Aquacultural Engineering 19 (1998) 29 – 39

Simple electrometric methods for estimating microbial

activity in aquaculture ponds
Delma Bratvold, Craig L. Browdy *
Waddell Mariculture Center, P.O. Box 809, Bluffton, SC 29910, USA

Received 1 February 1998; accepted 8 August 1998

Abstract

Simple electrometric methods were tested to estimate microbial activity in hypereutrophic

aquaculture ponds using equipment common at aquaculture facilities for biochemical oxygen
demand (BOD) measurements. Activity measurements tested included net photosynthesis
based on changes in dissolved oxygen (DO) concentrations in samples incubated under light
and dark conditions; general microbial activity as indicated by oxygen consumption rates in
the dark; nutrient limitations as indicated by changes in DO consumption rates with nutrient
additions; and nitrification based on the difference in DO consumption rates between
samples incubated with and without a nitrification inhibitor. Method evaluations in intensive
shrimp culture ponds suggest that simple electrometric methods are a good index of water
column microbial activity under a range of conditions. © 1998 Published by Elsevier Science
B.V. All rights reserved.

Keywords: Oxygen consumption; Microbial activity; Photosynthesis; Nitrification; Nutrient

limitations

1. Introduction

Many aquaculture systems rely upon the pond microbial community to enhance
growth, maintain water quality, control waste accumulation and limit opportunities
for disease. Traditionally, the microbial community in shrimp ponds has been
managed through the exchange of water with adjacent natural waters. In recent
years, problems associated with introduction of pathogens and eutrophication of
surrounding waters have led to concerns regarding the sustainability of this

* Corresponding author. Tel.: +1 843 8373795; fax: 843 8373487; e-mail: browdycl@musc.edu

0144-8609/98/$ - see front matter © 1998 Published by Elsevier Science B.V. All rights reserved.
PII S0144-8609(98)00040-5
30 D. Brat6old, C.L. Browdy / Aquacultural Engineering 19 (1998) 29–39

practice. Environmentally-friendly management strategies with reduced water ex-

change decrease the dilution of pond water. Possible negative repercussions of
reduced water exchange on water quality in systems that have traditionally relied
upon water exchange increases the importance of monitoring microbial processes to
aid in the establishment of alternative proactive strategies to maintain optimal
water quality.
Microbial processes can significantly influence water quality and target crop
health through affects on dissolved oxygen, ammonium, and labile (i.e. readily
biodegradable) organic matter, indirectly measured as biochemical oxygen demand
(BOD). The heterotrophic microbial community uses oxygen while feeding on
unconsumed feed, photosynthetic products, and target and non-target crop waste
products. Nitrogen waste products and carbon dioxide are used in phytoplankton
growth, which often includes the release of labile organic matter (Fogg, 1962;
Thomas, 1971; Sharp, 1975; Ittekkot, 1982). Labile organics, along with oxygen, are
consumed by water column bacteria (Ittekkot, 1982), thereby recycling carbon and
nitrogen in a microbial loop with continual oxygen demand.
General water column microbial activity can be indicated by oxygen consump-
tion rates (Coffin et al., 1993). Methods for measuring photosynthesis and general
activity from changes in DO concentration have often used Winkler titration
methods (e.g. Oviatt et al., 1986; Jensen et al., 1990; Hansen and Blackburn, 1992;
Coffin et al., 1993) because oxygen electrodes generally do not have sufficient
precision. However, the rapid rate of DO consumption in the hypereutrophic
aquaculture ponds may not require a high level of precision. Polarographic
electrodes have been employed for measurement of photosynthesis and oxygen
consumption in aquaculture ponds using a relatively sophisticated automatic sam-
pling and monitoring system (Szyper and Rosenfield, 1992).
The objective of this study was to develop and test simple, low cost methods as
an index of microbial activity in aquaculture ponds using equipment that is
commonly used at commercial farms. The methods tested employed standard
equipment for conducting biochemical oxygen demand (BOD) measurements.
Microbial processes assessed included net photosynthesis, general microbial activity
(as indicated by oxygen consumption rates), and nitrification. Field quantification
of microbial processes may facilitate development of flexible management strategies
that are responsive to natural fluctuations in microbial populations, and may
improve understanding of the causes of variation between ponds at a single site and
between ponds in different regions.

2. Materials and methods

Measurements of microbial processes were conducted with a YSI 5000 DO meter

with a YSI 5010 BOD probe (Yellow Springs Instruments, Yellow Springs, OH).
The meter was calibrated and operated according to manufacturer recommenda-
tions. Samples were vigorously shaken to assure initial DO measurements between
90 and 100% saturation. Unless specified otherwise, samples were incubated in
 D. Brat6old, C.L. Browdy / Aquacultural Engineering 19 (1998) 29–39 31

triplicate 300 ml BOD bottles washed with 10% hydrochloric acid, and rinsed with
tap water. Incubation time was based on the rate of DO change in the samples.
Incubation was typically stopped when dark controls fell to 2–4 mg/l, or light
bottles (for photosynthesis) increased to more than 10 mg/l. Incubation times
ranged from 1 to 18 h. To minimize instrument error, the DO meter was calibrated
every hour, at temperatures within 2°C of initial sample temperatures.
2.1. Sample collection and processing

Four-liter water samples were collected from 0.1 ha shrimp ponds with a dipper
lowered just below the surface of the pond from a drain riser that extends into the
deepest section of the pond. Additional sample sites, described below, were used to
measure in-pond variation. Upon sample collection, water was immediately passed
through a 150 mm mesh to remove large particulate matter. Samples were processed
and incubations begun within 1.5 h of sample collection.
2.2. Potential sources of measurement error

Several types of measurement error were tested, including DO meter precision

and bias due to incubation volume, introduced oxygen during multiple DO
measurements, temperature, and sample location within a pond. DO meter preci-
sion and the minimum change in DO that could be confidently measured was
determined by reading DO in air-saturated samples of the same salinity and
temperature 33 times. The potential significance of small amounts of oxygen
introduced into samples during periodic DO measurements such as used for general
activity measurements was tested by comparing final DO measurement of triplicate
samples measured every hour for 5 h to samples with only initial and final DO
measurements. The potential significance of the ‘bottle effect’ (i.e. increased bacte-
rial activity and growth in smaller volumes) was tested by measuring DO consump-
tion in triplicate 60 ml and 300 ml BOD bottles. Results of these tests were assessed
with a t-test.
To examine potential bias due to changes in incubation temperature, replicate
samples were incubated with and without nutrient supplements under three temper-
ature conditions: benchtop (25°C, approximate pond temperature), refrigerator
(5°C), and warm water bath (33°C). Initial DO measurements were made within
1°C of collection temperature. Sample temperatures gradually changed throughout
the incubation period. Results were assessed with a two-way ANOVA followed by
Dunnett’s All Pairwise Multiple Comparison.
Variation in oxygen consumption rates based on location in a 0.1-ha, PVC lined
pond with 15 hp/ha paddlewheel aeration was tested by collection of samples from
five locations within the pond: surface water in the middle of the pond and at the
pond drain riser; bottom water in the middle of the pond and at the pond drain
riser (i.e. the deepest part of the pond); and surface water from the edge of the
pond. Within-pond variation was tested on two dates, the first test included three
replicate samples per location, the second test included five replicate samples per
location. Results were assessed with a one-way ANOVA.
32 D. Brat6old, C.L. Browdy / Aquacultural Engineering 19 (1998) 29–39

2.3. Photosynthesis

Net photosynthesis was determined by calculating the difference in DO concen-

tration change in closed samples incubated under light and dark conditions
(Talling, 1973). Recognizing the dependance of photosynthesis on temperature and
light intensity (Bidwell, 1974), samples were simultaneously incubated in a pond,
just below the water surface.

2.4. General acti6ity

General water column activity was indicated by measurements of the DO

consumption rate (Coffin et al., 1993). Dissolved oxygen concentrations were
measured at least four times during incubation in the dark to allow calculation of
oxygen consumption rate by linear regression. A linear decrease of DO concentra-
tion in samples was assumed to indicate that overall microbial activity rate in the
samples was constant and similar to the ponds.

2.5. Nutrient supplements

To indicate the effect of nutrient additions on microbial activity, blackened BOD

bottles were incubated with and without potential nutrients: 1 mM ammonia, and
0.1 mM glucose. Increases in microbial activity due to nutrient addition were
indicated by increased oxygen consumption rates over controls. Rates were assessed
with an ANOVA followed by Dunnett’s All Pairwise Multiple Comparison.

2.6. Nitrification

Nitrification was measured by the difference in the DO consumption rate

between samples incubated in the dark with and without the nitrification inhibitor
2-chloro-6 (trichloromethyl) pyridine (Nitrification Inhibitor Formula 2533, Hach
Company, Ames, IA) (APHA, 1995). In a pilot study, this inhibitor did not
significantly affect respiration rates in five randomly selected, non-nitrifying bacte-
rial isolates from pond water, thus supporting inhibitor selectivity. Significant
differences between control and inhibited DO consumption rates were determined
using t-tests (alpha= 0.05). When significant differences were found, nitrification
was expressed as the rate of oxygen consumption due to nitrification, and the
proportion of total oxygen demand due to nitrification.

3. Results

The range between the highest and lowest DO in replicate, air saturated samples
was 0.07 mg/l, or about 1% of the average measurement. Based on this, a
conservative rule was established that DO must change by a minimum of 0.5 mg/l
during the incubation for estimation of DO consumption rates. Total DO changes
during the incubations of this study were several fold greater than 0.5 mg/l.
 D. Brat6old, C.L. Browdy / Aquacultural Engineering 19 (1998) 29–39 33

No statistically significant differences were found in comparisons of oxygen

consumption rates in 300 and 60 ml BOD bottles, but there was a tendency toward
greater variation among replicates in the smaller samples. No statistically significant
differences were found in DO concentrations in bottles that were measured five
times versus those that had only initial and final DO measurements. The power of
both of these tests was B 0.8, suggesting the possibility of a Type II error
(acceptance of a false null hypothesis).
Control and nutrient supplemented samples incubated at different temperatures
had significantly different rates due to temperature (P= 2.6×10 − 19) and nutrients
(P= 5.1× 10 − 9). There was significant interaction, i.e. the magnitude of rate
change with nutrient addition depended upon temperature (P= 1.3× 10 − 4). Dur-
ing these incubations, under cold conditions, sample temperatures gradually fell to
17°C, while they rose to 33°C under warm conditions. The resulting oxygen
consumption rates were nearly twice as great in the warmest group than in the
coolest group (1.28 mg/l per h and 0.67 mg/l per h, respectively). In the warmest
group, ammonia supplementation resulted in a 32% increase in the oxygen con-
sumption rate over controls at the same temperature, while only a 16% increase in
oxygen consumption rates was observed between cold supplemented and control
samples.
Finally, in tests of in-pond variation, no significant differences in oxygen con-
sumption were found in the first test conducted with triplicate samples from five
locations, however, the power of this test was B0.8. In the second experiment
using five replicates, significant differences (P=7.0× 10 − 4) were found in samples
from the same locations tested in the first experiment. The minimum and maximum
location means in this later experiment were 0.288 mg DO/l per h and 0.337 mg
DO/l per h, respectively.
Tables 1 – 3 show the results of tests conducted on three dates for measurements
of net photosynthesis, general microbial activity (i.e. oxygen consumption rate), and
nitrification rates in three replicate ponds (i.e. same size and management strategy).
Similar data regularly collected throughout the growout season could provide
relative indications of pond microbial activity. Throughout all measurements made
in this study, linear regression coefficients of determination from dissolved oxygen
versus time were typically \0.95, suggesting a constant rate of oxygen consump-
tion during the incubation period.

Table 1
Photosynthesis rates (mg/l/h) in three replicate shrimp ponds on three dates

Pond April 27 June 3 July 2

Mean S.D. Mean S.D. Mean S.D.

1 9.18 0.07 3.02 0.03 4.45 0.01

2 3.42 0.01 3.85 0.05 3.93 0.04
3 2.91 0.01 3.91 0.04 6.03 0.21
34 D. Brat6old, C.L. Browdy / Aquacultural Engineering 19 (1998) 29–39

Table 2
General activity as indicated by oxygen consumption rates (mg/l/h) in three replicate shrimp ponds on
three datesa

Pond August 20 August 22 September 8

Mean S.D. Mean S.D. Mean S.D.

1 1.19 0.04 1.01 0.02 0.59 0.02

2 0.61 0.03 0.68 0.03 0.45 0.01
3 0.59 0.02 0.73 0.04 0.56 0.01

a
Linear regression coefficients of determination ranged from 0.9956 to 0.9999.

Fig. 1 displays the percent change in oxygen consumption rates with glucose and
ammonia additions. Increased oxygen consumption with nutrient addition suggests
microbial activity was limited by the added nutrient. Ammonia limitation was
greatest in June and July. Labile carbon (i.e. glucose) appeared to be more limiting
in September.

4. Discussion

4.1. Meter error and incubation 6olume

The results of tests conducted to examine the significance of measurement errors

due to DO meter precision, and biases due to incubation volume and introduced
oxygen during multiple DO measurements suggest that these error sources were not
significant. However, the power of both of these tests was very poor (i.e. B 0.8),
suggesting a more than 20% chance of a Type II error. Tests with greater power
(i.e. replication) may yield the alternative statistical conclusion. From a practical
standpoint, small differences due to method variation can be controlled by using
the same procedure for all measurements to be compared. The BOD probe used in
this study was designed specifically to fit BOD bottles with no air-water interface
during measurement. This design may offer greater precision than general DO
probes. DO meter precision can be optimized by periodic calibration of the DO

Table 3
Nitrification rates (mg/l/h) in three replicate shrimp ponds on three dates

Pond August 7 September 4 September 17

Mean S.D. Mean S.D. Mean S.D.

1 0.02* 0.009 0.13 0.005 0.10 0.006

2 0.03* 0.005 0.15 0.005 0.13 0.003
3 0.04 0.005 0.21 0.002 0.12 0.001

* Too low for accurate measurement with incubation times used.

 D. Brat6old, C.L. Browdy / Aquacultural Engineering 19 (1998) 29–39 35

Fig. 1. Percent change in oxygen consumption rates with glucose and ammonia supplementation
compared to controls.

meter, and conducting meter calibration at the same temperatures as sample

incubation. Furthermore, recognizing the potential for measurement bias due to
oxygen introduction during repeated DO measurements, it is prudent to use
standard (300 ml) BOD bottles for incubation and not make excessive repeated
measurement from the same incubation bottle. For the purposes of this study, five
time points were used to confirm a linear response. After linearity is confirmed,
fewer time points may be acceptable for similar samples with similar incubation
times.

4.2. In-pond 6ariation

Sample location within a pond was a minor source of variation for DO

consumption measurements in the pond tested. Similar findings of minor variation
in well-mixed ponds have been reported for measurements of carbon, nitrogen, and
phosphorous (Wei and Laws, 1989). In the present study, error due to within-pond
variation was controlled by sampling all ponds in similar locations. However, in
larger, less well-mixed commercial systems there may not be a single site that
provides adequate representation of the entire pond. In these cases, multiple
samples from a single pond may be required to obtain representative measurements.

4.3. Temperature and processing time effects

Sample incubation temperature significantly affected the DO consumption rate,

as was seen by Hargrave (1969). This suggests the importance of tight temperature
36 D. Brat6old, C.L. Browdy / Aquacultural Engineering 19 (1998) 29–39

control. In the field, temperature control may be easily achieved by incubation of

samples in a pond. The increased percent change in oxygen consumption due to
nutrient addition suggests that incubations at temperatures greater than pond
temperature may be preferable when the objective is rapid determination of water
column nutrient limitations.
Throughout this study, care was taken to minimize sample processing time. In
experiments with many samples, replicates processed within the first 20–30 min of
sample collection generally had slightly greater rates of oxygen consumption than
those processed within 70 – 80 min of sample collection. This observation may be
the result of rapid depletion of the most labile organic matter. Potential bias due to
holding samples prior to incubation can be minimized by rapid sample processing.
In experiments with many samples, replicates should be prepared throughout the
processing time so all samples will have replicates prepared early and late during
processing.

4.4. Photosynthesis

Table 1 depicts significant differences in photosynthesis rates between replicate

ponds, and temporally in the same pond. For comparisons of photosynthesis rates
between different ponds, samples must be incubated under similar light conditions.
This can be achieved by incubating bottles simultaneously, just below the surface of
the same pond to minimize variations in turbidity and shading. The simple method
used in this study does not estimate average pond photosynthesis rates, which
would require incubations at different pond depths to account for decreasing light
with depth. Furthermore, this method does not allow precise comparisons between
dates, which would require incubations in a constant light and temperature
chamber. However, field incubations at similar sun altitudes with similar cloud
cover may provide an index of phytoplankton activity. In this study, care was taken
to avoid excessive photosynthesis incubation times that may result in DO concen-
trations above 14 mg/l because DO measurement precision is reduced as samples
become increasingly supersaturated. It should be noted that DO methods for
measuring photosynthesis are not appropriate for estimating photosynthesis of
macrophytes due to difficulties in obtaining representative samples in a bottle, and
delayed oxygen release in some species.

4.5. General acti6ity

Water column oxygen consumption rates are typically primarily due to oxygen
use for respiration by heterotrophic and autotrophic plankton. Autotrophic plank-
ton include phytoplankton and nitrifying bacteria. When nitrification rate is
subtracted from the total oxygen consumption rate, the resulting rate approximates
the respiration rate. Heterotrophic respiration rates provide only general decompo-
sition rates because the amount of carbon dioxide released per unit oxygen
consumed varies with different nutrients (Gnaiger, 1983), and the nutrient pool
varies among sites and over time (Painting et al., 1989). Water column respiration
 D. Brat6old, C.L. Browdy / Aquacultural Engineering 19 (1998) 29–39 37

rates also include phytoplankton respiration which uses organic carbon produced in
the cell by photosynthesis. Depending upon the species present, a portion of
phytoplankton respiration may also be based upon the use of extracellular carbon
(Lewitus and Kana, 1995). Recognizing the various processes affecting water
column oxygen consumption rates, the methods in this study provide a general
index of microbial activity. In Table 2, the large differences in rates of oxygen
consumption on different dates in replicate ponds suggests temporal variation
within a pond, and suggests the need for multiple measurements throughout a
season to confidently identify treatment differences.
Five-day BOD measurements are collected at some aquaculture facilities for
regulatory purposes. Although short-term oxygen consumption measurements may
correlate with standard 5-day BOD measurements, 5-day BOD measurements do
not indicate the rate of organic matter breakdown or microbial activity in the pond
because the microbial community and labile organic matter concentration changes
during incubations of several days. Short-term incubations with linear rates of
oxygen consumption can indicate pond oxygen consumption rates and relative rates
of organic matter decomposition. High rates of oxygen consumption suggest higher
levels of microbial activity and nutrient cycling.

4.6. Nutrient additions

Results shown in Fig. 1 suggest that ammonia tended to be more limiting in June
and July than in August and September. Algal populations, as indicated by pond
color and turbidity, were greatest during the times when ammonia was most
limiting. The nitrogen demand of the larger June and July algal populations may
have contributed to greater ammonia limitation during these months. Ammonia is
generally the preferred nitrogen source for phytoplankton (Syrett, 1981) and has
been found to be inversely related to algal density in aquaculture ponds (Tucker et
al., 1984). Dissolved organic matter (DOM) from phytoplankton is thought to be
a major source of nutrients for bacteria in eutrophic estuaries (Bjornsen et al., 1989;
Coffin et al., 1993). Oxygen consumption rate responses to specific nutrient
additions may vary with shifts in algal DOM release and the types of bacteria that
dominate the community (Painting et al., 1989). Thus, when the phytoplankton
community is at a phase in its population cycle that releases less DOM, or types of
DOM that are not a digestible for the current bacterial community, additions of
labile carbon may increase bacterial respiration.
Changes in oxygen consumption rates in response to nutrient additions may
provide a useful tool for pond management strategies that include stimulation of
pond activity by addition of limiting nutrients such as nitrogen and phosphorous
(e.g. fertilizer) and carbon (e.g. molasses). In aquaculture ponds with daily addi-
tions of feed, DOM from feed may present a more constant substrate than algal
derived DOM. Thus, additions of limiting nutrients to aquaculture ponds in
response to algal population cycles may reduce bacterial community changes due to
phytoplankton cycles.
38 D. Brat6old, C.L. Browdy / Aquacultural Engineering 19 (1998) 29–39

4.7. Nitrification

Nitrification rates shown in Table 3 suggest that electrometric methods can be

used to measure differences in water column nitrification rates between ponds and
temporal changes within a pond. Sediment nitrification rates often exceed water
column nitrification (Focht and Verstraete, 1977; Blackburn and Henriksen, 1983;
Henriksen and Kemp, 1988), but are not measured with the methods of this study.
However, these simple methods may provide an index of nitrification activity.
Nitrification rates can be expressed either in terms of oxygen consumption, or as
ammonia oxidation. Complete ammonia oxidation is calculated based on 2 mol
oxygen consumed for every mole of ammonia oxidized. Estimation of the rate of
total ammonia nitrogen (TAN) oxidation in conjunction with TAN concentrations
suggests the importance of nitrifying bacteria for nitrogen cycling in the pond
system. The ratio of NBOD to BOD may also be of interest from a pond oxygen
management standpoint because it suggests the proportion of total oxygen con-
sumption that is due to nitrification.
Nitrification is commonly a limiting step in nitrogen cycling (Focht and
Verstraete, 1977). Rates of denitrification generally correspond to nitrification rates
(Jenkins and Kemp, 1984). Thus, greater rates of nitrification suggest more rapid
nitrogen cycling. From a pond management standpoint, high ammonium concen-
trations in conjunction with low nitrification rates indicate the need to develop
strategies to promote nitrification, such as habitat expansion or bioaugmentation of
nitrifying bacteria.
Overall, the electrometric methods tested in this study were able to provide
general measurements of pond microbial activity using the same equipment used to
conduct standard BOD tests. These methods can be easily adapted to field
conditions with relatively little investment in development and training. These
simple tools may assist in the examination of differences in microbial activity
between aquaculture systems and indicating additional methods for controlling this
important community.

Acknowledgements

The authors would like to thank the staff of the South Carolina Department of
Natural Resources’ Waddell Mariculture Center without whom this research would
not have been possible. This research was supported by the USDA/CSRS US
Marine Shrimp farming Program. This paper is contribution number 413 from the
South Carolina Marine Resources Center.

References

APHA (American Public Health Association), American Water Works Association, and Water Pollution
Control Federation, 1995. Standard Methods for the Examination of Water and Wastewater. 19th
ed., American Public Health Association, Washington, DC, pp. 5-2 to 5-6.
 D. Brat6old, C.L. Browdy / Aquacultural Engineering 19 (1998) 29–39 39

Bidwell, R.G.S., 1974. Plant Physiology. Macmillan, New York, pp. 167 – 171.
Bjornsen, P.K., Riemann, B., Pock-Steen, J., Nielsen, T.G., Horsted, S.J., 1989. Regulation of bacterio-
plankton production and cell volume in a eutrophic estuary. Appl. Environ. Microbiol. 55,
1512 – 1518.
Blackburn, T.H., Henriksen, K., 1983. Nitrogen cycling in different types of sediments from Danish
waters. Limnol. Oceanogr. 28, 477–493.
Coffin, R.B., Connolly, J.P., Harris, P.S., 1993. Availability of dissolved organic carbon to bacterio-
plankton examined by oxygen utilization. Mar. Ecol. Prog. Ser. 101, 9 – 22.
Focht, D.D., Verstraete, W., 1977. Biochemical ecology of nitrification and denitrification. Adv.
Microbiol. Ecol. 1, 135–214.
Fogg, G.E., 1962. Extracellular products. In: A.R.E. Lewin (Ed.), Physiology and Biochemistry.
Acedemic Press, New York, pp. 475-489.
Gnaiger, E., 1983. Calculation of energetic and biochemical equivalents of respiratory oxygen consump-
tion. In: Gnaiger, E., Forstner (Eds.), Polarographic Oxygen Sensors. Springer-Verlag, Berlin, pp.
337-345.
Hansen, L.S., Blackburn, T.H., 1992. Effect of algal bloom deposition on sediment respiration and
fluxes. Mar. Biol. 112, 147–152.
Hargrave, B.T., 1969. Epibenthic algal production and community respiration in the sediments of
Marion Lake. J. Fish. Res. Board Can. 26, 2003 – 2026.
Henriksen, K., Kemp, W.M., 1988. Nitrification estuarine and coastal marine sediments. In: Blackburn,
Sørensen (Eds.), Nitrogen Cycling in Coastal Marine Environments. John Wiley and Sons, Ltd.,
New York, pp. 207-249.
Ittekkot, V., 1982. Variations in dissolved organic matter during a plankton bloom: Qualitative aspects,
based on sugar and amino acid analysis. Marine Chem. 11, 143 – 158.
Jenkins, M.C., Kemp, W.M., 1984. The coupling of nitrification and denitrification in two estuarine
sediments. Limnol. Oceanogr. 29, 609–619.
Jensen, L.M., Sand-Jensen, K., Marcher, S., Hansen, M., 1990. Plankton community respiration along
a nutrient gradient in a shallow Danish estuary. Mar. Ecol. Prog. Ser. 61, 75 – 85.
Lewitus, A.J., Kana, T.M., 1995. Light respiration in six estuarine phytoplankton species: contrasts
under photoautotrophic and mixotrophic growth conditions. J. Phycol. 31, 754 – 761.
Oviatt, C.A., Keller, A.A., Sampou, P.A., Beatty, L.L., 1986. Patterns of productivity during eutrophi-
cation: a mesocosm experiment. Mar. Ecol. Prog. Ser. 28, 69 – 80.
Painting, S.J., Lucas, M.I., Muir, D.G., 1989. Fluctuations in heterotrophic bacterial community
structure, activity and production in response to development and decay of phytoplankton in a
microcosm. Mar. Ecol. Prog. Ser. 53, 129–141.
Sharp, J.H., 1975. Gross analysis of organic matter in seawater: Why, how, and where from. In: Church
(Ed.), Marine Chemistry in the Coastal Environment. ACS Symposium Series 18, American
Chemical Society, pp. 682-696.
Szyper, J.P., Rosenfield, J.Z., 1992. Diel cycles of planctonic respiration rtaes in briefly incubated water
samples from fertile earthen ponds. Limnol. Oceanogr. 37, 1193 – 1201.
Syrett, P.J. 1981. Nitrogen metabolism of microalgae. In: Platt, T.J. (Ed.), Physiological Bases of
Phytoplankton Ecology. Canadian Fisheries Research Board Bulletin 210, Ottowa, pp. 182-210.
Talling, F., 1973. The application of some electrochemical methods to the measurement of photosynthe-
sis and respiration in fresh water. Freshwater Biol. 3, 335 – 362.
Thomas, J.P., 1971. Release of dissolved organic matter from natural populations of marine phytoplank-
ton. Mar. Biol. 11, 311–323.
Tucker, C.S., Lloyd, S.W., Busch, R.L., 1984. Relationships between phytoplankton periodicity and the
concentrations of total and unionized ammonia in channel catfish ponds. Hydrobiologia 111, 75 – 79.
Wei, S.L., Laws, E.A., 1989. Spatial and temporal variation of water column measurements in
aquaculture ponds. Aquaculture 78, 253–266.