WATER POLLUTION CONTROL RESEARCH SERIES • 16010 EHR 03/72
ROLE OF BACTERIA IN THE
NITROGEN CYCLE IN LAKES
U.S. ENVIRONMENTAL PROTECTION AGENCY
-------
WATER POLLUTION CONTROL RESEARCH SERIES
The Water Pollution Control Research Series describes the
results and. progress in the control and abatement of
poflution in our Nation’s waters. They provide a central
source of information on the research, development, and.
demonstration activities in the water research program of
the nvironmenta1 Protection Agency, through inhouse
research and. grants and. contracts with Federal, State, and
local agencies, research institutions, and industrial
organizations.
Inquiries pertaining to Water Pollution Control Research
Reports should be directed to the Chief, Publications Branch
(Water), Research Information Division, R&M, Environmental
Protection Agency, Washington, D.C. 2O -6O.
-------
ROLE OF BACTERIA
IN THE NITROGEN CYCLE IN LAKES
by
Elizabeth F. McCoy
Department of Bacteriology
University of Wisconsin - Madison
for the
Office of Research and Monitoring
ENVIRONMENTAL PROTECTION AGENCY
Program #16010 EHR
March 1972
For sale by the Superintendent of Documents, U.S. Government Printing Office
Washington, D.C., 20402 - Price 35 cents
-------
EPA Review Notice
This report has been reviewed by the Environmental
Protection Agency and approved for publication. Ap-
proval does not signify that the contents necessarily
reflect the views and policies of the Environmental
Protection Agency, nor does mention of trade names
or conimercial products constitute endorsement or
recommendation for use.
1].
-------
ABSTRACT
In a 3-year study, 1690 samples were tested in the
field for N0 3 -N, N0 2 -N and NH -N and in the laboratory
for nitrifying and denitrifying bacteria and fungi.
The sampling sites were fresh waters, underlying muds
and beaches.
Biological nitrification, both heterotrophic and auto-
trophic, was demonstrated. Values for N0 3 -N above
10 ppm were common; 30—60 ppm were often found on
beaches with decomposing organic matter.
Denitrifying bacteria were prevalent at the same sites;
more than 70% of 628 samples contained more than l0 /m1.
Nitrification and denitrification are opposing pro-
cesses but can coexist either in close succession or
in adjoining microhabitats. Thus, the field values
for N03-N and N0 2 -N vary considerably and must be
viewed as net values at any given time.
Experiments with 13 species of locally caught fishes
showed great difference in resistance to N02-N. Perch
and brook sticklebacks were killed in 3-5 hr at 5 ppm.
Carp and black bullheads tolerated 40 ppm for 2 wks
and 100 ppm for about 24 hr. The susceptibilities of
other species varied. Nitrite toxicity may influence
the dominance of fish species in a eutrophic lake.
Submitted in fulfillment of Program No. 16010 EHR
under sponsorship of the Environmental Protection
Agency.
iii
-------
CONTENTS
Section Page
I Conclusions 1
II Recommendations 3
III Introduction 5
IV Experimental 7
Nitrification Studies 7
Heterotrophic 7
1969 Field Studies 9
Denitrification Studies 12
1970—71 Field Studies 12
Nitrite Toxicity to Fishes 16
V Acknowledgments 21
VI References 23
V
-------
FIGURES
Page
1 Populations of Denitrifiers
in Marsh and Muddy Sites 14
2 Populations of Denitrifiers
in Water Sites 15
3 Seasonal Counts of Denitrifiers in
Three Sites--Marsh, Stream and Spring 17
vi
-------
TABLES
Page
1 Spring Survey 1969 10
2 Identity and Characteristics of Fishes 18
vii
-------
SECTION I
CONCLUSIONS
In the course of a 3-year study, 1690 samples were tested
in the field for NO 3 —N, NO 2 —N, and NHL -N in ppm and in the
laboratory for populations of nitrifying and denitrifying
bacteria and fungi. The sampling sites were mainly in the
Madison, Wisconsin area and included lakes, ponds and
streams, their waters, underlying muds and beaches.
1. Biological nitrification, both heterotrophic and auto-
trophic, was demonstrated by correlating laboratory data for
high populations of nitrifiers at sites where field data
showed NO3—N and NO 2 -N present.
2. Values for NO3—N above 10 ppm were common, and concen-
trations as high as 30-60 ppm were often found on beaches
with decomposing organic matter and in muds under shallow
water. Water samples from open lakes were generally nega-
tive; occasional samples from bays and backwaters were
positive at 2 to 3 ppm. Occasional beach sites with piles
of dying algae and aquatic weeds in the late summer were
positive at 180+ ppm (calculated on the water basis in the
beach sand)
3. Denitrifying bacteria, mainly Pseudomonas sp., were
prevalent at the same sites, and their numbers ranged from
less than 10 to 10 6 /ml; more than 70% of 623 samples showed
more than 10’ /m1. There was some indication of seasonal
fluctuation and of higher denitrifier counts at sites where
nitrification was high or had been high previously.
The two processes, nitrification and denitrification, are
opposing, but they can coexist either in succession or in
adjoining microhabitats at the same sampling site.
4. Toxicity of N0 2 -N to fish was shown in the laboratory
at levels which are commonly found in the field (formed
either by nitrification or by reduction from NO 3 -N).
Perch ( Percina caproides ) were most sensitive, dying in less
than 3 hr at 5 ppm; likewise the brook stickleback ( Eucalia
inconstans ) was killed by 5 ppm, but in the slightly longer
time of 3-5 hr. On the contrary, the carp ( Cyprinus carpio )
and the black bullhead ( Ictalarus melas ) tolerated 40 ppm
to the end of the 48-hr test. Even at 100 ppm the carp sur-
vived for 45 hr and the bullhead for 24 hr. The common
sucker ( Catastomus commersoni ) lived 48 hr at 100 ppm; and
1
-------
the quiliback ( Caproides cyprinus) , about 36 hr at 100 ppm.
The seven other species tested varied in tolerance, most of
them surviving less than 12-24 hr at 20 to 40 ppm.
5. A possible significance of nitrite toxicity is sug-
gested. Since the bottom feeders, such as carp, bullheads
and suckers, are most tolerant, their survival in shallow
eutrophic waters may be favored. And, conversely, the sen-
sitive fish, such as perch, are killed in a few hours at
levels of nitrate which are common, according to our field
data for shore samples of eutrophic waters. It is conceiv-
able, therefore, that the effect of N0 2 —N may be one factor
in the change of dominance of fish species as a lake
progresses in eutrophication.
2
-------
SECTION II
RECOMMENDATIONS
Nitrification and denitrification are opposing processes.
Both were shown to be active systems at sampling sites in
shallow waters and beaches of lakes and streams.
1. Since the chemical data for N0 3 -N, N0 2 -N, and NH -N
for field samples show only the net or temporary balance
between nitrification and denitrification (including nitrate
reduction) , such data are not meaningful alone. Rates of
formation and transformation must be obtained for modeling
the events of the N cycle in eutrophic lakes.
2. Since nitrification is dependent upon protein N (for
heterotrophic) and ammonia N (for autotrophic), nitrifica—
tion is most active on beaches and in shallow water with
dying algae, aquatic plants, or dead fish.
Therefore, such decomposing organic matter should not be al-
lowed to accumulate. Weeds cut from the water should not be
piled on shores which drain back to the lake. Masses of dead
fish should be removed. Even shoreline improvement by re-
moving fallen trees or other obstructions would help in
preventing the accumulation of dead plant material in shallow
water.
3. Failure to remove organic matter, as in (2) above, could
result in enough NO 2 -N (from either the first step of nitrif i-
cation or from N0 3 —N reduction) to cause kills of sensitive
fish or their fry in shallow waters. Such N0 2 -N, in levels
commonly found in the shore sites tested, would be enough to
affect the dominance of fish species by favoring carp and
other bottom feeders. For these reasons also, shoreline
“housekeeping” should be encouraged.
3
-------
SECTION III
INTRODUCTION
In water, as in soil, the inorganic compounds in the nitrogen
cycle are a particularly important nitrogen source for higher
plants and algae. Some problems of eutrophication depend
upon the rate at which such inorganic nitrogen becomes avail-
able, and this rate in turn depends upon bacterial action.
Thus, a study of populations of nitrifiers and denitrifiers
should reflect potential transformations for which they are
the agents.
The objectives of this study were: 1) to explore biological
nitrification as contributing nitrite and nitrate to lake
and stream waters by determining: a) types and numbers of
nitrifiers, and b) their sites of growth and activity; 2) to
determine numbers of denitrifying bacteria and their potential
activity as opposing nitrification; and 3) to obtain and
interpret field data on N0 2 -N and N0 3 -N in terms of popula-
tions of nitrifiers and denitrifiers.
The nitrogen species in question are NH -N, N0 2 -N and N0 3 —N.
The processes by which they are formed and transformed and
the bacterial agents concerned are as follows:
Ammonification is the release of ammonia by enzymatic degra-
dation of nitrogenous organic matter which is carried out by
many heterotrophic bacteria. The ammonia can remain in
aqueous solution or be released as gaseous NH 3 to the air.
It may later be returned to solution in rain or snow.
Nitrification is the stepwise oxidation from NH +-N to 1) N0 2 -N
and 2) N0 3 -N, which is carried out by specialized bacteria or
fungi; both autotrophic and heterotrophic types are known.
Nitrate reduction is the stepwise reduction of N0 3 -N to N0 2 -N
d usually to NH 4 -N. It is the reverse of nitrification.
Many bacteria in soil and water carry out nitrate reduction
as they metabolize the oxidized N for their own growth or for
an electron acceptor. It is important to note that N0 2 —N can
arise from either step 1 of nitrification or step 1 of nitrate
reduction.
Denitrification is the release of gaseous N2 or N 2 0 by the
reduction of either NOz-N or N0 3 -N. This can be accomplished
by relatively few but common types of bacteria in soil and
water under conditions when oxygen is otherwise limiting.
5
-------
In view of the above definitions, it is clear that field
data for N03-N, N02-N and NHi+—N do not reveal which process
is dominant at the site or time of sampling. Determination
of populations of nitrifying and denitrifying bacteria,
although not conclusive evidence as to their activity, helps
in the interpretation of data.
6
-------
SECTION IV
EXPERIMENTAL
The work reported here followed several lines which are pre—
sented separately, although in some cases the same samples
were used for different purposes. The methods are given
briefly in each section below.
Nitrification Studies
Heterotrophic
In the past a specialized group of autotrophic bacteria have
been considered the main agents of nitrification in soil
(Alexander, 1965) . Because conditions for their autotrophy
are restricted, little attention has been given to excessive
nitrification in soil. For the same reasons and, additionally,
because of the low level of nitrogen in natural waters of
lakes and streams, one would not anticipate strong nitrifica—
tion in these waters. Yet, high values for NO 3 -N in waters
do occur at times; and the question arises whether nitrifica-
tion can account for the levels found, or whether such N0 3 —N
has its origin external to the lake or stream (e.g., the
N0 3 —N of fertilizer or soil nitrification under unusual con-
ditions such as runoff from feed lots)
In addition to the autotrophic nitrifiers, certain hetero-
trophic bacteria, actinomycetes and fungi have now been
recognized as nitrifiers (Eylar and Schmidt, 1959; Alexander,
1965) . The genera concerned are Mycobacterium, Nocardia,
Streptomyces, Micromonospora and Streptosporangium (Hirsch
et al., 1961); Arthrobacter (Gunner, 1963); Agrobacterium,
Bacillus, Corynebacterium , and Pseudomonas (Alexander, 1965)
The first fungus to be recognized as a heterotrophic nitri-
fier was Aspergillus flavus (Schmidt, 1954, 1963; confirmed
by Marshall and Alexander, 1962). It produces nitrate from
amino nitrogen in Schmidt’s medium. Other soil fungi (e.g.,
Penicillium sp..) growing in Schmidt’s medium cannot form
N0 3 -N unless provided with NO 2 -N, as in step 2 of nitrifica-
tion. Reviewing the various systems, Alexander (1965) con-
cluded that heterotrophic nitrification occurs only when
nitrogen is present in excess of cellular needs and when an
energy source other than the oxidation of N is available.
Usually heterotrophic nitrification by bacteria in Amino N
medium stops at NO2-N formation, although N03-N can subse-
quently be formed by the Penicillium type system using N0 2 -N
7
-------
as substrate for N03-N formation. It may be noted, too, that
autotrophic nitrifiers could carry out this final oxidation.
Since many of these heterotrophic nitrifiers are common in
waters of lakes and streams and in the adjoining beaches and
soils of the immediate drainage basin, this investigation was
undertaken to assay their importance as contributors of nitrate
and nitrite to eutrophic waters.
Confirmation of heterotrophic nitrification by fungi . As a
starting point, the work of Schmidt with A. flavus was con-
firmed and his technique was then adapted to the testing of
water samples. The medium of Schmidt is:
PartA K 2 HPO .. LOg
MgSO 7H 2 O.. . .... 0.5
FeSO 7H 2 O 0.01
MnSO 4H 2 O 0.01
Distilled water.. 900 ml
Part B Glucose 4.0 g
Peptone 8.0
Distilled water.. 100 ml
The parts are sterilized separately and combined in the de-
sired quantity as needed. In the experiments reported below,
100 ml in 500 ml Erlenmeyer flasks were used.
A. flavus was tested along with a collection of 128 other
species and strains, kindly provided by Dr. Kenneth Raper,
Bacteriology Department, University of Wisconsin. They com-
prised all 8 morphological groups within the genus Asper-
gillus . They were grown in Schmidt’s medium on a rotary
shaker at 30°C for 5 days (aerated vigorously) or in standing
cultures at 30°C for 14 days (aerated gently). Concentra-
tions of N03-N, N02-N and NHi+-N were determined by the
Bremner-Keeney distillation method (Brernner and Keeney, 1965).
Approximately 70% of the cultures tested produced N0 3 -N in
amounts ranging 28-115 i.ig/ml. The highest producers were the
A. flavus and A. wentii groups; all strains produced 65-100 pg/mi.
öf 24 species and strains of Penicillium , 21 were able to pro-
duce N0 3 -N at 5-10 pg/mi levels but only when the amino N of
Schmidt’s medium was supplemented with N02-N. Similarly,
several cultures of Fusarium, Gliocladium, Memnoniella,
Scopulariopsis and Myrothecium produced 9-13 pg/ml of N0 3 -N
8
-------
when supplied with N02-N. These latter genera have not been
recognized before as nitrifiers.
1969 Field Studies
The study was begun by comparing heterotrophic vs autotrophic
activities of cultures in 2 series: Amino N medium of Schmidt
for the heterotrophic and NH+-N Synthetic medi um for the auto-
trophic. The latter medium was chosen from the general lit-
erature on nitrification by soil autotrophs; it contained:
Glucose 2.0 g
(NH ) 2 So 1.0
MgSO 7H 2 O 0.2
CaC1 2 •2H 2 0 0.1
FeSO .•7H 2 O 0.01
ZnSOLf•7H 2 0 0.001
MnSOz 4 •H 2 0 0.001
Na2MoOl. 2H 2 0 0.001
Distilled water.. 1000 ml
Again the glucose and mineral salts were separately steri-
lized and combined to give 100 ml of medium in 500 ml Erlen-
mdyer flasks. Incubation was at 30°C in shaken and standing
cultures for 14 days.
In the spring of 1969 , 191 samples were taken of waters,
underlying muds and beach sands from lakes, ponds and shallow
streams. On the larger lakes (Monona and Mendota) various
shore conditions were represented. Most of the samples were
obtained from the Madison area, but, in addition, 3 came from
the Wisconsin River, 6 from the Sioux River and 6 from Lake
Superior. As shown in Table 1, the greatest number of posi-
tives appeared in Schmidt’s medium. Microscopic examination
showed numerous gram—negative bacteria, actinomycetes and
vegetative fungal hyphae, which indicates active hetero-
trophic nitrification. Forty-seven of these cultures were
retested in Schmidt’s medium; although still only crude
enrichment cultures, 45 of them produced N0 2 -N in the range
of 2-154 pg/ml, with an average of 48 pg/mi. Only 2 of them
produced NO 3 -N (33 and 46 pg/ml). These 2 cultures contained
numerous fungal hyphae but the genus is unknown, since only
vegetative growth was seen. They could have been Aspergillus ,
probably not Penicillium sp. Thus, it appears that for these
191 samples the heterotrophic process of nitrification was
dominant and that most of the action stopped at NO 2 -N. How-
9
-------
ever, this is not conclusive evidence for lack of autotrophic
nitrification because of the difficulty in quantitative assays
for the autotrophic nitrifiers, Nevertheless, the autotrophic
system, if active, should have proceeded to N0 3 -N. In addi—
tion, the conditions in Schmidt’s medium favor the conclusion
that the heterotrophic system was present and presumably active
under the conditions of the test.
Table 1
Spring Survey 1969 . Testing of 191 samples for nitrifica—
tion in laboratory media, differential for autotrophic and
heterotrophic systems.
Samples
tested
Positive sam
NH N
(autotrophic)
pies in media
Schmidt’s
(heterotrophic)
Waters — 32
1
4
Shallow
lakes
mud of
and ponds
-
25
5
11
Beaches
— 38
1
28
Stream
mud and sand
- 23
17
15
“Positive” means any value in excess of minimum recorded,
>2 pg/ml for N0 2 -N and >5 pg/mi for N0 3 -N.
In the summer of 1969 , a more extended survey was made, in-
volving 700 samples taken from 66 stations in the Madison
lakes area. The sampling continued over a 3—1/2 mo period,
and most of the stations were visited from 6—10 times.
Records of the weather, water temperature and biota at the
stations were kept, and the NH -N, N0 2 -N and N0 3 -N tests
were done in the field. For the latter purpose, a field kit
and spot plate test were developed, using Nessler, dimethyl
a—naphthylamine and diphenylamine reagents for ammonia,
nitrite and nitrate respectively. Known standards of
(NHz ) 2 SOt , NaNO 2 , and NaNO 3 at 100 ppm as NHt -N, N0 2 -N and
N0 3 —N were carried in the kit. Proper dilutions were made
in the field, and color comparisons were made between the
test samples and known standards. An example of the pro-
cedure and sensitivity of matching in the field is as
follows. If the sample being tested for N0 3 -N showed a
distinct but light blue color, it would be compared with
0,1,2,5 ppm of the known standard. If dark or maximum blue
10
-------
color was seen in the first spot test, the sample would be
diluted to match the color for 2 ppm and the dilution factor
used to calculate the N03—N value for the sample. The
colors for “trace” and 1 ppm of the known standard were
appreciable, but the readings for less than 2 ppm were rather
subjective. Thus, such values were reported as less than
2 ppm without differentiation. All higher values are re-
ported as calculated data.
Water samples from open lakes and streams were generally
negative (no color by our spot test); occasional samples
from bays and backwaters were positive at less than 2 to
3 ppm as N0 3 —N. The data for shore samples were sorted
into 2 lists: those greater and less than 10 ppm. In
general, values less than 10 ppm resulted from samples col-
lected on clean sandy shores. Water weeds and algae were
identified and their presence at the shore sites recorded,
because of epiphytic bacteria on them. Beach samples,
particularly those taken under masses of dying algae,
yriophyllum or Lemna , were nearly always positive at levels
greater than 10 ppm. Occasionally such samples in late sum-
mer showed 180-200 ppm of N0 3 -N (calculated on the water basis
in the sandy beach sample) . Most such beach samples, however,
showed 30-60 ppm of N0 3 -N on the same basis. Ammonia was
high but the amount varied greatly. In one mass of dying
algae, an ammonia concentration of 600 ppm was found and
the pH was understandably high--pH 9-9.8. Thus, at certain
places at least, there is substrate N for both heterotrophic
(dead algal N) and autotrophic (NH -N) nitrification.
About one half of these samples were also tested in the lab-
oratory in shaken flasks at 30°C, to determine their potential
to support higher nitrification under aerated conditions.
Most of them (236 samples) did support higher nitrification.
They became strongly positive, but still yielded less than
10 ppm after 10 days. Many of these samples, when tested in
the field, showed very low NHt+-N values, ranging from trace
amounts to 5 ppm. Thus, their nitrification potential was
apparently limited by lack of N. Consequently, another set
of 82 samples taken at random were supplemented with 50 ppm
ammonia as (NHt+)2HPO and incubated on a rotary shaker at
30°C. Within five days 79 of the 82 became positive for
N03—N with an average of 6.2 ppm. Longer incubations were
not tested. Since no attempt was made to provide optimum
conditions for either heterotrophic or autotrophic nitri-
fication, it cannot be said how high the values might have
been. It is remarkable that N03-N was produced so rapidly,
and this appears to mean that autotrophic nitrifiers were
present and were able to respond immediately to NHi -N added
to the low—nutrient natural water samples.
11
-------
A further indication that available N is the determining fac-
tor in riitrifying activity can be seen in the following
laboratory observation. Six samples of Lake Mendota Bay
mud and related water were set up in aquaria and stocked with
small minnows. One aquarium initially had no detectable
N0 3 —N and was still negative at 30 days. All five others
were positive at levels greater than 10 ppm at 30 days, but
the time at which they became positive was quite variable.
One that was low (2 ppm at 30 days) was then supplemented with
50 ppm of NH -N as (NH ) 2 HPOt+ and became positive--6 ppm at
10 days, 22 ppm at 30 days.
In the course of use of some of the aquaria, the experimental
fishes died. One of them was removed to a beaker with 100 ml
of Lake Mendota water and allowed to decompose naturally (by
proteolysis and ammonification). After 5 days, the remains
of the fish and all floating debris were removed and the water
aerated at room temperature on a magnetic stirrer. A control
of the same water without the fish autolysate was run in
parallel, and it remained negative for nitrification at 20
days. The fish-water specimen became positive at 8 ppm by
5 days and greater than 20 ppm by 20 days. It was noted in
the summer survey that beach sand samples taken under dead
fish often were strongly positive in the 30-to--greater-than-
60 ppm range.
Denitrification Studies
1970-71 Field Studies
A small survey was made during June and early July to confirm
the results of the previous summer. A total of 175 samples
were taken from the same sites as in 1969, and 25 additional
samples were collected from new sites on lake shores and small
streams. The results showed that high N0 2 -N values (2 to
18 ppm) and less commonly high N0 3 -N values (greater than 10
to as high as 70 ppm) were related to organic matter on the
shore sites. However, in some cases it was found that repeti-
tion of sampling at the same site on successive days did not
always confirm the previous high value. Sometimes a storm or
visible change, such as wave action at the site, seemed to
account for the difference, but this was not always the case.
Thus, it was decided that denitrification should be studied
as the main program for the summer of 1970, and the results
were so interesting that an extension of time for study in
the summer of 1971 was sought. The discussion which follows
presents the results of both summers of study.
During late August to December 1970, 245 samples were tested
simultaneously for both nitrification and denitrification
12
-------
activity. These samples were collected from Lakes Nendota,
Mononar Waubesa and the Nine Spring Creek. The nitrification
tests consisted of field determinations as before, and the
results were consistent with the prior data. The denitrifi-
cation activity was judged from dilution counts on Giltay’s
medium (mineral salts with nitrate as the N source and citrate
as the C source) . Readings were taken on the basis of gas,
N 2 or N 2 0, and MPNs were calculated. The denitrifiers at the
different sites ranged from less than 10 to more than
5 x 10 6 /ml; 189 of the 245 samples had more than l0 3 /ml and
144 of them had more than 5 x l0 /ml. From the high dilution
gas—positive tubes, 3 to 5 serial transfers on Giltay’s
medium were made to provide enrichments from which to isolate
and identify the denitrifiers. Over 85% of the 300+ cultures
tested proved to be Pseudomonas sp. This was not unexpected,
since pseudomonads are the principal denitrifiers in the soil
and are numerous in all natural waters.
An analysis of the total body of data on denitrification was
made. This included the 245 samples of the summer of 1970,
followed by 378 samples taken mainly during the winter and
spring of 1970-71 and a few collected during the months of
July and August of 1971. The samples were sorted according
to the following types of sites:
Open lake waters
Stream and spring waters
Shore waters--turbid, suspended matter both living
and dead
Runoff waters
Beach waters——clear, off sandy beaches
Marsh muds, usually marl
Muds under shallow water
Sand on beaches, with small organic debris
Dry weeds and algae on beaches
The data pertaining to these sites are grouped as to waters
and muds in Figures 1 and 2. The graphs show numbers of
samples positive at each population level of denitrifiers
detected. These graphs are not based upon equal numbers of
samples, and in a few cases only a few samples were tested,
but nevertheless they do show several interesting points,
namely: 1) the great range of numbers of denitrifiers from
less than 10 to 5 x 10 6 /ml. Commonly, l0 to 10 5 /ml were
found regardless of whether the specimens were mud or water;
2) water samples yielded a greater number of positives than
mud samples; and 3) stream and spring water samples, and
shore water samples provided the largest number of positive
13
-------
D
r
U)
a)
r-H
E
8
a)
>
4 J
U)
0
0
Figure 1.
Populations of denitrifiers
in marsh and muddy sites
16
A Marsh marl
• Bottom mud and water
O Sand on beach
o Weeds and algae on shore
0
10
101
102
10
I
10
I
SxlO
I
I
5x10
106
Cell numbers per ml
-------
Figure 2.
Populations of denitrifiers
in water sites
o Beach water
o Open lake water
• Runoff water
V Shore water
• Streams and springs
16
rH
l2
6 )
rH
B
U)
6)
>
•H 8
U)
0
0
4
0
10
101 102 j 3
lO 5xl0 l0 5x10 5
106
Cell numbers per ml
-------
samples, This latter point i not a reflection of greater
numbers of such sampies apparently these sites are favor-
able for growth of pseudomonads, but the controlling factors
are not known.
Another approach to analyzing the ecology of the denitrifiers
was made by plotting the numbers/mi against seasons of year.
Figure 3 is a graph of data for 3 types of sites: streams,
springs and marshes. The populations of denitrifiers are
high in winter months, drop in late spring, and rise again
in late summer. The rise in late summer was confirmed by
the data from the second summer. The peak in winter is un—
derstandable because of cold—weather survival of the late
fall populations. The drop in spring is more difficult to
explain. It may be due to a low point in available N0 3 -N
substrate as water plants absorb their nitrogen. Another
possibility is antagonism of the general microbial flora
in the warming waters. Perhaps, readily available nutrients
become limiting, but this is not likely because of the di-
verse substrates utilizable by pseudomonads. Possibly it is
a combination of factors, such as lack of nitrate substrate
in late spring, which prevents the aerobic pseudomonads from
competing with the more facultative and fermentative bacteria
of the lake. Conversely, in winter under the ice where free
oxygen may become limiting to the general bacteria, the
nitrate-reducing denitrifiers would have an oxygen advantage
over many bacteria. There could then be growth of denitri—
fiers to account for the high winter populations. This is
possible, because many pseudomonads are psychrophiles. Un-
fortunately, we did not test the denitrifying pseudomonads
for growth at near freezing temperatures. All of the above
comments are speculative but could account for the apparent
seasonal changes in the populations of denitrifiers. High
populations and potential activity of denitrifiers are
important because of the possible opposition to nitrification
in natural water systems.
Nitrite Toxicity to Fishes
It was recognized that concentrations of both N03-N and N0 2 -N
in some samples were within the known levels for toxicity to
animals, and perhaps also to plants. Such toxicity could
potentially have considerable importance in eutrophic lakes.
The death of frypan and game fishes in certain shore sites
could be considerable and, if differences in susceptibility
could be shown, toxicity might be a factor in the change of
species dominance in eutrophic lakes.
16
-------
106
iü
rH
a)
10
U)
a)
0
0
H
H
a)
C-)
102
10
J
Figure 3.
Seasonal counts of denitrifiers in
three sites: marsh, stream, and spring
• Marsh
• Stream
A Spring
iar
Time of year
-------
Nitrite toxicity was studied first, because it is the more
toxic ion and, if the concentrations found in lake and
stream systems were not toxic to fish, there would be no
need to test the less toxic N0 3 —N.
Representatives of 13 species of fishes were used. The
species tested, their preferred habitats and food, and where
they were caught in the Madison area are given in Table 2.
Table 2
Identity and characteristics of the fishes
Per cid ae
Percina caproides Log perch
Collected from L Mendota and Mississippi R
On rocky shores, feeding on crustaceans, tubefix
and various invertebrates
Etheostoma nigrum Johnny darter
Collected from L Mendota but usually in streams
Feeding like perch
Centrachidae
Lepomis macrochirus Bluegill
Collected in slow and stagnant waters, feeding on
invertebrates in surface water
Lepomis gibbosus Pumpkinseed
Like bluegill but preferring many molluscs,
esp snails, in diet
Cyprinidae
Netropis spilopterus Spotf in shiner
Pelagic, spawning in rivers
Plankton feeders
Netropis stramineus Sand shiner
From Wisconsin R at Spring Green; in open sandy
areas of rivers
Plankton feeders
Cyprinus carpio Carp
Collected from Sugar R and Black Earth Cr
Bottom feeders
18
-------
Table 2 cont,
Gasterosteidae
Eucalis inconstans Brook stickleback
Collected from Dunn’s marsh, in weed-clogged channels
Voracious feeder on live food only, but after feeding
often kills all remaining live food in sight; rather
inactive between feedings
Catastomidae
Hypentilium nigricans Hog sucker
Collected from Black Earth Cr
Bottom feeder on aquatic plants and detritus
Catastomus cornmersoni Common white sucker
Collected from L Mendota
Bottom feeder on detritus, often in polluted water
Caproides cyprinus Quillback
Collected at Wisconsin R
Bottom feeders in sandy silt
Ictaluridae
Ictalurus melas Black bullhead
Collected in L Waubesa, Hog Island
Bottom feeders, predacious on molluscs, fish,
invertebrates, also saprophytic feeders on detritus
Noturus flavus Stone cat
Collected in Black Earth Cr
Feeding in rocky-riffle areas on drifting matter
The fish were held in the laboratory in Madison city water
in well aerated aquaria until tested, usually within 2-3 days
of capture. Toxicity tests were conducted with 3 small
fingerlings or minnows of the test species in gallon jars.
The sides of the glass jars were painted to minimize dis-
turbances during the test period. The levels of N0 2 -N tested
were 5,10,20,40,100,200 ppm, a range covering the N0 2 —N values
actually found in the field (max 73 ppm) or potentially found
in aerated lake water in the laboratory (max 154 ppm). As
stated earlier, 30-60 ppm of N0 3 —N (readily reducible to
N0 2 -N) is the range more commonly found at shore sites with
decomposing organic matter. The range from 30 to 45 ppm of
N0 3 -N in water is generally regarded as potentially toxic to
some animals.
19
-------
Great differences were found in the susceptibility of fishes
to N02-N. Perch ( Percina caproides ) were most sensitive,
dying in less than 3 hr with 5 ppm. Likewise, the brook
stickleback ( Eucalia inconstans ) was killed by 5 ppm, but in
the slightly longer time of 3 to 5 hr. On the contrary, the
carp ( Cyprinus carpio ) and the black bullhead ( Ictalarus melas )
tolerated 40 ppm to the end of the 48—hr test. Even at
100 ppm the carp lived for. about 45 hr and the black bullhead
24 hr. The common sucker ( Catastomus coinmersoni ) lived 48 hr
at 100 ppm, and the quillback ( Caproides cyprinus ) about
36 hr. The other species tested ranged between these extremes,
most of them surviving less than 12-24 hr with N02-N in the
range 20 to 40 ppm. Control fishes were kept under the same
conditions except for the N0 2 -N challenge, and they were living
and fully active after 4 weeks.
The green frog (only one tested) was remarkably resistant.
It lived 4 weeks in water with added NO2-N at 100 ppm; in
fact, the N02-N was added several times to make up for losses
by microbial action. Finally, when subjected to a challenge
dose of 200 ppm, the frog died within 22 hr.
The differences in susceptibility of the fishes cannot now be
explained. It is interesting that the bottom feeders were
the most resistant, and this fact may have some bearing upon
their ability to survive and displace the more desirable
species in eutrophic waters. It is also interesting that the
perch and several species of minnows were susceptible to as
low as 5 ppm N02-N, and our field data show that this range
of concentration is common in shore samples from eutrophic
lakes. It is conceivable that nitrite toxicity could be very
serious to the fry in their shoreline habitat.
20
-------
SECTION V
ACKNOWLEDGMENTS
It is a pleasure to acknowledge support of this
work by Program 16010 EHR of the Environmental
Protection Agency. The stimulus received from
Dr. Paul Uttormark and other colleagues in the
eutrophication program in the past three years
has been very beneficial. Thanks are also due
the graduate student assistants, Margaret
Heimbrook, Terry Thompson and James Kenyon, who
participated in the field and analytical work,
and to Don Samuelson for his aid in obtaining
the fish and testing nitrite toxicity upon them.
21
-------
SECTION VI
REFERENCES
Alexander, M. 1965. Nitrification. Chapter 8 in Soil
Nitrogen , Bartholomew, W. V., and F. E. Clark, eds.
Amer. Soc. Agron., Inc., Madison, Wis.
Bremner, J. M., and D. R. Keeney. 1965. Steam distilla-
tion methods for determination of arnmonium, nitrate
and nitrite. Anal. Chim. Acta 32:485-495.
Eylar, 0. R., Jr., and E. L. Schmidt. 1959. A survey of
heterotrophic microorganisms from the soil for ability
to form nitrite and nitrate. J. Gen. Microbiol. 20:
473—481.
Gunner, H. B. 1963. Nitrification by Arthrobacter
globiformis . Nature 197:1127—1128 .
Hirsch, P., L. Overrein, and M. Alexander. 1961. Formation
of nitrite and nitrate by actinomycetes and fungi.
J. Bacteriol. 82:442—448.
Marshall, K. C., and N. Alexander. 1962. Nitrification
by Aspergillus flavus . J. Bacteriol. 83:572—578.
Schmidt, E. L. 1954. Nitrate formation by a soil fungus.
Science 119:187—189 .
Schmidt, E. L. 1963. Cultural conditions influencing
nitrate formation by Aspergillus flavus . J. Bacteriol.
79:553—557.
GO’ERN 1EN1 PRINTIN F ICFJ 1972—4 44 6/24’
23
-------
SELECTED WA TER 1. Report No. 2. 3. Accession No.
RESOURCES ABSTRACTS W
INPUT TRANSACTION FORM
4. Title ROLE OF BACTERIA IN THE 5. Report Date
NITROGEN CYCLE IN LAKES 6.
_______________________________________________________________________________ 8. Performing Organization
7. A uthor(s) Report No.
McCoy, Elizabeth F. 10. ProjectNo.
16010 EHR
9. Organization ___________________________
University of Wisconsin, Madison 11. Contract/GrantNo.
Water Resources Center
13. Type of Report and
Period Covered
1?. Sponsoring Organization
15. Supplementary Notes
16. Abstract In a 3-year study, 1690 samples were tested in the field for
N03-N, N02-N and NH4 N and in the laboratory for nitrifying and denitri-
fying bacteria and fungi. The sampling sites were fresh waters, under-
lying muds and beaches.
Biological nitrification, both heterotrophic and autotrophic, was
demonstrated. Values for NO3-N above 10 ppm were common; 30-60 ppm were
often found on beaches with decomposing organic matter.
Denitrifying bacteria were prevalent at the same sites; more than
70% of 628 samples contained more than 10,000/mi. Nitrification and de-
nitrification are opposing processes but can coexist either in close
succession or in adjoining microhabitats. Thus, the field values for
NO3-N and N02-N vary considerably and must be viewed as net values at
any given time.
Experiments with 13 species of locally caught fishes showed great
differences in resistance to NO2-N. Perch and brook sticklebacks were
killed in 3-5 hr at 5 ppm. Carp and black bullheads tolerated 40 ppm
for 2 wks and 100 ppm for about 24 hr. The susceptibilities of other
species varied. Nitrite toxicity may influence the dominance of fish
species in a eutrophic lake.
17a. Descriptors
*Aquatic bacteria, *Aquatic fungi, *Denitrificatjon,
*Nitrate reduction, Nitrification, *Nitrite toxicity, *Nitrogen
compounds, *Nitrogen cycle, *Nitrogen fixation, Ammonia, Decomposing
organic matter, Eutrophication, Nitrates, Nitrites, Toxicity.
17b. Identifiers
Autotrophic bacteria, Heterotrophic bacteria.
I 7c. CO WRR Field & Group 0 SC
18. Availability 19. Security Class. 21. No. of Send To:
(Report) Pages ____________________________________________ -
WATER RESOURCES SCIENTIFIC INFORMATION CENTER
20. Security Class. 22. Price U.S. IDEPARTMENT OF THE INTERIOR
(Page) WASHINGTON, D. C. 20240
Abstractor Elizabeth McCoy Institution University of Wisconsin-Madison
WRSIC 102 (REV. JUNE 1971) GPO 913.261
------- |