880, the German agrono-

mist Weaver suggested that nitrogen mass balances be

performed on plants in pots and cylinder to determine the

relative efficacy of different forms of nitrogenous fertiliz-

ers (Lipman and Blair 1916). The greenhouse pot experiment

has since been one of the mainstays of plant physiology

research. Mass balances on the pot or cylinder scale are

relatively straightforward. Errors result only from in-

correct analyses or, more often, ignoring a flux altogether.

Unfortunately, however, these experiments normally modify

the soil structure considerably, and the small scale pre-

cludes studying the effect of local hydrology, micro-

meteorology, and community interactions.

Field plot studies solve some of the problems encoun-

tered in greenhouse work. Lipman and Blair (1916) did

extensive pioneering work in this area, and Russel's work

on field plots at Rothamstead extended over the entire first

half of this century (summarized in Russel 1961). These

early field plot studies were constructed primarily to

study the response of crop yield to fertilizers, however,

and the mass balances were seldom complete. The earliest

nitrogen mass balances included only fertilizer additions,

crop removal, and changes in soil storage. Russel included

rainfall additions and the addition of nitrogen in seeds at

planting time. Leaching losses and gaseous fluxes were not

included, however, due mainly to a lack of interest in re-

ceiving water quality and the lack of a sensitive technique

for measuring the fluxes of various nitrogen gases.

It was frequently noted, however, that measured nitro-

gen removal from pots, cylinders, or plots failed to equal

the measured inputs to the system. This phenomenon was

referred to by Allison (1955), in his classic review of the

early literature, as the "enigma of the soil balance

sheet." Allison attributed the unaccounted for disappear-

ance of nitrogen to net denitrification. Allison again

reviewed the literature in 1965 and in 1973 with the same

conclusion. The impact of nitrogen on environmental qual-

ity depends on nitrogen lost from the system, rather than

that retained by the crops. It is this "enigmatic" loss

of nitrogen that concerns the environmentalist. Large-

scale nitrogen mass balances have done much to clarify the

nature of these losses.

Some of the recent nitrogen mass balances will be re-

viewed in the remainder of this chapter. The balances are

divided into four geographic scales reflecting increasing

levels of aggregation. These are (1) local terrestrial

models covering large field plots or small watersheds,

(2) lake models, and (3) regional models describing large

watersheds or multibasin systems, and finally (4) global

mass balances. The first three categories roughly cor-

respond to the sugarcane farm, Lake Okeechobee, and the

Florida peninsula models presented in Chapters 4, 5, and

6, respectively.

Local Terrestrial Systems

The simplest nitrogen mass balances are constructed

for "local" terrestrial systems, the field or small water-

shed. These systems involve a minimum level of aggregation,

and it is frequently possible to measure many of the stor-

ages and fluxes with considerable precision and accuracy.

Microclimate and soil type are more or less homogeneous, and

intensive sampling programs are usually possible. Hydraulic

outputs for the entire system can often be collected at one

or a few points. These models are of particular utility

in homogeneous agricultural areas.

Small scale modeling has frequently been directed at

measuring seepage losses from agricultural plots receiving

nitrogen fertilizers. The general practice has been to

ignore atmospheric inputs and denitrification and to calcu-

late mineralization by difference. The assumption is made

that old systems are in a steady state with respect to soil

organic nitrogen (SON) and that younger systems exhibit net

mineralization, approaching steady state asymptotically.

Thus the great majority of the nitrogen input is in the

form of fertilizer. Exceptions to the general rule include

the histosols of Lake Apopka (Hortenstine and Forbes 1972),

the Everglades Agricultural Area (Volk and Sartain 1976;

also Chapter 4), and the nitrogen rich fossil loess soils

of Nebraska (Muir et al. 1976).

Nitrogen losses from agricultural systems depend on

climate, hydrology, and size of the system. Losses can be

calculated by estimating the drainage excess from the differ-

ence between precipitation and evapotranspiration (ET).

This value is multiplied by appropriate soil water nitrogen

concentrations obtained using wells or porous ceramic soil

moisture samplers. This approach has been used by Forbes

et al. (1974) for orange groves, by Graetz et al. (1974) for

small millet plots, and by Hook and Kardos (1978) for wood-

land plots receiving spray irrigation of sewage. The

hydrologic balance accounts for all water leaving the root

zone and thus enjoys an advantage of empirical methods

which may miss deep seepage and soil nitrate moving

laterally. The accuracy of the hydrologic balance method

is limited by the extent to which actual ET equals the

theoretical potential evapotranspiration (PET) rate (cf.

Eagleson 1970, pp. 226-234). Frequently, in areas which

exhibit net ET during much of the year, soil nitrate pro-

files alone may integrate nitrogen losses from the root

zone, since no deep seepage occurs (Ludwick et al. 1976).

In small agricultural systems, empirical measurements

of nitrogen losses are often made using tile-drains.

Calvert and Phung (1972) measured nitrogen losses from a

tiled orange grove. Duffy et al. (1975), Gambrell et al.

(1975b), Baker et al. (1975), Gast et al. (1978) and

Klepper (1978) have measured nitrogen losses from small

fields cropped to different grains. Of all of these studies,

however, only that of Gambrell et al. (1975a,b) stands out

as a true mass balance. Small plots (10 x 6 m) planted with

corn or wheat and horizontally isolated by vertical plastic

barriers were monitored for nitrogen fertilizer inputs and

removal in the form of grain, surface runoff, and sub-

surface drainage. Control plots indicated no net minerali-

zation of SON. The experiments revealed that 13 percent of

the nitrogen input was lost to the deep aquifer through

seepage in a moderately drained plot. In a less well-

drained soil, 31 percent of the nitrogen input was unac-

counted for and apparently lost by denitrification. While

tile-drain studies are useful, their main limitation is that

they are economically possible only where tile-drains are

already in use for agricultural applications.

Larger scale investigations of leaching losses are

possible by measuring baseflow and runoff losses from de-

fined watersheds. Recent examples of this type of model

have been published by Thomas and Crutchfield (1974),

Olness et al. (1975), Klausner et al. (1974), Burwell et al.

(1976), Chichester (1976), Kissell et al. (1976), Allen

et al. (1976), Volk and Sartain (1976), Nielsen and

MacKenzie (1977), Miller et al. (1977, 1978), Alberts et al.

(1978), and Klepper (1978). The systems studies ranged from

Florida to Canada and involved a variety of citrus, vege-

table, and field crops together with pasture and rangeland.

The heterogeneity of land uses, climate and hydrology of

these models preclude an in depth comparison, but some gen-

eral trends were noticeable. These are (1) nitrate is the

principle inorganic constituent in runoff; (2) particulate

nitrogen in runoff varies with catchment slope, climate,

hydrology, and land use; (3) nitrate losses increase with

increasing fertilizer application, drainage excess, and

native SON; and (4) low nitrate losses are characteristic

of aggrading (recovering) agricultural systems or of low

net hydraulic loading. Denitrification losses were not

addressed in detail in any of these studies.


As early as 1887, the ecologist S. A. Forbes noted that

lakes represented a convenient laboratory in which to study

the functioning of ecological systems. Lake boundaries are

well-defined, compared to many other natural systems; the

medium is homogeneous compared to soil, and thus it is

easily sampled; and the size is manageable, at least for

small lakes. The pressures of urbanization and intensified

agriculture have resulted in the cultural eutrophication of

many lakes over the past several decades, leading to a

decline in their water quality. Consequently, knowledge of

the transport and cycling of algal nutrients such as nitro-

gen and phosphorus has become an important aspect of water

quality management, and mass balance models for these

nutrients have played a major role in lake management in

recent years.

In 1939, Mortimer constructed a nitrogen mass balance

for L. Windermere, England, in order to compare the rela-

tive importance of runoff, precipitation, and nitrogen

fixation in supplying fixed nitrogen to the lake (Hutchinson

1957). He also compared nitrogen concentrations in in-

fluents and effluents of nine other English lakes of dif-

ferent productivity. He concluded that, in oligotrophic

lakes, nitrogen inputs and outputs closely balanced,and

that denitrification, if it occurred, approximately bal-

anced nitrogen fixation. In the more productive lakes,

however, effluents were lower than influents in total N, and

net denitrification must occur. The only (relatively)

complete mass balance, constructed for L. Windermere, in-

cluded losses from sedimentation and indicated that from 7

to 14 percent of the incoming nitrogen was denitrified,

assuming no in situ nitrogen fixation.

Sawyer (1947) pointed out that by the beginning of the

twentieth century nuisance blooms of blue-green algae had

prompted a committee to study nutrient inputs to the lakes

in Madison, Wisconsin. Lake Mendota is a 3,900 ha, hard-

water eutrophic lake with a mean depth of 13 m and a maximum

depth of 23 m (Torrey and Lee 1976). The watershed is

dominated by rural land, with only 12 percent of the water-

shed urbanized (Sonzogni and Lee 1974). The limnologist

E. A. Birge conducted many of his pioneering studies on the

lake, and it continues to be one of the most widely studied

in the world. Since the report of a consulting engineer to

the Madison City Council in 1918, there have been several

calculations of nitrogen flux through the lake.

Emelty and Hanson (1949) calculated a nitrogen balance

on the lake and found that only 23 percent of the calculated

inflow of nitrogen left the lake in the effluent. By com-

parison, Sawyer (1947) found 30 to 60 percent retention of

nitrogen in the lakes downstream from Mendota.

None of the early studies considered nutrient fluxes

from nonstream sources, so the Nutrient Sources Subcommittee

of the Lake Mendota Problems Committee (Lee 1966)

attempted to estimate these nonpointt" sources by using data

from the literature. Brezonik and Lee (1968) completed the

nitrogen balance by accounted for nitrogen losses from the

lake. Of the 21.69 x 104 kg of nitrogen calculated to

enter the lake annually by Lee (1966), 16 percent

was lost in effluent streams, 11 percent was lost by de-

nitrification, and 67 percent was removed by sedimentation.

The remainder was removed by fish catch and aquatic weed

removal. Sonzogni and Lee (1974) recomputed the nitrogen

loading using more recent runoff coefficients and also

including inputs from seepage and baseflow. The newer

figure totals 567 MT of nitrogen annually, and the implica-

tion follows that nitrogen removal rates by denitrification

must be adjusted upward to account for the additional

nitrogen input. The historical development of the nitrogen

mass balances for L. Mendota has been considered in detail

in a recent report of the National Research Council (NRC


Watershed nutrient export coefficients, such as were

used in the more recent Wisconsin lake balances, are useful

in determining nitrogen inputs to lakes dominated by diffuse

sources, or when monitoring programs are uneconomical. A

recent report by Rast and Lee (1978) summarized nitrogen

loadings to a variety of North American lakes using this

approach. One must be careful, however, in interpreting N

balances based on such data, however, since considerable

variability exists in the data upon which the coefficients

are based. For this reason, mass balances based on empiri-

cal estimates of fluxes and storage should be examined

wherever possible. Unfortunately, few such balances exist.

The most frequent situation is that only data on hydrologic

nutrient imports and exports exist. Such information is

sufficient for phosphorus mass balances, since the differ-

ence between imports and exports plus change in storage

can be attributed solely to sedimentation. Since nitrogen

has several gaseous pathways through which fluxes across

the lake surface may occur, only a net term including

nitrogen fixation, dry gas deposition of NO2 and NH3,

ammonia volatilization, denitrification, and sedimentation

can be calculated. Recent examples of this situation can

be found in nitrogen balances for lakes in the National Lake

Eutrophication Survey (EPA 1975c), for Char Lake,

N.W.T., Canada, (deMarch 1975) and for L. Sodra Bergundasjon,

Sweden, a polluted lake undergoing sewage diversion

(Bengtsson 1978). Occasionally, N sedimentation is meas-

ured using paleolimnological techniques, and the net

gaseous flux can be estimated. The work of Mortimer on

L. Windermere cited earlier and a nitrogen balance on the

Rybinsk Reservoir, U.S.S.R., constructed by Kuznetsov

(1968) provide examples of this type of balance.

Ahlgren (1967) proposed a clever method for determining

nitrogen sedimentation from sediment N:P ratios. Assuming

all of the phosphorus retained in the lake is sedimented,

the N:P ratio of the sediments multiplied by the flux of

phosphorus to the sediment is an estimate of the nitrogen

lost to the sediment. Using this method of calculating

sedimentation Ahlgren found that 37 percent of the nitrogen

entering eutrophic L. Norrviken was lost by denitrification.

Vollenweider (1968) applied Ahlgren's method to existing N

and P balances for six European lakes and found that 45 to

81 percent of the nitrogen inputs were lost to denitrifica-

tion, and that denitrification rates were roughly propor-

tional to the N retention in the lakes. Thus he considered

the nitrogen concentration in these lakes to be self-

regulating. Andersen (1974) applied Ahlgren's approach to

six shallow Danish lakes, mistakenly crediting Vollenweider

with the original method. Anderson calculated that 0 to

54 percent of the nitrogen input was lost from these lakes

by denitrification, despite the absence of stratification.

There was no apparent correlation between nitrogen reten-

tion and denitrification. Jorgensen et al. (1973) found

very high rates of denitrification in a small productive

Danish lake. Larsen (1975) found that 23 and 61 percent of

the total nitrogen input to Lakes Glenstrup and Hald,

respectively, were lost through denitrification. Parallel

experiments indicated that most of the denitrification oc-

curred in the sediments of these lakes. Serruya (1975)

used the sediment N:P approach to calculate denitrification

in L. Kinneret, Israel. Denitrification was found to remove

58 to 62 percent of the annual nitrogen input to the lake.

Most of the denitrification occurred in the later summer

when the hypolimnion became anoxic.

Finally, Brezonik and Lee (1968) and Larsen (1975)

used a mass balance approach on anoxic lake hypolimnia to

determine denitrification rates. Since nitrate inputs are

virtually nil because of isolation of the hypolimnion,

nitrate losses result only from assimilatory nitrate reduc-

tion and denitrification. By measuring the former process

using 1NO3-tracer, the magnitude of denitrification can be

calculated. Brezonik and Lee estimated that 11 percent of

the nitrogen loading to L. Mendota was lost via denitrifica-

tion. Although Larson did not express denitrification

measured by 15NO3 techniques as a percentage of input, he

found that 80 to 95 percent of the nitrate in the hypolimnion

was lost through this pathway.

A summary of the nitrogen balances from the lakes dis-

cussed above is presented in Table 2-1. The variation in

rates between years has been averaged for ease in compari-

son. The nitrogen sedimentation rate is a function of the

sediment N:P ratio, and since the variation of this ratio

on an annual basis cannot be measured, annual estimates of

denitrification are somewhat tenuous. The general absence

of a correlation between areal nitrogen loading rate and the

fraction of the input denitrified is apparent. As was pre-

viously noted, the losses can be quite large. There is

similarly no apparent correlation between the rate of ni-

trogen CEN sedimentation and the nitrogen loading rate.

The lack of relationships points to the individuality of

each lake with respect to nitrogen dynamics. It is inter-

esting that the nitrogen retention (plus denitrification)

rate decreased in L. Sodra Bergundasjon following sewage

diversion. The possible causes for the variability will be

considered in detail in Chapter 7.

Watershed Models

River basin models have a somewhat higher degree of

aggregation, and the heterogeneity of the system modeled is

far greater than is the case for lake models. As a system

becomes larger, the accurate determination of fluxes and

Table 2-1. Summary of nitrogen balances on several lakes1

Net N Lost to %N Input
Lake N Input N Output Net N Lost to N Denitrified %N Input Source
Input Sediments Denitrified









Byrup Langsj

Kvind So

Kul So








Salten Langs$ 25.:

Halle So 85.

Stigsholm So 80.

Glenstrup 13.
Hald 15.
Char 0.41
Kinnaret 16.7
Sbdra Bergundasjbn
Bef. sewage diversion 21.5
Aft. sewage diversion 11.4
















































Mortimer (in
Hutchinson 1957)

Bezonik and
Lee (1966)

















Ahlgren (1967)

Anderson (1974)

Larsen (1975)

De March (1975)
Serruya (1975)

Bengtsson (1978)



storage becomes increasingly difficult. The many micro-

environments provided by different soil types, plant cover,

and microclimate complicate the estimation of atmospheric

fluxes. Croplands become interspersed with other crops,

woodlands, or dwellings, and animal populations, including

humans, become increasingly patchy. Furthermore, system

boundaries are more difficult to define accurately in these

systems than-in lakes. On the other hand, terrestrial en-

vironments offer easier working conditions for field re-

search, and some kinds of data are easier to collect.

Medium and large watershed models thus differ in both con-

struction and interpretation from local or lake models.

Only a few large scale models have been reported in the

literature. Models for both natural and heavily populated

watersheds have been published, and a literature exists for

bioclimatic type systems. One statewide balance and two

national balances (multiple watershed models) have been

published for the United States. At the upper size limit,

global mass balances for nitrogen have been produced since

1944. The global models are compartmentalized according to

phase (rock, surface water, atmosphere, etc.), rather than

geography. Their construction and interpretation is there-

fore different than the watershed model. We might opera-

tionally define the regional model, then, as ranging from

100 km2 upwards, to the size of a continent rather completely

isolated from its neighbors by the sea.

Watershed-scale nitrogen mass balances for relatively

undisturbed systems have recently been reviewed by F. Harris

(NRC 1978). These models are frequently based on limited

data collected at remote field stations or on expeditions.

Grand extrapolation and "rough" estimation seems to be

characteristic of these models, where data regarding simple

fluxes like nitrogen fixation and precipitation are diffi-

cult to obtain or altogether lacking. The results of these

balances are summarized in Table 2-2. Except for the

desert system, with its dearth of soil organic matter, the

systems are characterized by large storage of soil organic

nitrogen (SON) and biomass. These large storage, together

with the small natural fluxes of nitrogen in precipitation

and through internal cycling, lead to long residence times

for nitrogen in these systems. To the balances in Table 2.2

should be added the IBP Tundra Biome Balance constructed by

Barsdate and Alexander (1975).

A discussion of watershed models would be incomplete

without a brief mention of the Hubbard Brook, N. H.,

experiment. Extensive ecological and mineral cycling

studies have been conducted there for about 15 years. The

forested catchments in the watershed are underlain with

impermeable granite, and thus seepage can be ignored.

Surface outflows are gauged and analyzed at stem outflows.

Because no agriculture is practiced and because the catch-

ment is unpopulated, complicated estimates of fertilizer,

manure, and wastewater loadings are unnecessary. The most

Table 2-2. Estimated nitrogen balances for some bioclimatic types. Data from
authors. Values are in kg/ha (pools) or kg/ha-yr (fluxes)1


Deciduous Western Tropical
Deious Tundra Desert Coniferous Grassland Tropical
ForestForest Forest
Inputs 16 1 --
Wetfall 1
NO3-N 4 <1
NH4-N 5 <1
NH3 5
Total N 13 27
Nitrogen Fixation 2-90 4 88

Plant Tissue 492 90 41 1230
Detritus 119 32 68 9 126
Soil Organic Matter 5080 3350 162 2809 91
NH4-N 75 72 326
NO3-N 3 0 175 <1

Internal Fluxes
Uptake 124 11 13 38 102
Mineralization 115 11 13 -

Losses 1
Streamflow -29
NO3-N <1
NH4-N 1
Total N 3 10

Denitrification -0 14 56

Ecosystem Residence
Time (yr) 1370 19 4680

1Slightly modified from NRC (1978).

ambitious mineral mass balances in the watershed to date

have been for conservative or nonvolatile minerals. A first

attempt has been made at a nitrogen mass balance (Bormann

et al. 1977) in one of the catchments, watershed No. 6.

This watershed comprises an aggrading beech forest,

and the nitrogen mass balance indicated that the system

was "tight" with respect to retention of nitrogen. Pre-

cipitation delivered 32 percent of the system's nitrogen

while (net) biological fixation was calculated by difference

to supply the remaining 68 percent. Only 20 percent of the

nitrogen escaped the system, while 80 percent accumulated

in the system. Of the retained nitrogen, 54 percent was

added to living biomass and 46 percent stored as organic

matter. Dry gas deposition and denitrification, which were

not measured, were assumed to be minor pathways. In light

of the gas flux estimates discussed in Chapter 6, it is

likely that this assumption may be considerably modified as

data become available.

Concern over elevated concentrations of nitrate in

drinking water supplies have provided impetus for the con-

struction of nitrogen mass balances for three predominantly

agricultural watersheds in the United States. These are a

small (330 km2) Fall Creek watershed in central New York

state (Johnson et al. 1976) and the much larger upper Santa

Ana River Basin (Ayers and Branson 1973) and the southern

San Joaquin Valley (Miller and Smith 1976), both in

California. The Fall Creek watershed model did not

attempt a total nitrogen balance, nor were pool sizes

estimated. A small mass balance model on nitrate was con-

structed, however, to indicate the relative contribution

of corn, sewage, and precipitation to the nitrate load of

Fall Creek. The balance was used to predict a carrying

capacity for corn production that would not lead to excessive

(>10 mg/L) NO-N concentrations at the mouth of the creek.

The authors noted that only 75 percent of the inorganic

nitrogen estimated to be supplied by precipitation left the

watershed through Fall Creek. Considering that fertilizer

and sewage were not included in the input, considerable

large gaseous losses or changes in storage must have oc-

curred. Experiments indicated that net denitrification was

unlikely in the creek, but it was noted that 40 percent of

the land in the watershed was abandoned farmland, reverting

to woodland. It is therefore not unlikely that significant

increases in stored SON were occurring.

The primary purpose of the two California studies was

to examine small-scale mass-balance models for crop nitrogen

use and human and animal wastes, in order to establish

management practices and research guidelines. Both studies

used comparable methodology and involved extensive ground-

water and soil sampling programs. These measurements

provided unusually reliable data on aquatic fluxes of


The upper Santa Ana River Basin (Ayers and Branson

1973) encompasses 356,000 acres (144,000 ha) and is

characterized by shifting land-use from citrus to urban and

cattle/poultry. Certain areas of the basin are experienc-

ing elevated nitrate levels in groundwater. The southern

San Joaquin Valley (Miller and Smith 1976) is a much larger

area (1.8 million ha, with 1 million ha under irrigation).

This basin has a lower population density and relatively

fewer livestock, poultry, and dairy operations, but much

greater production of food and fiber crops compared to the

Santa Ana basin. Figure 2-1 summarizes the nitrogen cycles

in the two basins. Despite the detail of these analyses,

there are still unknown fluxes (e.g., the partitioning of

sources of nitrogen in subsurface and groundwater flow).

The overall mass balance produced useful equations relating

groundwater nitrate concentrations to fertilizer application

rates and irrigation water excess. The agricultural mass

balances indicated that although the efficiency of nitrogen

use by some crop varieties increased slightly, increased

application of nitrogenous fertilizers between 1961 and 1971

far outstripped that increased efficiency, and percent

nitrogen removal by crops decreased during this period.

Both basins have comparable areal nitrogen loading

rates and potential for leachate loss. Areal surface out-

flows, however, vary by an order of magnitude. The explana-

tion for this variation is not readily apparent from mass

balances alone, but the variable contribution of groundwater

to surface water could account for some of the differences

that were observed. The nitrate concentration of groundwater


U) 0
0 C



Ztr -j(

Jz E


0 E


0 E

Z a:

(B) -- I --
w E


~ dE

Losses ATM. N POOL e irn
4.2 1.13X 107 4.2

-- ------ --------
M Ni --Export
S. Interbasin Transfers 4.

Precip. Gas Losses SURFACE
Water Surnly 1.9 LAND SURFACE 5.8 WATER
4.7 Wate N Frt 264 Return Flow N POOL
15.8 0.6 5.4 07 -
- -- --_. 8. Plant .. . ..1. Pl
Uptake '
--- -- Gas Losses
SOIL N P -L 2 n4
18.1 2.6 X 103 Infiltration
-- -- -- --- -- -_4.7 -- ----
SUBSTRATA Leaching 7.6 Subsurface
N POOL Recharge F lo
3.64 X 104 0.0 F
-- ------ GROUNDWATER ------ ----
Pumped N POOL Groundwater
4.6 Total 53 Flow
Satd. 45
Unsatd. 7.9

Losses ATM. N POOL Gains
57.8 1.390 X 108 57.8

S X tr Interbasin Transfers
t6 1.5
Precip. Ga Losses SURFACE
Water Supply 6.6 LAND SURFACE 28.1 WATER
37.1 N POOL Return Flow N POOL
37.1 teN Fert. 2.218X 102 0 3.807 X 10 '-
V 0-3 ---. -- 9 09 5
151. Plant -- "
S-- - -- -151.5-.Plant _- -- --- -.-.o
Uptake "!
Uptake Gas Losses
221.4 1.878 X 104 Infiltration
----- ------- --37.2-- --
SUBSTRATA Leaching 79.3 Subsurface
N POOL. Recharge -
2.761 X 105 G Flow
-- - GROUNOWATER - -- - -
Pumped POOL Groundwater
30.4 5.230 X 102 Flow

Figure 2-1. Nitrogen pools and fluxes within (A), the
Upper Santa Ana River Basin, and (B), the
Lower San Joaquin River Basin, in California.
Values of pools and fluxes are in thousands
of metric tons and thousands of metric tons
per year, respectively. (A) based on 1960 and
(B) based on 1971 level of development [NRC
(1978), after Ayers and Branson (1973) and
Miller and Smith (1976)].

that was used to calculate the mass balances for the two

river basins varied by nearly an order of magnitude, and

the proportion of pumped to imported water varied between

the two basins. Other factors such as differences in the

size of the basins also could contribute to the differences

in surface outflow. The large size of the San Joaquin Basin

could provide some buffering capacity. Finally, there is

a need to consider spatial distribution of land use, soil

characteristics, and other variables that could influence

the dynamics of the nitrogen cycle of a basin and result in

differences in concentrations of nitrogen in surface water.

A statewide nitrogen mass balance was conducted for

Wisconsin by D. Keeney (NRC 1978), concurrently with an

early version of the Florida peninsula mass balance pre-

sented in Chapter 6. The model was constructed using avail-

able statistical data and thus represents a relatively low-

cost effort. A summary diagram of the Wisconsin mass

balance is presented in Figure 2-2. Many of the assumptions

used in generating the balance were similar to those used

in the Florida balance, and they will not be discussed

separately here. One of the interesting findings revealed

by the model is the nitrogen fixation pathway through

legumes. This nitrogen input to the system represented

55 percent of the annual total and was over twice that from

fertilizer applications. Legume-fixed nitrogen is fed to

cattle, following which it is returned to the SON pool for

subsequent remineralization.

c Mineralization Wastes 260 1 I
(1 percentiyr)
=' ) I Fixation 225 Nonlegumes 15
Alfalfa 79 O
SZ by Legumes 317 Soybeans -2 3.52
> 0 Deposition
(NH4, NO3) 39

Crop Removal
and Residual
Fertilizer 122 Available Soil Ia n R 469s
127 Nitrogen 3615.246
127 386
L -Denitrification,
Leaching Volatilization mportsGO

Total Losses

Figure 2-2. The flow of nitrogen in Wisconsin agriculture in 1974. Pool sizes
and fluxes are expressed in 106 kg N (from NRC 1978).

Several features of the Wisconsin balance that are

typical of mass balances on large systems deserve special

mention. First, the mineralization pathway was estimated

to release 1 percent of the SON annually. Since the SON

compartment dwarfs the remaining pools, a small error in

this estimate will have a significant effect on fluxes

calculated by difference on any compartment driven by

mineralization. The second interesting feature is the large

loss of N through volatilization of inorganic N in manure

(139 x 106 kgN/yr). Finally, 18 percent of the nitrogen

flux through the available soil nitrogen compartment is

unaccounted for and is presumably lost by denitrification

or leaching. Erosional, hydraulic losses are estimated

from soil loss equations in the Wisconsin balance rather

than from water quality data. The calculated loss through

this pathway, 18 x 106 kgN/yr, represents only 3 percent of

the system input. This estimate is almost certainly low

and may point to the importance of nitrate leaching and

transport relative to transport of nitrogen in the suspended

particulate load of rivers.

Two national nitrogen mass balances have been con-

structed. The first was by Lipman and Conybeare (1936) and

is summarized in Table 2-3. System inputs from precipita-

tion and nitrogen fixation were based on meagre data. The

relative lack of importance of fertilizer compared to legume

fixation and manures contrasts the modern situation. River

losses were estimated from water quality data, and leaching

Table 2-3. Soil nitrogen balance for the United States
(Modified from Lipman and Conybeare 1936).

Additions of Nitrogen 109 kg

Fertilizers 0.28
Manures 2.33
Seeds 0.15
Rainfall 3.04
Irrigation 0.02
Legumes 4.96
Nonsymbiotic 3.96

Total 14.7

Losses of Nitrogen

Crops. 4.18
Grazing 7.52
Erosion 4.53
Leaching 4.54

Total 20.8

Net Loss 6.03

Net Loss


losses calculated as the difference between the total loss

of riverine nitrogen and nitrogen loss predicted by soil

loss equations. These authors found a net export of

6.6 x 106 short tons or 6.0 x 10 kg of nitrogen from the

U.S. annually. Whether this represents early mining of

SON may never be known.

A more recent calculation of nitrogen inputs and out-

puts for the U.S. was constructed by NRC (1972)(see Table

2-4). Total inputs were calculated to be 21 x 109 kg,

31 percent higher than the estimate of Lipman and Conybeare.

Thirty-six percent of this input is from chemical fixation,

as opposed to 2 percent in the earlier study. Denitrifica-

tion was assumed to remove 66 percent of the annual nitrogen

input or 13.9 x 109 kg/yr. While Lipman and Conybeare

(1936) calculated a net loss of nitrogen, the NRC model

showed a retention of nitrogen, 1.5 x 109 kg N/yr. The

authors of the balance noted that the most severe problem

involving nitrate pollution involved the production of

animal protein.

Global Mass Balances for Nitrogen

Mass Balances for nitrogen at the global scale have

several methodological advantages. The system boundaries

are the most natural, and the degree of aggregation is the

highest, of nitrogen models on all scales. Furthermore,

most of the globe is covered by oceans, which are consider-

ably more homogeneous than terrestrial environments, making

Table 2-4. Estimates of nitrogen inputs and returns to the
total land area of the United States, 1970 (millions of
metric tons of N) (from NRC 1978)

Annual Inputs to Soil Compartment
Nonsymbiotic N2 fixation 1.2
Symbiotic N2 fixation 3.6
Rainfall 5.6
Chemical fixation 7.5
Mineralization of soil organic nitrogen 3.1
Total input 21.0
Utilization in Plant-Animal-Human Food and
Fiber Chains
Production of fiber 0.2
Production of sugar 0.6
Production of plant protein 0.9
Production of animal protein 15.1
Total 16.8
Total input 21.0
Not utilized in food chains 4.2
Fate of Nitrogen in Food and Fiber Chains
Excreted by humans 1.2
Excreted by animals 4.2
Other nitrogenous wastes 15.6
Total 21.0
Annual Returns to Atmospheric Compartment
As ammonia or oxides to atmosphere moisture 5.6
By denitrification after loss to waterways 5.0
By denitrification from soil 8.9
Total 19.5
Total input 1.0
Net Retention in Soil and Water per Annum 1.5

single measurements more generally applicable to wide areas.

Also, meso-scale atmospheric transport processes (e.g.,

between industrial and rural areas) can be ignored, and

attention can be focused at the scale of transport processes

between maritime and continental air masses.

Balancing these advantages, however, is the reality

that basic data on nitrogen fluxes and reservoir sizes are

lacking for most of the Southern Hemisphere, for most devel-

oping and Communist-Bloc countries of the Northern Hemisphere,

and for much of the world ocean. The high cost of research

cruises limits knowledge of marine fluxes and storage to

relatively few sampling points and concentration profiles.

Lack of data has not discouraged the development of

several global nitrogen mass balances during the past half

century, however, and useful insights have been gained from

such efforts. The strategy of most authors has been to

refine preexisting balances as new data become available,

rather than to use radically new approaches. Clarke's

(1924) geochemical data on riverine mineral transport was

used by Hutchinson (1944), together with various rainfall

data, to calculate global rates of nitrogen fixation and

other atmospheric fluxes. Virtanen (1952) suggested gather-

ing data for a mass balance to determine the rate of global

fixation. Emery et al. (1955) used data from a variety of

sources to produce a nitrogen mass balance for the World

Ocean. They concluded that rivers delivered considerably

less nitrogen annually than was needed for primary production

in the ocean but that oceanic reserves exceeded the annual

requirement by two orders of magnitude. They also noted

that riverine inputs exceeded losses by sedimentation and

hypothesized that denitrification was necessary to balance

the oceanic N budget. Nitrogen fixation was assumed to be

insignificant. Eriksson (1959) produced an interesting

oceanic mass balance model that predicted a net atmospheric

flux of 3 kg/ha-yr of fixed nitrogen to the ocean surface,

a flux he considered reasonable for rainfall. Although

Eriksson's model is not an algebraic mass balance, but a

dynamic differential equation model, it deserves mention

because it presages Ahlgren's (1967) use of N:P ratios to

determine sedimentation rates for nitrogen.

The 1970's saw the first relatively complex global

mass balances for nitrogen. Delwiche (1970) and Wlotzka

(1969) produced balances using data published for the most

part during the 1960's. Burns and Hardy (1975) extended the

global mass balance of nitrogen to include recent data on

nitrogen fixation. Robinson and Robbins (1971) emphasized

atmospheric fluxes in a model that is controversial for its

differences from other analyses. Siderlund and Svensson's

(1976) balance, which has been modestly updated by NRC

(1978), is the most ambitious analysis to date. These

recent balances will doubtless be supplemented and altered

in turn, as more data become available.

The global nitrogen balances are characterized by

large, unaccounted for losses of fixed nitrogen from land

surfaces. Like the models done on an intermediate scale, the

global mass balances indicate that if soil nitrogen is at

steady state, denitrification must be an important sink for

fixed nitrogen. However, almost 98 percent of the total

global nitrogen occurs in tightly bound forms, and about

2 percent occurs in the atmosphere and ocean as N2, which

has low biological availability. Less than one out of 104

nitrogen atoms exist as labile, fixed forms. It is clear

that small changes in these relatively inert pools, caused

by changing tectonics or climatic variations, could con-

tribute significantly as sources or sinks for the global

system. Furthermore, the small reservoirs of fixed nitrogen

forms in the atmosphere have the shortest turnover times.

This suggests potential instability and variability in these

compartments. The variability implies that intensive

monitoring is required for accurate modeling of the flows

through these compartments.


Regardless of model scale, the nitrogen mass balances

in the literature point to large, unaccounted for losses of

nitrogen from the system. Since surface water discharges

are easily measured, the losses apparently occur to ground-

water or the atmosphere. Alternatively, the nitrogen may

accumulate in the system in the form of soil organic nitro-

gen or sediments, although this seems not to be the case

for small plot agricultural systems where estimating the

change in SON accurately is possible. Furthermore, most

nitrogen mass balances ignore dry gas deposition, which

would increase the estimated rate of denitrification in a

steady-state system. Ammonia volatilization is also fre-

quently neglected as a system sink, although this flux is

constrained by local pH and can occasionally be ruled out.

Florida depends heavily upon groundwater for its

drinking water supplies, and its surface waters are im-

portant to recreation and tourism. Because elevated con-

centration of nitrates in groundwater and in lakes may

have serious deleterious environmental effects, it is

important to determine the relative magnitude of the nitro-

gen fluxes in the state and to establish whether imbalances

between input and output fluxes are likely to result in

changes in water quality. In the following chapters,

nitrogen mass balance models for three Florida ecosystems

are constructed to accomplish these ends.


Chapter 3 summarizes the experimental methods used to

collect data for the nitrogen mass balances described in the

following chapters. Sampling and field methods, analytical

and microbiological techniques, and the designs of fre-

quently repeated experiments dealing with nitrogen trans-

formations are described. Short experiments of simple

design that were not repeated are described in the results

sections of Chapters 4 and 5. The data gathering process

for Chapter 6 is strongly oriented toward searching pre-

existing data storage systems rather than conducting lab-

oratory experiments. The methods used for the search will

be described in Chapter 6 along with the results.

Sampling and Field Methods

Soil samples were taken with a Hiler peat auger or by

digging a test hole with a shovel. The sites were randomly

chosen and were always at least 10 m from the nearest canal

or pond. Samples were taken in fields with growing cane

unless otherwise noted. Water samples were taken with a

2 L Van Dorn bottle. Soil samples were transported to the

lab in plastic bags in an ice chest and extracted within

24 hours or were extracted in the field. Water samples were

preserved with 1 mL saturated HgC12 per liter (except for

microbiological samples) and were analyzed for nutrients

within five days. Soil temperatures were determined using

a shaded mercury thermometer. Water temperatures and dis-

solved oxygen were determined in the field with a YSI Model

51 oxygen meter. Redox potential, EH, was determined in the

field using an Orion Model 401 Ionalyzer equipped with a

platinum electrode and a calomel reference electrode.

Sediment cores were taken with a piston corer and kept cool

during transport to the laboratory for extrusion.


Soil and sediment samples were extracted by shaking in

a flask either with distilled water (= 1/20, w/v) or with

2M KC1 (for exchangeable NH4) for 2 hr on a rotary shaker.

The suspensions were centrifuged for 20 min at 6000 RPM and

the supernatant preserved with HgCl2 at 4C for analysis

within 3 days. Nitrate and nitrite were determined using

an automated cadmium reduction method; ammonium ion was de-

termined by the automated colorimetric phenate method and

organic-N using Kjeldahl digestion followed by an automated

colorimetric analysis (EPA 1976). Dissolved organic carbon

was determined on a Beckman Model 915 TOC analyzer following

removal of inorganic carbon by acidification. Dry weights

were determined by oven drying at 1050C for 24 hr. Ash

weights were determined by ashing in a muffle furnace at

5500C for 1 hr. Bulk densities were determined from the

dry weight of a sample following volume determination by

water displacement or by driving metal cannisters into the

soil. Total soil carbon was determined using the Walkley-

Black method (Allison, 1965b)and a conversion factor of

1.33. Total organic nitrogen in whole soil samples was

determined by semimicro Kjeldahl digestion followed by

steam distillation and titration of ammonium ion (EPA

1976). Sediment TKN was determined using micro-Kjeldahl

digestion using the colorimetric finish (EPA 1976).


Samples to be analyzed for denitrifiers or nitrifiers

were taken from the center of cores with a sterile spatula

and transferred to dilution bottles. Denitrifiers were

enumerated by an MPN technique (Focht and Joseph 1973).

Relative nitrifier activity in soil profiles was assessed

by incubating dilutions used for denitrifier MPNs in

Alexander and Clark's (1965) nitrifier broth and monitoring

for NO2 and N03 production.

Mineralization Experiments

Mineralization of soil nitrogen in the drained surface

soils was assessed following the basic method of Stanford

et al. (1974). Eight 25 g samples of interrow soils and

8 samples of soil shaken from sugarcane roots were packed

in 2.54 cm diameter PVC tubes. Glass wool plugs were put

above and below the soil, and a rubber stopper with a glass

tube was fitted to the bottom of the PVC tube. All tubes

were placed in a wooden rack which was enclosed in a

plastic bag during incubations to retard evaporation.

Samples were leached with 10 mL of a nutrient solution

containing 344 mg CaSO4*2H20, 241 mg MgSO4, 1.72 g

CaHPO4 2H20, and 435 mg K2SO4 per liter. Following incu-

bation for a period of several days, the tubes were leached

with 200 mL of 0.01 M CaC12. Initial experiments indicated

that virtually all of the mobile nitrogen could be re-

moved by applying the solution in 50 mL increments, allow-

ing 15 minutes for natural percolation and then applying

vacuum for 5 minutes. The soil columns were conditioned in

this way for one month before data were taken. Samples

were leached from the tubes at 6- to 10-day intervals, fol-

lowing which 10 mL of the nutrient solution was applied to

each tube prior to the next incubation period. The tubes

were then drained, using a vacuum pump, to a given suction,

which varied with the experiment. The leachate was ana-

lyzed for nitrate plus nitrite (NO3 + NO2), ammonium

(NH4), and occasionally, for dissolved organic carbon

(DOC). Anaerobic mineralization experiments were carried

out by incubation in either glass canning jars or stoppered

flasks with samples taken at varying intervals for deter-

mination of NH .

Nitrate Disappearance Experiments

Samples of soil were placed in airtight glass jars,

flasks, or in plastic vials with snap-on caps. The cultures

were spiked with either ammonium nitrate (333 mg/L or with

glucose (684 mg/L) plus ammonium nitrate (333 mg/L). Dupli-

cate jars were analyzed for NO3 + NO2 initially and after

incubation for given periods to determine rates of dis-

appearance. Sampling was accomplished by centrifuging

following extraction on a shaker table or by centrifuging

following autoclaving. The supernatant was then analyzed

for NO2 + NO3.

Denitrification by Acetylene Blockage

Soil columns for acetylene blockage assays were taken

in the field by driving 3" or 6" I.D. PVC pipes into the

soil or by taking sediment cores by hand or with a piston

corer. The columns were perforated at 10 cm intervals

from top to bottom with holes fitted with rubber serum

stoppers. At the beginning of the experiment, 10 mL of

acetylene was injected into the soil column through each

serum stopper. Nitrous oxide was monitored in the closed

headspace by withdrawing a 1 mL sample with a plastic

syringe at various times during the incubation. Experi-

ments were carried out on both closed headspace systems and

with the head gas replaced at a constant rate by means of

a peristaltic pump.

Analysis of N20 was done by gas chromatography using

a Tracor 550 gas chromatograph equipped with a 6Ni electron

capture detector operated at 3500C in pulsed mode with an

argon-methane (95.5) carrier gas. A 6-ft column of Poropak

Q operated at 55C was used to separate N20 from the other

gases (Rasmussen et al. 1976).

Sediment Cores

Sediment cores were taken from the lake using several

different hand-held piston-type coring devices. The cores

were returned to the laboratory and stored with the top

uncapped in a 4C walk-in cooler. Sediments were extruded

within 48 hrs of collection and cross sectioned at 1, 2, 5,

or 10 cm intervals. Cores LO1, L02, L03, L04, L05, and L06

were sectioned medially into right and left halves prior to

cross sectioning in order to provide duplicate total

nitrogen and phosphorus analyses.

In the cores to be analyzed for total Kjeldahl nitro-

gen and total phosphorus only, aliquots of sediment from

the center of the core were dried at 700C for 24 hrs to

determine moisture content and, following weighing, were

ground to a fine powder in a mortar .and pestle. Drying

at 700C prevented excessive loss of nitrogen compounds that

may volatilize at 1050C. Phosphorus content was determined

by suspending aliquots of approximately 50 mg dry weight

in 100 mL of deionized water and analyzed for total

phosphorus (TP) using the acid persulfate digestion,

followed by measurement of reactive phosphate using the

ascorbic acid method (APHA 1975). Total Kjeldahl nitrogen

(TKN) was determined using a macro-Kjeldahl technique.

Aliquots of approximately 40 mg dry weight were placed in

16 cm test tubes, 10 mL of deionized water was added, and

then 2 mL of digestion reagent and a few selenized boiling

chips. The tubes were digested in an aluminum block

digester until they cleared (about 18 hr). The digest

was then brought up to volume, diluted, and ammonium nitro-

gen was determined spectrophotometrically using a Technicon

Auto Analyzer II (EPA 1976). Nitrate was assumed to

contribute negligibly to total N in the cores.

Volumetric aliquots from the center of each cross

section were taken from the cores to be analyzed for inter-

stital water chemistry and microbial populations using a

sterilized plastic disposable syringe from which the needle

end had been cut off to form a cylinder of uniform diameter.

Aliquots were placed in flasks for extraction of nutrients

by shaking with water (or KC1) inoculated into milk dilu-

tion bottles for determination of microbial populations, or

innoculated into serum bottles capped with rubber stoppers

for metabolic measurements. This procedure was done as

soon as the section of core was exposed, and inoculations

were completed within 15 min.

Sediment Traps

Each sediment trap consisted of three wide-mouth

polyethylene bottles suspended in a triangular pattern in

a wood and styrofoam bracket. The trap held the bottles

upright by buoyancy of the bracket, and the trap was held

to the bottom using a metal anchor. The traps were placed

by a diver using SCUBA, and a sample of water was taken to

correct for seston present in the water filling the trap at

the time it was placed. Later the diver returned to the

traps, screwed the caps onto the tops of the bottles in

one trap, and returned the bottles to the surface. The

samples were preserved with 1 mL/L saturated HgCl2 solution,

refrigerated, and returned to the laboratory, where they

were filtered through tared, preashed Whatman GF/A glass

fiber filters in Gooch crucibles at 700 mm Hg suction.

Dry weight and ash weight (5500C, 24 hr) were determined on

the residue.


The Florida Everglades occupies a shallow, 65 km wide

trough extending from the southern shore of Lake Okeechobee

southward to Florida Bay at the tip of the peninsula.

Historically this trough carried the overflow from the lake

as a southerly sheetflow, supporting a luxuriant growth of

sawgrass, Cladium jamaiciense Crantz, which Marjorie

Stoneman Douglas picturesquely termed the "river of grass."

The deposition and incomplete decay of this sawgrass over

the past 4,000 years resulted in the accumulation of a rich

organic layer of rheophilous muck and peat as deep as 3.5 m

in some areas (Davis 1946, Cohen and Spackman 1974). Be-

ginning in 1907, the Everglades Drainage District began to

excavate drainage canals in the area south and east of Lake

Okeechobee and the Everglades Agricultural Area (EAA) was

created (Carter 1974). Today, the EAA comprises 319,000 ha

of partially drained histosols. About 182,000 ha of this

land is presently devoted to sugarcane, vegetables, and

cattle ranching (Zelazny and Carlisle 1974).

Drainage of the EAA has not been without environmental

costs, however. The Everglades histosols are subsiding at

a rate of 3 cm/yr in some areas (Stephens 1974). Although

some of this subsidence is caused by compaction, much of

the loss is caused by oxidation of the soil organic matter

(Volk 1974). This oxidation results in mineralization and

release of inorganic nitrogen in quantities more than suf-

ficient to support the crop, judging from the high nitrogen

concentrations of dissolved nitrogen in the drainage waters

(Brezonik and Federico 1975). During periods of excess

precipitation, drainage water is "backpumped" from the EAA

canals into L. Okeechobee. Because nitrogen may be an

important algal nutrient, and because some phosphorus is

also found in the backpump water, there has been increasing

concern over the effects of backpumping on water quality in

Lake Okeechobee (MacGill et al. 1976). Some of the min-

eralized nitrogen may be nitrified, whereupon it may even-

tually cause problems in South Florida water supplies.

Alternatively, denitrification may produce N2, thus acting

to improve water quality by removing the polluting nitrate.

These concerns have led to the construction of nitrogen and

phosphorus mass balances for several land use types in the

EAA (Shannon 1977).

In an intensive 15 month study, Shannon (1977) col-

lected nitrogen flux data for a hydrologically isolated

sugarcane plantation. These data included nitrogen fluxes

in irrigation and backpump water, and in bulk precipitation

on the plantation. In order to complete the balance, field

and laboratory studies were conducted to determine the

amount of inorganic nitrogen produced by net mineralization

in the plantation each year and to see how much of the

nitrogen input could be lost by denitrification. In situ

denitrification experiments were very difficult to perform

given the remote location of the field site. Therefore,

first an attempt was made to estimate the quantity of

nitrate supplied to the denitrifiers by nitrification.

Then potential denitrification rates under varying environ-

mental conditions of temperature, moisture content, and

the like were estimated. Finally, the results of the

laboratory experiments were related to conditions observed

in the field in order to estimate the most reasonable

denitrification rates in situ. A large soil column ex-

periment was conducted in the laboratory to closely simu-

late field conditions in order to provide an additional

estimate of denitrification rate.

In this chapter, the study site is described, followed

by a brief introduction to the literature on nitrogen

cycling in histosols. Subsequent sections describe the

general characteristics of the plantation soil as well as

laboratory experiments and field observations used to de-

termine the magnitude of nitrogen mineralization, nitrifi-

cation, and denitrification. A nitrogen mass balance for

the plantation is presented in the final section. Discussion

of the environmental implications of the model, along with

possible management strategies, is deferred to Chapter 7.

Model veracity and the reliability of data will also be

considered in Chapter 7.

Site Description

The sugarcane plantation which was studied is located

along the Miami Canal in the north-central part of the EAA

(T. 44 S. and R. 35 E.), south of the town of Lake Harbor

in Palm Beach Co. It is bounded on the north by Palm Beach

County Road 832, on the east by the Miami Canal, on the

south by the Bolles Canal, and on the west by a Seaboard

Coast Line railroad spur. The soils are largely histosols

derived from sawgrass, Cladium jamaicense Crantz. They

range from 70 to 155 centimeters in depth and overlie the

Fort Thompson Limestone formation. The geology, local

vegetation, and soil types of the EAA were reviewed by

Davis (1946), by Jones (1948), and more recently by sev-

eral authors in Gleason (1974). The plantation is drained

and irrigated by a system of ditches and canals which are

connected through a pumping station to the Miami Canal.

The plantation is cropped exclusively to sugarcane

(Saccharum officianarum). A soils map showing the approxi-

mate location of sampling sites is presented in Figure 4-1.

The sugarcane plantation, along with details of the

drainage and irrigation system, is presented in Figure 4-2.

The main canal is approximately 4 m deep and is connected

to the drainage ditches by lateral canals. All of the

drainage structures are dug into the limestone and have

steep sides and a thin layer of sediment which appears to

be derived from slumping of the sides, erosion, etc. The

Lauderhill muck
Terra ceia muck
:: Pahokee muck
Organic soil depth:

S 1 mile

Figure 4-1. Sugarcane plantation. Soils map. Letters A-H
represent sampling sites. (after Shannon 1977).


Main canal
Laten! nnals Cjflhl
Levoes 0 .00 2.':0 ft
Lrt bounoaries
A 00 Mcnitcrinq FPint

Figure 4-2. Map of the sugarcane field showing the
drainage and irrigation system, sampling
wells, and sampling points for hydrologic

ditch sediment does not appear to be highly altered from

the parent peat.

Nitrogen Transformations in Histosols

General Literature

Lowmoor peat soils, or histosols, have been studied

primarily by agriculturists interested in crop yields,

geologists, who see histosols as a laboratory for the study

of coal formation, and soil taxonomists involved in revis-

ing classificatory schemes and nomenclature. The taxonomists

and geologists have not considered nitrogen transforma-

tions. Agriculturists, who are concerned with crop yields,

have not focused attention on nitrogen, since it is usually

available in excess. This has narrowed the literature to

a few papers of a general nature.

Waksman and Stevens (1928 a,b; 1929 a,b) and Tenney

and Waksman (1930) published a set of papers on the chem-

istry and microbiology of both lowmoor and highmoor

(Sphagnum) peats. These papers review the earlier litera-

ture, much of which is difficult to interpret because of

the use of questionable techniques. Davis (1946) and Jones

(1948) published extensive works regarding the Everglades

peats. The former paper includes chemical and some micro-

biological analyses, while the latter is more concerned

with drainage and agronomic factors. Kononova (1961)

published a classic work on organic soils; indeed most of

the relevant histosol literature is published in Russia and

is largely unavailable. Neller (1943), Stephens (1974),

Knipling et.al. (1970), and Volk (1974) have studied the

Everglades subsidence problem but did not consider nitrogen

transformations. Zelazny and Carlisle (1974) published

representative chemical analyses of four Everglades histo-

sols. Hortenstine and Forbes (1972) and Sinclair (1976)

have examined soil nitrogen and drainage water nitrogen

concentrations in L. Apopka histosols (also formed from

sawgrass), and Volk and Sartain (1976) and Brezonik and

Federico (1975) have performed similar studies on an Ever-

glades drainage waters. Given and Dickenson (1973) have

reviewed much of the recent general literature (including

Russian) on peat formation and processes. Finally, valuable

general information can be found in publications by Allison

(1973) on organic soils, Tusneem and Patrick (1971) on

flooded soils, and Alexander (1977) on soil biology and


Nitrogen Mineralization

Mineralization. Mineralization is the process whereby

soil organic matter is converted by the soil microflora to

soluble molecules which can either be further metabolized

or leave the soil system. To accomplish this conversion,

bacteria and fungi excrete extracellular hydrolytic enzymes

which reduce large organic molecules to smaller intermedi-

ates that can be transported across the cell membrane.

Carbohydrates are reduced to mono- and disaccharides,

celluloses, and hemicelluloses. Proteins are hydrolyzed to

amino acids, amides, nucleoproteins, and amino sugars.

Lipids are converted to organic acids and alcohols, and

aromatic compounds, lignins, and waxes are converted to

simpler, more or less refractory derivatives. These low

molecular weight hydrolysates are taken up by the cell,

whereupon they are either immobilized (assimilated into

microbial biomass), or they are oxidized to produce energy

for metabolic processes. If the rate of energetic processes

exceeds that of biosynthetic processes, net mineralization

occurs, and CO2 and ammonium ions are released during the

degradation of carbon skeletons and amino acids, respective-

ly. Carbon dioxide may escape from the system in the gas

phase, but under conditions of neutral pH, ammonium ion is

retained in the soil, whereupon it is subject to nitrifica-

tion and subsequent denitrification or leaching.

Ammonification. The removal of ammonium ion from amino

acids and amino sugars is termed ammonification. Ammoni-

fication rate is determined by several environmental vari-

ables, including the quantity of potentially mineralizable

soil organic nitrogen (SON), cropping practice, aeration,

pH, and C:N ratio. Stanford et al. (1974) demonstrated that

in laboratory incubations, the quantity of inorganic nitro-

gen released from different soils was a first order function

of the quantity of "potentially mineralizable" nitrogen

present in the soil. It is frequently found that untilled

soil initially high in SON shows a steadily decreasing SON

concentration under tillage until a new, lower steady-state

concentration is reached (Stevenson 1965). This pattern

would result if net ammonification were proportional to some

fraction of the SON. The mineralized nitrogen could be

removed from the system by crop uptake, leaching, or


The generally high concentration of SON in the EAA

histosols would thus be expected to result in high ammonifi-

cation rates. The ammonification rate would be further

enhanced by the increased level of metabolism brought about

by aeration of a portion of the soil profile following

drainage. The fact that crops grown on the EAA histosols

require little or no nitrogen fertilizer is prima facie

evidence that net nitrogen mineralization is still occurring

in these soils.

Soil aeration plays a significant role in determining

the ammonification rate by influencing the relative rates

of assimilation and dissimilatory energy production in the

soil microflora. Aerobic metabolism proceeds via the

Kreb's cycle and respiratory electron transport chains,

with oxygen acting as the terminal electron acceptor. The

oxidation of one mole of glucose using 02 as an electron

acceptor yields a free energy change of -687 kcal/mol (pH7)

which drives the production of 38 moles of ATP. A portion

of this ATP is subsequently used in anabolic processes to

produce new cell protoplasm. As 02 becomes depleted in the

environment, oxic bacteria are replaced by denitrifiers,

sulfate reducers, and ultimately by fermentative bacteria.

The substitution of nitrate for 02 as the terminal electron

acceptor results in a slightly lower free energy change

(-649 kcal/mole). Furthermore, the ATP yield is propor-

tionally even smaller than the decrease in free energy

would indicate, because of a lower efficiency of oxidative

phosphorylation in the denitrification electron transport

pathway (cf. Focht and Verstraete 1977). Dissimilatory

sulfate reduction yields even less free energy (-116 kcal/

mole). Fermentative microorganisms lack an electron trans-

port system and generate only 8 moles of ATP per mole of

glucose oxidized (AGo = -100 kcal/mole), assuming the final

products are CH4 and CO2.

The anaerobic pathways thus generate considerably less

energy and ATP than does aerobic oxidation. Consequently,

relatively large amounts of carbon, nitrogen, and other

elements must be mineralized to produce a unit of biomass

under aerobic conditions. Alexander (1977) defines the

carbon utilized orcarbon assimilated ratio as efficiency,

and notes that it is affected not only by soil aeration,

but also by phylogeny, generally being higher in fungi

than in actinomycetes or bacteria. Since fungi are not

present in anaerobic soils, however, it is generally true

that anaerobic soils exhibit the lowest efficiency and the

highest net:gross mineralization ratios.

It is apparently a corollary of the low efficiency of

anaerobic mineralization that complex molecules are more

refractory than under oxic conditions. Tenney and Waksman

(1930) showed that water soluble carbohydrates became less

soluble under waterlogged conditions than under air. This

effect was more pronounced with cellulose and hemicellulose,

and the more resistant fractions of SON were virtually inert.

Under both aerobic and anaerobic conditions, amino

acids, phenols, and flavenoid decomposition products, to-

gether with complex metabolites produced by the soil

microflora, combine to form condensed polyphenolic com-

pounds referred to as soil humus (Given and Dickenson 1973,

Allison 1973). These compounds are refractory even under

aerobic conditions and make up 5 to 20 percent of the SOM

in the Everglades. Given and Dickenson (1973) found no

evidence in the literature that lignins, humic acids, or

fulvic acids were degraded under anaerobic conditions. One

may speculate that complex molecules in SOM are cometabolized

during aerobic metabolism, but that under conditions of low

energy production, cometabolism is energetically wasteful

or even impossible. At any rate, after the easily oxidized

carbohydrates were metabolized during deposition of the

original sawgrass, metabolism of the remaining constituents

under anaerobic conditions in the waterlogged soils pro-

ceeded at a greatly reduced rate. This phenomenon, of

course, allowed the accumulation of the Everglades peat

deposits during the past millenia.

The C:N ratio of the SOM being utilized, and of the

decomposer microflora itself, also affects the net rate of

nitrogen mineralization. C:N ratios of microorganisms are

in the range of 5-15, being slightly higher in fungi than

in bacteria or actinomycetes. If the C:N ratio in the

fraction of SOM being utilized for assimilation exceeds

approximately 15:1, the nitrogen obtained from catabolism

will be incorporated into biomass and the excess carbon

lost as CO2. Progressive loss of CO2 will occur during

subsequent mineralization-immobilization cycles until the

C:N ratio reaches approximately 10:1. Waksman and Stevens

(1929a) found that sawgrass plants themselves had a much

higher C:'N ratio than the peat that resulted from their

decomposition. As a result of this mechanism of nitrogen

enrichment, the Everglades acted historically as a nitro-

gen trap for water leaving the southern shore of Lake



The process of nitrification commonly is mediated by

certain gram-negative, chemoautotrophic,rod-shaped bacteria

that obtain energy for C02-fixation by oxidizing inorganic

nitrogen forms. Ammonium is oxidized to nitrite by

Nitrosomonas (and less common forms such as Nitrosococcus

and Nitrosolobus), and the nitrite is oxidized to nitrate

by Nitrobacter and other less common autotrophs (Alexander

1977). Organic nitrogen and ammonia can also be oxidized

by various heterotrophic microorganisms. Nitrification is

an important process in the transport of nitrogen in the

environment. The negatively charged oxidized forms are not

held by the cation exchange sites of clays and SOM, and

thus they are highly mobile in groundwater. Furthermore,

nitrate and nitrite are necessary for the occurrence of


The requirements of autotrophic nitrifiers are an

ammonium substrate, CO2, and oxygen. It generally has been

noted in culture experiments that Nitrosomonas and Nitro-

bacter oxidize about 35 and 100 atoms of nitrogen, re-

spectively, to fix one molecule of CO2 (Alexander 1965b).

This results in the production of 1 4 x 104 cells of

Nitrobacter or 3 10 x 104 cells of Nitrosomonas per ug of

N oxidized (Ardakani et al. 1974a)in log phase cultures.

The lower yield of the nitrite oxidizers undoubtedly re-

sults from the smaller amount of free.energy released in

nitrite to nitrate (=20 kcal/mole) than is obtained in

oxidizing ammonium to nitrite (=65 kcal/mole). Cell yield

is lower in resting or senescent cultures (Hofman and Lees

1952). The Michaelis constant (Km) for ammonium and

nitrite oxidation at 200 to 300C range from about 1 to

10 mg N/L (Loveless and Painter 1968; Gould and Lees 1960).

Ardakani et al. (1974b) found Km values near the high end

of these ranges under field conditions. Product inhibi-

tion in Nitrosomonas has been observed only at nitrite

concentrations above 500 mg/L (Painter 1970).

While most bacteria have Km values of approximately

3.2 x 10-4 mg/L for 2, Nitrosomonas and Nitrobacter have

Km values for 02 in the range of 0.3-0.1 mg/L (Painter

1970). Thus, under optimum conditions for mineralization,

the respiratory demands of the heterotrophs may well cause

a cessation of nitrification (Focht and Verstraete 1977).

Although autotrophic nitrifiers may exist under anaerobic

conditions, possibly by facultative heterotrophy, nitrifica-

tion is not observed in environments with an EH below 250 mV

(Zobell 1935). Based on a review of the literature, Focht

and Verstraete (1977) reported that nitrification gen-

erally follows the Arrhenius equation in the temperature

range of 150-350C but that autotrophic nitrification ef-

fectively ceases above 400C.

Waksman and Stevens (1929b) found an abundant nitri-

fying microflora in samples from the surface layer of

cultivated Everglades peat. The nitrifier population

decreased with depth, but nitrifiers were isolated as deep

as 110-120 cm in the soil column. The observation that

nitrate is abundant in drained, cultivated sawgrass peat

(Hortenstine and Forbes 1972; Sinclair 1976) is evidence

of the ability of the nitrifiers to oxidize ammonium in

these soils.

Waksman and Stevens (1929b) found that after 50 days

of incubation at 25-280C, 95 percent of the inorganic

nitrogen was released as nitrate from a surface sample of

rheophilous New Jersey peat with an initial water content

of 67 percent. Samples from 60 and 120 cm with 78 percent

moisture produced only 13-14 percent of their inorganic N

as nitrate. A sample from 165 cm produced no nitrate

despite its low (65 percent) water content, indicating the

absence of a nitrifying flora at this depth. Five percent

of the total soil nitrogen was released from the surface

layer after nine months of incubation. Air dried samples

moistened to 200 percent water or more after air drying

produced little nitrate.

In experiments with peats from the Hula River Valley

in Israel, Avnimelech (1971) found that nitrate accumula-

tion (nitrification minus denitrification) was linear over

a 90-day period. The optimum moisture content was 105-

125 percent, and rates decreased at both higher and lower

contents. The rate of accumulation increased by a factor

of 2-3 from 240 to 350C. The soil was calculated to accumu-

late 4500 ppm nitrate annually, in agreement with field

data. This value is approximately 15 percent of the total

soil nitrogen.

While the most familiar nitrifying flora are the auto-

trophic lithotrophs, such as Nitrosomonas and Nitrobacter,

some heterotrophic bacteria and fungi are capable of pro-

ducing nitrate or nitrite during catabolism of organic

nitrogen compounds. Tate (1977) studied autotrophic nitri-

fier populations in samples of Pahokee muck and found that

they were insufficient to produce the amount of nitrate

found in the soil profile, assuming that the autotrophs

are as efficient in the field as they are in pure culture.

Arthrobacter sp., a known heterotrophic nitrifier, were

isolated from the soil in large numbers. Addition of

acetate plus ammonium greatly stimulated the Arthrobacter.

Addition of N-serve, an inhibitor of autotrophic nitrifica-

tion, diminished but did not prevent nitrate accumulation

in cultures of Nitrosomonas. These data led Tate to suggest

that heterotrophic nitrification may be an important pathway

in Everglades peat soils.


Decomposer organisms oxidize organic compounds pref-

erably by using atmospheric oxygen. In the absence of 02,

however, certain gram-negative bacteria can use nitrate or

nitrite as a terminal electron acceptor, reducing it to

N2 or N20. Payne (1973a) listed 17 genera of bacteria in

which at least some species can carry out this process.

Denitrification yields slightly less energy than does oxida-

tion of organic matter with 02, and thus denitrification

does not occur in the presence of appreciable 02. Denitri-

fying bacteria use 02 when it is present, and they do not

exhibit facultative fermentation (Payne 1973a). Extensive

reviews of the biochemistry and microbiology of denitri-

fication recently were published by Payne (1973a), Brezonik

(1977) and Focht and Verstraete (1977). Factors affecting

rates of denitrification and the environmental significance

of the process were also reviewed in the latter two papers.

Generally, the same nutrient kinetics apply to the

denitrifiers as to aerobic heterotrophs. Stanford et al.

(1975c) and Starr and Parlange (1975) reported first order

kinetics at nitrate concentrations below 32 and 40 mg N/L,

respectively. Focht and Chang (1975) observed that first

order kinetics could be extended to nitrate concentrations

greater than 170 mg N/L by supplementing the soil with

glucose. Burford and Bremner (1975) found that soluble

carbon in various virgin and agricultural soils was insuf-

ficient for reduction of the available nitrate and that some

mineralization of carbon was required for complete denitri-

fication. There is often good agreement, however, between

soluble organic carbon levels and denitrification rates

(Stanford et al. 1975c; Burford and Bremner 1975). Modeling

the transient kinetics of denitrification in soil systems

where carbon is slowly mineralized is beyond present

capabilities (Focht and Verstraete 1977).

Although there has been considerable debate in the

past over the occurrence of aerobic denitrification, these

arguments have been answered by the concept of the anaerobic

microsite (Greenwood 1961). Soil particles may contain

anoxic centers, the size of which depends on oxidation rates

within the particle. Denitrification can occur in apparently

"aerobic" soils if anaerobic microsites are present. It

has been found that significant rates of denitrification

can occur in environments in which the gross EH falls below

400 mV (Pilot and Patrick 1972; Van Cleemput et al. 1976),

which is generally taken to be the minimum EH of an oxic

system. The difficulties of measuring the EH in the micro-

environment are reviewed by Flihler et al. (1976).

Stumm and Morgan (1970) reviewed the theoretical difficul-

ties of interpreting EH measurements in general.

Because of the diversity of species involved in

denitrification, Q10 values range from 1.4 to 3.4 for the

temperature range 120-350C. The high optimum temperature

(>350C) for denitrification may result from species

replacement; e.g., Alcaligenes species are probably replaced

by Bacillus species (Focht and Chang 1975); Q10 values are

also affected by substrate and aeration (Novak 1974; Misra

et al. 1974). The diversity of species capable of denitri-

fication also leads to wide pH tolerance, but Wiljer and

Delwiche(1954) showed that significantly more N20 than N2

is produced at pH < 6. Focht and Verstraete (1977) suggest,

however, that short-term incubations may allow insufficient

time for the development of acid-tolerant strains that

could complete the reduction of N20 to N2 in situ. The pH

of the sugarcane soils are generally above 6.0 because of

the buffering action of calcium carbonate; thus, the effect

of pH on product ratios in microbial denitrification and

the chemical production of NO from NO2 in acid soils (Kim

1973) are not of concern in this study.

Molybdenum could theoretically limit denitrification

since it is an integral part of the nitrate reductase

enzyme system, but no studies have reported such limitation

in situ. Certain fungicides and, to a lesser extent,

2,4-D, can inhibit denitrification in situ when applied at

a rate of 100 ppm (Bollag and Henninger 1976). It is not

known whether these chemicals are applied to the EAA soils.

The accumulation of nitrate in histosols, together

with the high oxygen demand of SOM and ammonium oxidation,

would appear to make denitrification highly likely in these

soils. Although the soils literature abounds with informa-

tion of mineral soils, the literature on denitrification in

peats is sparse. Waksman and Stevens (1929b) demonstrated

the presence of an abundant anaerobic soil flora from

110-120 cm depth in an Everglades peat, but most bacteria-

counting methods do not distinguish active from resting life

stages and thus indicate only the potential for growth of

the species in question. Given (unpublished; cited in Given

and Dickenson 1973) reported that no respiratory transport

system activity was found below a depth of 50 cm in

undisturbed Everglades soils.

Avnimelech (1971) reported that the rate of nitrate

loss from anaerobically incubated peat from the Hula Valley

in Israel increased with nitrate concentration and with

temperature. Denitrification was virtually absent at 5C,

and it increased with temperature up to 360C at high nitrate

concentration (4784 ppm). The high potential for denitri-

fication in the Israeli peats led Avnimelech and Raveh

(1974) to develop a method for inducing denitrification in

field plots by sprinkler irrigation. By wetting the soil

down to 60 cm they were able to facilitate the denitrifica-

tion of as much as 850 kg/ha of NO3-N in the fields, or

72 percent of the nitrate originally present.

General Characteristics of the Sugarcane
Plantation Histosols

In this section, the general characteristics of the

sugarcane plantation histosols will be described. The data

derive from more than 250 soil samples collected over the

15 months of the project. These data provide two important

functions in the quantification of nitrogen fluxes in the

plantation. Data on environmental variables such as

temperature, soil moisture, redox potential, and soil

solutes set potential limits on reaction rates and can be

used to extrapolate the results of laboratory experiments

to field conditions. Also, patterns of nutrient concen-

trations in the soil profile provide semiquantitative insight

into the spatial and temporal patterns of nitrogen metabolism

occurring in the soil column.

General Composition of the Soil Solution

Seven soil cores were taken on each of two synoptic

sampling dates (7-29-76 and 2-06-77), and two cores were

taken on five other dates. Vertical soil cores were

sampled for NO2, NO and NH4 at approximately bimonthly

intervals. Total extractable organic N, extractable TOC,

and exchangeable NH4 (with 2 M KC1) were determined less

frequently. The detailed nutrient distributions for the

sampling period June 1976 to September 1977 are presented

in Appendix A. Table 4-1 gives the grand mean values for

each parameter over depth for the entire sampling period.

The mean fraction of water in each sample is also shown.

Table 4-1. Mean concentration of nitrogen forms, total
organic carbon, and physical composition of
soil samples for the entire study period

Nitrogen Forms
(g/m3 S.D.)

NO3-N 39.6 79.9

NO2-N 0.44 0.59

Soluble NH -N 1.99 1.81

Exchangeable NH4-N 14.6 12.0

Soluble TON 22.1 15.2

SOLUBLE TOC 472.0 413.0

Physical Component Fraction of
Fresh Weight

Water (330%)1 0.77 0.13

Oven-dry material (1050C) 0.23 0.13

Ash (5500C) 0.05 0.07

1Percent water on dry weight basis.

The number of samples represented ranges from 256 for nitrate

to 94 for exchangeable ammonium. The high standard devia-

tions of all of the chemical parameters, particularly

nitrate, result from the nonrandom distributions of the

species with depth.

The nutrient concentrations in Table 4-1 are expressed

in g/m3 of soil. This unit is preferred over a mass/mass

unit because the fibrous, organic nature of peat obviates

the expression of nutrients on a g/dry wt basis as is done

for mineral soils. As Given and Dickenson (1973, p. 127) state,

"It is difficult to picture a gram of dried peat." There-

fore, the concentrations on a g/dry wt basis (given in the

Appendix) were converted to volumetric weights by multiply-

ing by the appropriate soil density. The bulk densities

for each depth increment were calculated using the data in

Table 4-2, which were obtained by driving cannisters of

known volume into the walls of test holes at various depths

and determining oven-dry weights of the contents. The

tabulated bulk densities agree with the results of Zelazny

and Carlisle (1974) for similar peats. Since no pattern

of bulk density-depth was discerned below 30 cm, the deeper

parts of the peat profiles were assumed to have a mean bulk

density of 150 kg dry wt/m3

It is clear from Table 4-1 that the majority of nitro-

gen dissolved in the soil water is present as nitrate.

Soluble organic nitrogen is present at a mean 22.1 g/m3

bulk soil, only one-half the concentration of NO3-N

Table 4-2. Bulk density of sugarcane plantation soils1


Depth A B C D E Mean

0 cm 350 360 270 400 420 360

10 cm 380 330 250 340 320

30 cm 370 130 150 220

50 cm 200 150 110

70 cm 130 140 160 160

90 cm 120 130 120 140 150

110 cm 130 150 130

130 cm 190 170 180

Marl 490 490

IValues in kg/m3 (oven-dry weight).

2Stations as in Figure 4-1.

(39.6 g/m3). Most'of the ammonium (88 percent of the

14.6 g/m3 present) is held on the ion exchange sites of

the soil colloids. This is to be expected because of the

typically high cation exchange capacities of organic soils

(Allison 1973). Nitrite is present at low concentration

(0.44 g/m3), since it is readily oxidized by Nitrobacter

and related organisms (Alexander 1965b). Total organic car-

bon had a mean value of 472 g/m3, and the mean C:N ratio for

the soil solution was 6.0. This ratio can be compared with

a mean bacterial biomass ratio of 10 (Alexander 1977),

corroborating the nitrogen enrichment process outlined in

the preceding section. Flooding of the soil is indicated

by its high water fraction (0.77 of the gross weight).

This corresponds to 330 percent water in the convention of

the traditional soil science literature. The highly or-

ganic nature of the histosols and, indirectly, their

relative lack of decomposition is indicated by the low ash

content (22 percent of oven-dry weight). Oven dry weights

of samples collected in September of 1977 were based on

drying for 24 hr at 700C. This procedure was used to allow

minimum volatilization of nitrogen during drying, so that

the samples could be used later for determination of soil

organic nitrogen (SON) (Bremner 1965). Experiments indi-

cated that this lower drying temperature removed 93 percent

of the water that would be removed by drying at 1050C.

Seasonal Variation

The seasonal variations of inorganic nitrogen forms in

sugarcane soil are depicted in Figure 4-3, along with season-

al changes in water content. The infrequent sampling,

together with the high variability between sampling sites,

makes such data useful for observing general trends only.

Nonetheless, it appears that May of 1977 was characterized

by high NO3 and NH4 concentrations, while the low NO2 and

NO3 concentrations during summer of 1976 were accompanied

by slightly elevated NH4 concentrations. Lower nitrifica-

tion rates during the wet summer (1976) period could account

for these trends. The 1977 drought, along with a dry

winter, resulted in generally lower soil moisture after

September of 1976. The high NO and NH+ concentrations in
3 4
May might have resulted from concentration by evaporation,

since soil moisture was low on this date. The relatively

high moisture content measured in early February followed

several days of heavy rain.

Spatial Patterns

Typical depth profiles of nitrogen forms and TOC for

winter (February) and late summer (September) are shown in

Figure 4-4. The February samples were not analyzed for

TON or exchangeable NH4. These sampling dates were chosen

because eight replicate cores were taken, thus giving a

more reliable estimate of the mean concentration profiles

in soils of the plantation. The typical profiles show high







0 0.5


z 1

S-0 --- --

0 0.7580



1976 1977

Figure 4-3. Mean concentrations of nitrogen species and
fraction of water over all depths in soil
over the study period.


N(g/m3) 10 20
TOC(g/m3)200 400

Figure 4-4. Mean distribution
0-D) TON (0-O),



l I I

W u

50 N(g/m3) 10
TOC(g/m3) 200




of NO2+NO -N (--*), soluble NH -N (A-A), total NH -N
and TOC (I-8) in all cores with depth on two dates.

nitrate levels in the surface soils, with concentrations

decreasing markedly below 60 cm. Nitrite was present at a

maximum of 2 g/m3 at the 10-cm depth and otherwise was low.

Ammonium ion was generally low throughout the column, but

it exhibited a slight increase below 100 cm in the October


High concentrations of DOC at 10 cm and at 90 cm were

observed in the February samples, and an increase in soluble

organic N near the bottom of the profile was noted in

September. Two mechanisms are possible to explain the peaks

in concentration of a given nitrogen or carbon form within

the soil profile: (1) the physical process of evaporation

of water, which leaves a concentrated soil solution behind;

and (2) microbial production in excess of decomposition of

the substance at the depth where the peak occurs. When

evapotranspiration.exceeds precipitation, water migrates

upward by bulk flow along the wetted surface of particles,

transporting salts in solution. As the soil becomes drier

(closer to the surface), an increasing fraction of the water

evaporates and moves through the soil as a vapor (cf.

Marshall and Gurr 1954). This leaves the remaining soil

water enriched in the migrating ions. This process of

concentration can occur only above the water table, and

the resulting concentration profile of an inert ion might

be expected to show an inverse relationship to the profile

of soil moisture. A composite plot of moisture content

versus depth, shown in Figure 4-5, indicates that evaporation

60 t

80 1


120 -


Figure 4-5.

Mean moisture content of all soil samples taken
during the study versus depth in the sugarcane










I t t r

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs