|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
QUANTIFYING THE MAGNITUDE OF NUTRIENT LIMITATION ON
PHYTOPLANKTON IN KINGS BAY, FLORIDA, USA
A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE
UNIVERSITY OF FLORIDA
This document is dedicated to my family: Mike, Jeannie, Anna, and Travina, and to the
past, present, and future generations of staff and graduate students of the Frazer Lab at
the University of Florida, especially Steph, Sky, Kelly, Kristin, Vince, Emily, and Ray.
I thank my committee members for their support and guidance. I also thank the
students and staff of the University of Florida's Department of Fisheries and Aquatic
Science for use of their equipment and their assistance in running this experiment.
Funding for this project was provided by the University of Florida's Institute of Food and
Agricultural Science and the Swisher Water Resources Graduate Research Assistantship.
TABLE OF CONTENTS
A C K N O W L E D G E M E N T S ..................................................................... .................... iv
LIST OF TABLES ............... .................................... vi
LIST OF FIGURE S ......... ..................................... ........... vii
A B S T R A C T .......................................... .................................................. v iii
1 IN TRODU CTION ................................................. ...... .................
2 ST U D Y L O C A T IO N ................................................................................ ............5
3 M E T H O D S ............................................................................. 7
Water Collection...............................7.... .........7
Laboratory Assays of Nutrient Limitation......... ................... ......... ...................7
C hlorophyll A naly sis .. .. ...... ............................................................ ...... ....... .. ..
N utrient Analysis ................................................................ ........ 10
Calculation of Growth Rates and Maximum Biomass ........................................... 10
Statistical A n aly sis ................................................. ...................10
N utrient Lim itation A ssessm ent ....................................................... ..... ........... 11
4 RESULTS ..................................... .................................. ........... 13
A m bient C condition s ................................................................ .. ...... ... ...... 13
Experim mental A ssay .................. ..................................... .. .......... .... 14
5 DISCUSSION .................. .................................... ........... .... .......... 23
Comparing Ambient and Experimental Results ............................... ................ 23
E xperim ental A ssay Im plications.................................................................... .......23
L IST O F R E FE R E N C E S ............................................................................. .............. 28
B IO G R A PH IC A L SK E TCH ..................................................................... ..................32
LIST OF TABLES
3-1 Experimental design with bioavailable N:P ratios at the start of the assays for
water collected from sites 23 (a) and 81(b)....................................................... 12
4-1 Summary of ambient water quality parameters for sites 23 and 81 ...........................16
4-2 Summary of statistical results for the overall effects of nitrogen and phosphorus on
phytoplankton growth rates and maximum biomass........................................17
4-3 Summary of statistical results for the effects of nitrogen and phosphorus on
phytoplankton maximum biomass at fixed levels of the other nutrient................... 18
LIST OF FIGURES
2-1 M ap of K ings B ay ................ .................................... ...... ........ ... .. .. ..6
4-1 Two dimensional growth rate response surface for site 23 .............. ..................19
4-2 Two dimensional growth rate response surface for site 81 ......................................20
4-3 Two dimensional maximum biomass response surface for site 23 ...........................21
4-4 Two dimensional maximum biomass response surface for site 81 ............................22
Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science
QUANTIFYING THE MAGNITUDE OF NUTRIENT LIMITATION ON
PHYTOPLANKTON IN KINGS BAY, FLORIDA, USA
Chair: Tom Frazer
Co-Chair: Craig Osenberg
Major Department: Fisheries and Aquatic Sciences
The recreational and economic viability of many aquatic systems is linked to water
quality in general and water clarity in particular. In cases where water clarity has been
impaired as a consequence of excessive phytoplankton production, a common
management strategy is to reduce the load of the limiting nutrient. In this study, nutrient
addition and dilution assays were used to quantify the magnitude of nutrient (nitrogen
and phosphorus) limitation, and characterize the two-dimensional response surface of
phytoplankton across a broad range of nutrient concentrations and N:P ratios. Site water
from two sites in Kings Bay, Florida (USA) was filtered using a Millipore stirred cell
concentrator, which allowed a fixed percentage of site water and nutrients to be removed
while maintaining ambient abundances of plankton. DI water and stock solutions of
nitrogen and phosphorus were added to return samples to their original volumes and
create nutrient concentration treatments that ranged from -12 to 10 times the ambient
concentration of phosphorus, and 1/4 to 15 times the ambient concentration of nitrogen.
Growth rates and maximum biomass were estimated using in vivo fluorescence measures.
The magnitude of limitation was quantified by estimating how much the response
variable changed per unit of nutrient concentration. Phytoplankton growth rates were not
limited by nitrogen or phosphorus at one site, but limited by both nitrogen and
phosphorus at the other site. The overall estimated magnitude of limitation for this latter
site was a 0.1011 change in per capital algal growth (d-1) for each mg L-1 of nitrogen, and
a 1.1268 change in per capital algal growth (d-1) for each mg L-1 of phosphorus. Neither
site showed significant nitrogen/phosphorus interaction effects on growth rate responses.
However, there was a significant interactive effect of nitrogen and phosphorus on
maximum biomass responses at both sites. A qualitative assessment of the response
surfaces for maximum biomass suggests that this variable is affected by both the N:P
ratio of nutrient treatments as well as the total amount of nutrients supplied. Since the
magnitude of limitation of one nutrient is determined by the concentration of the other
nutrient, an overall magnitude of limitation may either overestimate or underestimate the
expected maximum biomass response for specific co-occurring changes in nutrient
concentrations. Instead, magnitude of limitation should be determined on a case by case
basis, taking into consideration specific nutrient remediation scenarios and creating a
response surface with those co-occurring changes in nutrient concentrations in mind. The
combination of quantitative analyses and characterization of two-dimensional response
surfaces can be used by water resource managers as a tool to develop and implement
nutrient reduction strategies with predictable outcomes.
Eutrophication of aquatic ecosystems, as a consequence of increased nutrient
delivery, occurs worldwide. It can result in an increase in the frequency and intensity of
algal blooms and fish kills, as well as a decrease in biodiversity with losses of key species
such as rooted aquatic plants and corals (Grall and Chauvaud 2002, Baird et al. 2004,
Hauxwell et al. 2004, Lapointe et al. 2004, Miller and Stadelmann 2004). The resulting
decline in water quality and ecosystem function can have pronounced socioeconomic
impacts that negatively affect a large number of recreational, agricultural, and industrial
interests (Lund 1972, Carpenter et al. 1998, Smith et al. 1999, Tett et al. 2003). In
response to the many issues related to cultural eutrophication, there has been a concerted
effort on the part of scientists and water resource managers to improve water clarity and
reverse the effects of eutrophication. Nutrient reduction strategies, however, have
resulted in only limited success. In fact, many systems show no response to nutrient
reduction. Carpenter et al. (1999), for example, showed that the effects of prior nutrient
enrichment could be (1) reversible (recovery is immediate and proportional to the nutrient
reduction), (2) hysteretic (recovery requires extreme reductions in nutrient input for a
period of time), or (3) irreversible (recovery cannot be accomplished by reducing nutrient
input alone). To determine why hysteretic and irreversible systems do not respond to
nutrient remediation and reduction strategies, scientists and managers are now focused on
better understanding the complex suite of physical, chemical, and biological interactions
that affect nutrient uptake and assimilation.
An understanding of the relationship between phytoplankton abundance and
nutrient availability is vital to our ability to predict the effects of either nutrient increases
or reductions. This relationship is especially important for systems where phytoplankton
are one of the primary factors reducing water clarity. Of particular importance is the
ability to determine whether or not phytoplankton growth is limited by a particular
nutrient and identifying what that nutrient is. Nutrient limitation may be inferred from an
analysis of nutrient availability and elemental ratios in phytoplankton or bulk suspended
material in the water column. A more direct experimental evaluation can be made by
adding nutrients to a water sample and evaluating if the growth, biomass, or other
physiological parameter (e.g., respiration or carbon fixation rate) of phytoplankton differs
significantly from a control treatment (Gerhart and Likens 1975, Beardall et al. 2001).
Results from these types of studies can yield important insights to the current limiting
status of a system, but may be of limited utility to managers that need to make
quantitative predictions of a system's behavior to specific management actions.
There are three primary limitations of this simple experimental approach. First, it
focuses on two nutrient concentrations (added and control), which may or may not be
relevant to the situations of interest to the managers. Second, many studies are not able to
capture any complex interactions that may be occurring among nutrients. Some studies
focus on the manipulation of a single nutrient, whereas others have limited nutrient
treatment combinations (2x2 factorial with added and control treatments for each
nutrient). In either case, interactions among nutrients can not be characterized. By
manipulating >1 nutrient, all at a range of concentrations, multi-dimensional response
surfaces can be obtained and characterized to reveal potentially complex nutrient
interactions. Finally, the standard experimental approach focuses on the detection of
statistical differences between treatments, which may not reveal biologically important
effects which may or may not be statistically significant (e.g., Stewart-Oaten 1996).
Instead, quantitative estimates of the magnitude of limitation (e.g., Downing et al. 1999,
and Osenberg et al. 2002) may be more useful because these estimates reveal how much
systems might respond to changes in nutrients. As a result, scientists and managers can
directly test responses of phytoplankton to the range of nutrient concentrations they are
currently exposed to or may be exposed to in the future. This information may provide
managers with the detailed information necessary to implement policies that have an
increased probability of success.
This study introduces a method combining nutrient addition and dilution assays to
create a two-dimensional response surface, and thus quantify the magnitude of nitrogen
and phosphorus limitation. Both phytoplankton growth rates and maximum achieved
biomass were measured to better understand how phytoplankton might respond to
changes in nitrogen and phosphorus concentrations. Conceptually, there are four possible
qualitative outcomes of this experiment based on the two response variables: (1) nutrients
are never limiting, i.e. initial growth rates and maximum achieved biomass are similar
across all nutrient treatments, (2) nutrients are not limiting at first, but become limiting as
nutrients are depleted, i.e. initial growth rates are similar across treatments, but maximum
achieved biomass increases with increasing nutrient concentration, (3) the nutrient is
limiting at first, but some other factor consistent across treatments becomes limiting
when the culture reaches a particular biomass, i.e. initial growth rates increase with
increasing nutrient concentration, but maximum achieved biomass are similar across all
nutrient treatments, and (4) the nutrient is always limiting; i.e., both initial growth rates
and maximum achieved biomass increase with increasing nutrient concentration. The
potential effects of nutrient manipulation on phytoplankton growth and biomass have
seldom been quantified. Thus, the methodology and results reported herein will be of
interest to a broad group of scientists and water resource managers.
Kings Bay (Figure 2-1) is a tidally influenced system (ca. 1.75 km2; see Haller et
al. 1983) that is connected to the Gulf of Mexico by Crystal River on the west coast of
Florida, about 96-km north of Tampa (Frazer et al. 2001). Greater than 30 springs in
Kings Bay have a combined average total discharge of -27 m3 s-1 accounting for
approximately ninety-nine percent of the freshwater entering the system (Yobbi and
Knochenmus 1989, Hammett et al. 1996). Over 100,000 boaters, anglers, and sport divers
visit these springs each year (USFWS 2005). Kings Bay is within the confines of the
Crystal River National Wildlife Refuge and provides valuable habitat for a variety of
native and endangered species. Since 1999, greater than 250 West Indian manatees
(Trichechus manatus) take refuge annually in the relatively warm spring waters during
winter months (J. Kleen, USFWS, pers. comm.). This system is a recreationally,
economically, and ecologically important area that relies on maintaining water clarity and
functional habitats to continue being utilized in this manner.
As part of the Surface Water Improvement and Management (SWIM) program, one
of the primary management goals for Kings Bay is to improve water clarity and wildlife
habitat (SWFWMD 2003). Hoyer et al. (1997) found that the water clarity in Kings Bay
is primarily determined by algal suspended particles, and moderately affected by non-
volatile and detrital suspended particles. In an effort to reduce eutrophication, point
sources of nutrients (e.g. effluent from the City of Crystal River Sewage Treatment Plant)
were removed in 1992. As a result, average total phosphorus was reduced from 105 ug
L-1 to 27 ug L-1 and average total nitrogen was reduced from 620 ug L-1 to 220 ug L-1 in
Cedar Cove located in the northern region of Kings Bay (Terrell and Canfield 1996).
Water clarity, however, has not improved (Bishop and Canfield 1995, Hoyer et al. 1997,
0 0.5 1 Kilometers b1 f.l
Figure 2-1 Map of Kings Bay with sampling sites. There are > 30 springs located
throughout Kings Bay with the main spring for Kings Bay located in the
southern region, southeast of Site 81. Prior to 1992, the Crystal River Sewage
Treatment Plant discharged effluent into Cedar Cove in the northeast region of
Kings Bay. The dark grey/ black areas surrounding the bay correspond to
developed coastline, whereas the light grey areas correspond to marshland.
Two sites within Kings Bay were selected for water collection and subsequent
nutrient manipulation. These two sites, numbers 23 and 81 (see Figure 2-1), correspond
geographically to those established by Frazer and Hale (2001) as part of a longer-term
vegetative monitoring program (see Notestein et al. 2005). On October 11, 2004, surface
water samples (- 0.5 m depth) from both sites were collected and analyzed within 24h for
nitrate + nitrite (N03- + NO2-) and soluble reactive phosphorus (SRP) to determine
ambient concentrations. These values were then used to estimate the amount of stock
solutions needed to create designated treatments (see Table 3-1). The next day, water was
again collected at each site (20-L samples). Water samples were transported immediately,
in covered carboys, to the laboratory and allowed to set in the dark overnight in an
incubation room before initiation of the experiment the following morning.
Laboratory Assays of Nutrient Limitation
Phytoplankton growth rate (d-1) and maximum algal biomass, as indicated by in
vivo fluorescence (IVF) measures of chlorophyll-a, were determined over a gradient of
nitrogen and phosphorus concentrations. The experiment was set up as a fully crossed
factorial design, with 5 manipulated nitrogen and 5 manipulated phosphorus treatments.
Each treatment was replicated three times. Target phosphorus treatments corresponded
approximately to ambient SRP (X), 50% dilution (0.5 X), 25% dilution (0.75 X), 25%
addition (1.25 X), and 50% addition (1.5 X). Nitrogen treatments were chosen to
correspond to ambient nitrate + nitrite (X), 50% dilution (0.5 X), and additions that were
double (2 X), four times (4 X), and eight times (8 X) ambient, reflecting a particular
interest in and concern for increased nitrate in this and other spring-fed systems in Florida
(SWFWMD 2003). An additional control (run in triplicate) that lacked a nutrient
manipulation was used to determine ambient conditions and verify that the responses of
phytoplankton to experimental nutrient manipulation were realistic. Because the target
nutrient treatment levels were chosen based on water analyses run the previous day, the
ambient nutrient levels ("X") were not necessarily identical to the nutrient levels in the
ambient control treatment.
To establish the nutrient gradients, water collected from the two sites was filtered
using a Millipore stirred cell concentrator (model 8400), with Biomax-30 filters
(NMWL: 30,000). This allowed a pre-determined percentage of site water and nutrients
to be removed while maintaining ambient abundances of plankton. For all replicates
within a treatment, except the ambient control, a 400-ml aliquot of site water was filtered
to 50% of its original volume to control for filtration effects. The unfiltered portion, with
ambient concentration of plankton, was then placed in a 500-ml Erlenmeyer flask. DI
water and stock solutions of known nitrogen (KNO3) and phosphorus (K2HPO4)
concentration were added to return samples to the original volume of 400 ml and
establish the nutrient concentration treatments (see Table 3-1). For the ambient control,
site water was filtered to 50% of the original volume to control for filtration effects, and
the filtrate, rather than DI water, added back to the sample to maintain both ambient
nutrient and plankton concentrations. This control was used to determine ambient growth
rate and biomass, and provided a comparison to determine if there was an effect of
removing micronutrients from the experimental assays. Within 5 min of the nutrients
being added, 100 ml of sample water from each flask was withdrawn and later analyzed
for initial chlorophyll-a and nutrient concentrations. Nutrient measurements include total
nitrogen (TN), nitrate + nitrite (NO3- +NO2-), ammonium (NH4+), total phosphorus (TP),
and soluble reactive phosphorus (SRP). The nitrogen and phosphorus treatments reported
herein are the averaged initial nitrate + nitrite and SRP concentrations measured from the
All Erlenmeyer flasks were subsequently incubated in temperature-controlled water
baths with bottom illumination (-120 Em-2s-1). Incubation temperatures were maintained
at ambient temperatures (24C) recorded on the water collection date with a photoperiod
of 11/13 dark/light hours. Algal biomass was estimated using IVF measures at the time of
nutrient addition (time 0), and at 12 h intervals thereafter until IVF values in all
experimental flasks had peaked. IVF was measured using a Turner Designs Model 10
fluorometer with a 1-cm path length. At the end of the incubation period, water remaining
in the flasks was used for final IVF, chlorophyll-a, and nutrient measurements.
Initial and final chlorophyll-a measurements were obtained by filtering a minimum
of 50 ml of sample water through a 47-mm Whatman GF/F filter. Filters were stored
over desiccant and frozen prior to analysis. Chlorophyll-a was extracted in a 90% heated
ethanol solution, and chlorophyll-a concentrations were determined
spectrophotometrically using a Digilab Hitachi U-2810 spectrophotometer with a 4-cm
path length (APHA 1998).
Total nitrogen and total phosphorus concentrations were determined from whole
water samples digested with potassium persulfate. Ammonium, nitrate + nitrite, and SRP
were determined from water samples filtered through a 47-mm Whatman GF/F filter.
Total nitrogen, ammonium, and nitrate + nitrite analyses were determined
spectrophotometrically on a Bran-Luebbe autoanalyzer with a cadmium column
reduction method. Total phosphorus and SRP were determined spectrophotometrically on
a Digilab Hitachi U-2810 spectrophotometer with a 4-cm path length (APHA 1998).
Calculation of Growth Rates and Maximum Biomass
Phytoplankton growth rates (d-1) were estimated as ln(IVFt/IVFo)/t, where IVFo was
the initial fluorescence and IVFt was the fluorescence after t days, where t was
determined visually as the time period during which In(IVF) was an approximately linear
function of time. Maximum biomass was determined as the peak fluorescence measured
just before IVF began to decrease as the algal assemblage crashed. This peak sometimes
coincided with the end of exponential growth, but usually lagged by 0.5-1.5 days.
Because nitrogen and phosphorus treatments were not set at the same
concentrations for the two sites, an analysis of co-variance (ANCOVA, with [N] and [P]
as the covariates) was used to determine whether the two sites showed similar responses
to nutrient treatments. An analysis of variance (ANOVA) was then run for each site
individually to determine how nutrient treatments affected phytoplankton growth rates
and maximum biomass. For all analyses, the simplest model was used by removing
higher order treatment effects that were not significant; nutrient effects were determined
to be significant at a = 0.05 (SAS Institute 2004).
Nutrient Limitation Assessment
The magnitude of nutrient limitation is reflected as the change of the response
variable (i.e. growth rate or maximum biomass) for a given unit change of nitrogen or
phosphorus concentration. Graphically, the magnitude of limitation is the slope of the
relationship between growth rate (or maximum biomass) and nutrient concentration. If
there is no interaction between nitrogen and phosphorus, then the magnitude of limitation
for one nutrient can be expressed independently of the concentration of the other nutrient.
Where there was no nitrogen/phosphorus interactions, estimates of the treatment effects
determined by regression analysis were used as the overall estimate of magnitude of
limitation. If there is an interaction between N and P, then the magnitude of limitation for
one nutrient is influenced by the concentration of the other nutrient. Where there were
nitrogen/phosphorus interactions, estimates of treatment effects were also determined
using regression analysis for nutrients at each concentration level of the other nutrient.
Bioavailable N:P ratios at the start of the assays for water collected from sites
23 (a) and 81 (b). N:P ratios < 7 (shaded) suggest nitrogen limited conditions;
N:P ratios > 7 unshadedd) suggest phosphorus limited conditions. Nitrogen
and phosphorus treatments reported are the averaged beginning nitrate +
nitrite and SRP concentrations for each target treatment (n=15). Standard
deviations are given in parentheses and the experimental treatments closest to
ambient nutrient concentrations for each site are indicated with an asterisk.
(a) Site 23 Nitrogen (ug/L)
25 65 167 385 794
(7.0) (2.3) (11.1) (9.6) (20.6)
7(1.6) 3.57 9.29 23.86 55.00 113.43
CM 12(1.1) 2.08 *5.42 13.92 32.08 66.17
18 (1.0) 1.39 3.61 9.28 21.39 44.11
0 25(1.0) 1.00 2.60 6.68 15.40 31.76
32 (1.1) 0.78 2.03 5.22 12.03 24.81
(b) Site 81 Nitrogen (ug/L)
7 83 191 393 802
(3.8) (3.6) (8.1) (23.3) (42.4)
4 (1.4) 1.75 20.75 47.75 98.25 200.50
CM 9(1.2) 0.78 *9.22 21.22 43.67 89.11
16(1.1) 0.44 5.19 11.94 24.56 50.13
0 23 (1.1) 0.30 3.61 8.30 17.09 34.87
30 (1.1) 0.23 2.77 6.37 13.10 26.73
Sites 23 and 81 had similar temperature, salinity, dissolved oxygen, and pH, but
different nutrient concentrations (see Table 4-1). Ambient nitrate + nitrite and SRP
concentrations at site 23 (83 ug L-1 and 11 ug L-1, respectively) were approximately twice
those at site 81 (39 ug L-1 and 6 ug L-1, respectively). The bioavailable N:P ratio (ug:ug;
nitrate + nitrite:SRP) at site 23 was 7.6, while at site 81 the N:P ratio was 6.5. Both sites
23 and 81 were near the expected 7:1 N:P weight ratio for the chemical composition of
phytoplankton (Redfield 1934, Duarte 1992). This 7:1 N:P ratio is generally considered
the threshold where phytoplankton switch from nitrogen limitation (N:P < 7) to
phosphorus limitation (N:P > 7). Qualitative assessment of the plankton community
suggests that both sites were comprised predominately of small centric diatoms and
cryptophytes. There was, however, a lower concentration of chlorophyll-a at site 23
(10.92 ug L-) than at site 81 (14.59 ug L-1).
By using the model created by the experimental treatments, the ambient control
responses can be compared to the experimental treatment responses to determine if the
methodology may have affected the results. The maximum biomass achieved in the
ambient control assays were 0.8023 for site 23 and 0.6117 for site 81, as indicated by IVF
measures. By using the regression model created from the experimental assays (see Table
4-2), the maximum biomass (1SE) for site 23 was expected to be 0.7042 0.1118, and
site 81 was expected to be 0.7121 0.1187. Since the actual maximum biomass
measured in the ambient control assays fall within these ranges, the methodology did not
appear to bias the maximum biomass results. On the other hand, ambient growth rates for
phytoplankton at sites 23 and 81 were 0.3494 and 0.4836 d-1, respectively. Again, by
using the regression model created from the experimental assays (see Table 4-2), the
growth rate (1SE) for site 23 was expected to be 0.3019 0.0300, and site 81 was
expected to be 0.4318 0.0400. Growth rates measured in the ambient control assays
were greater than the expected, suggesting that the methodology employed may have
slightly biased the growth rate results.
Mean algal growth rates in the assays ranged from 0.2493 to 0.3406 d-1 for site 23,
and 0.3826 to 0.5349 d-1 for site 81 (Figures 4-1 and 4-2, respectively). The mean
maximum biomass achieved across treatments, as indicated by IVF measures, ranged
from 0.5667 to 1.0411 for site 23, and 0.6167 to 1.0179 for site 81 (Figures 4-3 and 4-4,
respectively). There was a significant site effect on growth rate (F2, 142 = 939.47, P <
0.0001) and maximum biomass (F2, 138 = 161.89, P < 0.0001). In addition, there was a
significant site and nitrogen treatment interaction effect (F1, 142= 5.18, P = 0.0243) on
growth rate, as well as significant nitrogen x phosphorus (F1, 138 = 13.99, P = 0.0003) and
nitrogen2 x phosphorus (F1, 138= 12.53, P = 0.0005) effects on maximum biomass. As a
consequence of the significant site effects, phytoplankton responses to nutrient treatments
were assessed for each site individually.
Nitrogen and phosphorus had significant effects on phytoplankton growth rates
only at site 81 (Table 4-2; Figures 4-1 and 4-2). The overall magnitude of nitrogen
limitation (+1SE) at this site was estimated as 0.1011 0.0160 L mg-1 d-1 (i.e. a 10%
increase in growth for each 1 mg L-1 increase in nitrogen concentration). Magnitude of
limitation for phosphorus (1SE), at site 81, was estimated as 1.1268 0.4866 L mg-' d-.
For site 23, the estimated magnitude of limitation (1SE) by both nitrogen (-0.0112 +
0.0120) and phosphorus (-0.1899 0.3792) included zero; hence the absence of a
significant effect of the nitrogen or phosphorus gradient.
With regard to the accumulation of phytoplankton biomass (Figures 4-3 and 4-4),
there was a significant interaction between nitrogen and phosphorus on the maximum
biomass response at both sites (see Table 4-2). Due to this interaction, estimates of the
treatment effect were determined for each nutrient at fixed levels of the other nutrient
(see Table 4-3). Treatment effects of phosphorus at fixed levels of nitrogen ranged from
-0.9333 to 9.9987 (IVF) L mg-1 for site 23, and -2.3460 to 10.2664 (IVF) L mg-1 for site
81. Treatment effects of nitrogen at fixed levels of phosphorus ranged from 0.4786 to
2.0827 (IVF) L mg-1 for site 23, and 0.2339 to 1.6421 (IVF) L mg-1 for site 81. These
results are consistent with the potential role of both nitrogen and phosphorus on
phytoplankton responses as indicated with the growth rate results above.
Qualitative assessment of the response surfaces for maximum biomass suggests
that this variable is affected by both the N:P ratio of nutrient treatments as well as the
total amount of nutrients supplied. Maximum biomass appeared to be maintained at low
concentrations when the N:P weight ratios were either very low (N:P < 3) or very high
(N:P > 27). At the lowest concentrations of each nutrient (i.e. 7 ug L-1 phosphorus and 25
ug L-~ nitrogen for site 23; 4 ug L-U phosphorus and 7 ug L-U nitrogen for site 81),
maximum biomass does not significantly change regardless of the concentration of the
other nutrient. When the N:P weight ratios were between 3 and 27, maximum biomass
appeared to increase with increasing nutrient availability.
Table 4-1 Summary of ambient water quality parameters for sites 23 and 81.
Temperature, salinity, dissolved oxygen, and pH were measured in the field at
the time of collection. All other parameters were taken from a subsample of
the ambient control treatments.
Temperature (C) 24.18 24.15
Salinity (ppt) 0.81 0.69
Dissolved Oxygen (mg L1) 6.3 6.99
pH 8.19 8.18
Total Nitrogen (ug L1) 183 147
Nitrate + Nitrite (ug L') 83 39
Ammonia (ug L"') 7 13
Total Phosphorus (ug L1) 27 27
SRP (ug L'1) 11 6
Chlorophyll-a (ug L1) 10.92 14.59
Predominant Species diat dit
Composition cryptophytes cryptophytes
Maximum Biomass (IVF) 0.8023 0.6117
Growth Rate (d-1) 0.3494 0.4836
Doubling rate (d) 2.0 1.4
Table 4-2 Factorial ANOVA results for sites 23 and 81 including estimates and standard
error (SE) of the treatment effects. Treatment effects are considered
significant at a < 0.05 (shaded). The simplest model was used for each
analysis by removing non-significant higher order treatment interactions. The
model for growth includes nitrogen (N) and phosphorus (P) treatment effects.
The model for maximum biomass also includes an interaction of nitrogen and
phosphorus (N*P), a quadratic term for nitrogen (N*N), and an interaction of
that quadratic term with phosphorus (N*N*P).
Treatment Effect Estimate of Effect SE F1, 72-value P-value
Site 23 N -0.0112 0.0120 0.88 0.3518
P -0.1899 0.3792 0.25 0.6180
Treatment Effect Estimate of Effect SE F1, 72-value P-value
Site 81 N 0.1011 0.0160 40.05 < 0.0001
P 1.1268 0.4866 5.36 0.0234
Treatment Effect Estimate of Effect SE F1, 69-value P-value
N 0.2088 0.4470 0.22 0.6419
Site 23 N*N -0.0898 0.5256 0.03 0.8648
P -3.8406 2.7967 1.89 0.1741
N*P 56.7800 21.4759 6.99 0.0101
N*N*P -64.8845 25.2536 6.60 0.0124
Treatment Effect Estimate of Effect SE F1, 69-value P-value
N 0.1025 0.3821 0.07 0.7894
Site 81 N*N -0.1015 0.4461 0.05 0.8206
P -2.3749 2.7789 0.73 0.3957
N*P 54.0362 20.2441 7.12 0.0095
N*N*P -58.0074 23.6320 6.03 0.0166
Table 4-3 ANOVA regression analysis results for sites 23 and 81 including estimates and
standard error (SE) of the treatment effects of one nutrient at fixed levels of
the other nutrient. Treatment effects are considered significant at ca 0.05
(shaded). Treatment effects include nitrogen (N) and the quadratic term for
nitrogen (N*N) at fixed phosphorus levels (ug L-1); and phosphorus (P)
treatment effects at fixed nitrogen levels (ug L1).
Phosphorus Treatment Effect Estimate of Effect SE F1,s1-value P-value
7 N 0.4786 0.3469 1.9 0.1928
N*N -0.3931 0.4079 0.93 0.3542
12 N 0.954 0.3195 8.92 0.0114
N*N -0.8705 0.3757 5.37 0.0390
18 N 1.4974 0.4520 10.97 0.0062
N*N -1.701 0.5315 10.24 0.0076
25 N 1.3684 0.5300 6.67 0.0240
N*N -1.3589 0.6232 4.75 0.0499
32 N 2.0827 0.5059 16.95 0.0014
N*N -2.2249 0.5949 13.99 0.0028
Nitrogen Treatment Effect Estimate of Effect SE F1,s1-value P-value
25 P 1.9124 1.3931 1.88 0.193
65 P -4.9333 1.8949 6.78 0.0219
167 P 2.6687 3.3061 0.65 0.4341
385 P 9.9987 5.2922 3.57 0.0814
794 P 0.0392 2.1218 0.00 0.9856
Phosphorus Treatment Effect Estimate of Effect SE F1,a1-value P-value
4 N 0.2518 0.2260 1.24 0.287
N*N -0.2660 0.2638 1.02 0.3332
9 N 0.2339 0.3223 0.53 0.482
N*N -0.0613 0.3762 0.03 0.8733
16 N 1.7768 0.3537 25.23 0.0003
N*N -2.1123 0.4129 26.17 0.0003
23 N 1.0387 0.4317 5.79 0.0331
N*N -1.2083 0.5040 5.75 0.0337
30 N 1.6421 0.5455 9.06 0.0109
N*N -1.6164 0.6368 6.44 0.026
Nitrogen Treatment Effect Estimate of Effect SE F113-value P-value
7 P -2.346 1.2318 3.63 0.0792
83 P -0.2692 3 0.01 0.9227
191 P 10.2664 2.9691 11.96 0.0042
393 P 7.428 4.5433 2.67 0.126
802 P 4.0152 3.8804 1.07 0.3197
Figure 4-1 Growth rates for site 23 at given phosphorus and nitrogen treatments (SE +
0.0300). There were no significant nitrogen or phosphorus effects on
phytoplankton growth rates (see Table 3-2). Colors represent growth rates
within the indicated interval and are not intended to imply statistically
16 M Nitrogen
Phosphorus 23 30 ogen
(ug L)ug L
Figure 4-2 Growth rates for site 81 at given phosphorus and nitrogen treatments (SE +
0.0400). There was a significant nitrogen (F1, 72 = 40.05, P < 0.0001), and
phosphorus (Fi, 72= 5.36, P = 0.0234) effect on phytoplankton growth rates
(Table 3-2). Magnitude of limitation was estimated as a 0.1011 change in
growth rate (d-1) for each mg L-1 change in nitrogen, and an estimated 1.1268
change in growth rate (d-1) for each mg L-1 change in phosphorus. Colors
represent growth rates within the indicated interval and are not intended to
imply statistically significant differences.
Figure 4-3 Maximum biomass, as indicated by in vivo fluorescence (IVF), for site 23 at
given phosphorus and nitrogen treatments (SE + 0.1118). There was a
significant nitrogen x phosphorus interaction effect (Fl, 69 = 6.99, P = 0.0101)
as well as a significant nitrogen2 x phosphorus interaction effect (Fl, 69 = 6.60,
P = 0.0124; Table 3-2). Colors represent maximum biomass within the
indicated interval and are not intended to imply statistically significant
162 c Nitrogen
Phosphorus 23 0 rogen
(ug L1) (ug L1)
Figure 4-4 Maximum biomass, as indicated by in vivo fluorescence for site 81 at given
phosphorus and nitrogen treatments (SE + 0.1187). There was a significant
nitrogen x phosphorus interaction effect (Fl, 69 = 7.12, P = 0.0095) as well as a
significant nitrogen2 x phosphorus interaction effect (F1, 69 = 6.03, P = 0.0166;
Table 3-2). Colors represent maximum biomass within the indicated interval
and are not intended to imply statistically significant differences.
Comparing Ambient and Experimental Results
Maximum biomass did not appear to be affected by the methodology used in this
study, the actual maximum biomass values measured in the ambient control were within
the ranges predicted by the models created from the experimental assays. There was,
however, a discrepancy between the ambient control and the expected values from the
experimental assay models for growth rate responses; the models underestimated the
actual responses of phytoplankton growth rates for both sites. This discrepancy may be
due to the removal of micronutrients during the filtration process. Further
experimentation would have to be done to evaluate this hypothesis and determine what
other nutrients, if any, limit phytoplankton growth rates in Kings Bay.
Experimental Assay Implications
The results of this study provide important insights into how phytoplankton growth
rates and biomass are affected by nitrogen and phosphorus in Kings Bay. Phytoplankton
growth rates were not necessarily directly correlated with maximum biomass, i.e. growth
rate did not determine the maximum achieved biomass, just how long it took to reach that
biomass. The phytoplankton growth rate responses were spatially heterogeneous. Growth
rates were either affected by both nitrogen and phosphorus (site 81) or were not affected
by nitrogen or phosphorus (site 23). In addition, there was no significant
nitrogen/phosphorus interaction, indicating that the N:P ratio did not have a strong
influence on phytoplankton growth rates. On the other hand, maximum biomass was
primarily affected by interactions between nitrogen and phosphorus, which appeared to
result from responses to both the amount of nutrients and their relative abundances.
The difference in minimum and maximum values for algal growth rates (d-1) at
sites 23 and 81 were 0.09 and 0.15, respectively. These nutrient-induced changes in
growth rate are similar to those reported in other studies. For example, Downing et al.
(1999) reported that algal assemblages responded with changes in growth rates that
ranged from 0.0 0.2 d-1 following the addition of surplus nutrients, while Elser and
Frees (1995) reported changes in growth rates of 0.02-0.25 d-1 with the addition of
nutrients. The range in growth rates (0.25 0.53 d-1) and the range in chlorophyll-a
measured from the initial and final treatment subsamples (0.40-51.44 ug L1) in this study
are not unusual when compared to those reported for other systems. Growth rates (d-1)
reported in the literature surveyed ranged from 0.10 0.80 (Elser and Frees 1995, Hein
and Riemann 1995, Phlips et al. 2002), and Frazer et al. (2004) report chlorophyll-a
values that range from 0.2 74.4 ug L-1 in estuarine systems adjacent to three counties
along Florida's west coast.
Characterizing the two-dimensional response surfaces created in this study both
qualitatively and quantitatively, can provide valuable information in identifying ways of
managing nitrogen and phosphorus input to achieve management goals. For systems
where algal biomass is a primary contributor to water clarity, like Kings Bay, the
maximum biomass results of this study could be used to calculate the algal biomass that
could be expected given a particular nutrient reduction strategy. Currently, nutrient
concentrations in Kings Bay are relatively low in comparison to other coastal, spring-fed
systems in Florida (see Frazer et al. 2001). However, nutrient concentrations (nitrate in
particular) are elevated compared to the presumed historical values (see Jones and
Upchurch 1994). As point sources of nutrient load to the Kings Bay system have been
removed, additional nutrient reduction will require reduction of non-point sources. This
may prove to be a difficult task considering the rapid rate of population growth and the
associated changes in land-use within the broader Kings Bay watershed (Jones and
Upchurch 1994, SWFWMD 2003). This being the case, managers seeking to achieve an
increase in water clarity in Kings Bay may have to focus on other factors that influence
the accumulation of phytoplankton biomass such as removal processes.
In shallow estuarine systems like Kings Bay, phytoplankton are typically removed
by physical processes, e.g., flushing, and potentially by grazing (Phlips et al. 2002).
Although little is known about grazing effects in Kings Bay, there have been modeling
studies that characterize circulation patterns and calculate flushing rates. Open waters of
the bay are thought to have relatively short particle residence times, between 50-59 hrs (2
- 2.5 days), while dye injection studies suggest that Kings Bay is flushed every 71-94 hrs
(3 4 days), depending on the magnitude of spring discharge (Hammett et al. 1996).
Ambient phytoplankton growth rates measured in this study suggest that phytoplankton
can double in as little as 1.4 days. At present, phytoplankton doubling times are close to
the flushing rate. It may be possible to reduce phytoplankton growth rates to the point
where doubling times are greater than flushing rates, thereby potentially reducing the
accumulation of phytoplankton biomass within the bay. The magnitude of nutrient
limitation can be used to determine the reduction in nutrient concentration necessary to
achieve the desired response in phytoplankton growth rates. For example, the overall
magnitude of nitrogen limitation for site 81 estimates a 0.1011 change in growth rate per
mg L-1 of nitrogen. To increase doubling times from 1.4 days to 2 days, phytoplankton
growth rate would have to decrease from 0.4836 to 0.3466. The magnitude of nitrogen
limitation suggests that site 81 would have to undergo a 1.4-mg L-1 reduction of nitrogen
if flushing rates are the only factor considered to remove phytoplankton. However, the
nitrogen concentration in Kings Bay is already so low (TN -0.25 mg L-1, Hoyer et al.
1997), that a 1.4-mg L-1 reduction in nitrogen is not possible. A similar calculation with
phosphorus indicates that a decrease in phytoplankton growth rate from 0.4836 to 0.3466
at site 81 would require an estimated 0.12-mg L-1 reduction of phosphorus. The average
TP in Kings Bay is -0.029 mg L-1 (Hoyer et al. 1997), thus eliminating the reduction of
phosphorus as a possibility. Instead a more realistic approach to improving water clarity
in Kings Bay may include a combination of nutrient remediation strategies for reducing
phytoplankton as well as other means of improving water clarity such as sediment
stabilization or increased water flow.
An important factor to consider when using estimates like the ones calculated in
this study is the spatial and temporal variation of phytoplankton responses. When making
management decisions, it is not only important to understand what factors are affecting
the biogeochemical processes of the system, but where and how these processes are
affected. Although only two sites were used in this study, the results suggest that different
areas within the same system may be controlled by different limiting factors. This
phenomenon is not restricted to Kings Bay; Lake Okeechobee and Chesapeake Bay have
also been found to show similar spatial heterogeneity of nutrient limitation (Aldridge et
al. 1995, Fisher et al. 1999). It is important that these patterns, along with seasonal and
annual variations are taken into consideration in order to make effective management
decisions (Vanni and Temte 1990, Lewitus et al. 1998).
LIST OF REFERENCES
Aldridge, F. J., E. J. Phlips, and C. L. Schelske. 1995. The use of nutrient enrichment
bioassays to test for spatial and temporal distribution of limiting factors affecting
phytoplankton dynamics in Lake Okeechobee, Florida. Archives Hydrobiologica
Special Issues on Advanced Limnology 45: 177-190.
American Public Health Association (APHA). 1998. Standard methods for the
examination of water and wastewater. 20th Ed. American Public Health
Association, Washington, DC.
Baird, D., R. R. Christian, C. H. Peterson, and G. A. Johnson. 2004. Consequences of
hypoxia on estuarine function: energy diversion from consumers to microbes.
Ecological Applications 14(3): 805-822.
Beardall, J., E. Young, and S. Roberts. 2001. Approaches for determining phytoplankton
nutrient limitation. Aquatic Sciences 63: 44-69.
Bishop, J. H., and D. E. Canfield. 1995. Volunteer water quality monitoring at Crystal
River, Florida (August 1992 August 1995). Final Report. Southwest Florida
Water Management District, Brooksville, FL.
Carpenter, S. R., N. F. Caraco, D. L. Correll, R. W. Howarth, A. N. Sharpley, and V. H.
Smith. 1998. Nonpoint pollution of surface waters with phosphorus and nitrogen.
Ecological Applications 8(3): 559-568.
Carpenter, S. R., D. Ludwig, and W. A. Brock. 1999. Management of eutrophication of
lakes subject to potentially irreversible change. Ecological Applications 9(3): 751-
Downing, J. A., C. W. Osenberg, O. Sarnelle. 1999. Meta-analysis of marine nutrient-
enrichment experiments: variation in the magnitude of nutrient limitation. Ecology
Duarte, C. M. 1992. Nutrient concentration of aquatic plants: patterns across species.
Limnology and Oceanography 37(4): 882-889.
Elser, J. J., and D. L. Frees. 1995. Microconsumer grazing and sources of limiting
nutrients for phytoplankton growth: application and complications of a nutrient-
deletion/dilution-gradient technique. Limnology and Oceanography 40(1): 1-16.
Fisher, T. R., A. B. Gustafson, K. Sellner, R. Lacouture, L. W. Haas, R. L. Wetzel, R.
Magnien, D. Everitt, B. Michaels, and R. Karrh. 1999. Spatial and temporal
variation of resource limitation in Chesapeake Bay. Marine Biology 133: 763-778.
Frazer, T. K., and J. A. Hale. 2001. An atlas of submersed aquatic vegetation in Kings
Bay (Citrus County, Florida). Southwest Florida Water Management District,
Brooksville, FL. Report 99CON000041. pp. 14.
Frazer, T. K., M. V. Hoyer, S. K. Notestein, J. A. Hale, and D. E. Canfield. 2001.
Physical, chemical and vegetative characteristics of five Gulf coast rivers. Final
Project Report. Southwest Florida Water Management District, Brooksville, FL.
Frazer, T. K., S. K. Notestein, C. A. Jacoby, and J. A. Hale. 2004. Water quality
characteristics of the nearshore gulf coast waters adjacent to Citrus, Hernando, and
Levy Counties: Project COAST 1997 to 2003. Annual Report. Southwest Florida
Water Management District, Brooksville, FL.
Frazer, T. K., E. J. Phlips, S. K. Notestein, and C. Jett. 2002. Nutrient limiting status of
phytoplankton in five gulf coast rivers and their associated estuaries. Final Report.
Southwest Florida Water Management District, Brooksville, FL.
Gerhart, D. Z., and G. E. Likens. 1975. Enrichment experiments for determining nutrient
limitation: four methods compared. Limnology and Oceanography 20(4): 649-653.
Grall, J., and L. Chauvaud. 2002. Marine eutrophication and benthos: the need for new
approaches and concepts. Global Change Biology 8: 813-830.
Haller, W. T., J. V. Shireman, and D. E. Canfield, Jr. 1983. Vegetation and herbicide
monitoring study in Kings Bay, Crystal River, FL. U.S. Army Corps of Engineers,
Jacksonville, FL. Report DACW17-80-C-0062. pp. 169.
Hammett, K. M., C. R. Goodwin, and G. L. Sanders. 1996. Tidal-flow, circulation, and
flushing characteristics of Kings Bay, Citrus County, Florida. U. S. Geological
Survey, Tallahassee, FL. Report 96-230. pp. 63.
Hauxwell, J., C. W. Osenberg, and T. K. Frazer. 2004. Conflicting management goals:
manatees and invasive competitors inhibit restoration of a native macrophyte.
Ecological Applications 14(2): 571-586.
Hein, M., and B. Riemann. 1995. Nutrient limitation of phytoplankton biomass or growth
rate: an experimental approach using marine enclosures. Journal of Experimental
Marine Biology and Ecology 188: 167-180.
Hoyer, M. V., L. K. Mataraza, A. B. Munson, and D. E. Canfield, Jr. 1997. Water clarity
in Kings Bay/ Crystal River. Southwest Florida Water Management District,
Brooksville, FL. pp.22.
Jones, G. W., and S. B. Upchurch. 1994. Origin of nutrients in ground water discharging
from the Kings Bay springs. Southwest Florida Water Management District,
Brooksville, FL. pp. 120.
Lapointe, B. E., P. J. Barile, and W. R. Matzie. 2004. Anthropogenic nutrient enrichment
of seagrass and coral reef communities in the Lower Florida Keys: discrimination
of local versus regional nitrogen sources. Journal of Experimental Marine Biology
and Ecology 308: 23-58.
Lewitus, A. J., E. T. Koepfler, and J. T. Morris. 1998. Seasonal variation in the regulation
of phytoplankton by nitrogen and grazing in a salt-marsh estuary. Limnology and
Oceanography 43(4): 636-646.
Lund, J. W. G. 1972. Eutrophication. Proceedings of the Royal Society of London. Series
B, Biological Sciences 180(1061): 371-382.
Miller, R., and P. Stadelmann. 2004. Fish habitat requirements as the basis for
rehabilitation of eutrophic lakes by oxygenation. Fisheries Management and
Ecology 11: 251-260.
Notestein, S. K., T. K. Frazer, S. R. Keller, and R. A. Swett. 2005. Kings Bay Vegetation
Evaluation 2004. Southwest Florida Water Management District, Brooksville, FL.
Osenberg, C. W., C. M. St. Mary, J. A. Wilson, and W. J. Lindberg. 2002. A quantitative
framework to evaluate the attraction-production controversy. ICES Journal of
Marine Science 59: S214-S221.
Phlips, E. J., S. Badylak, and T. Grosskopf. 2002. Factors affecting the abundance of
phytoplankton in a restricted subtropical lagoon, the Indian River Lagoon, Florida,
USA. Estuarine, Coastal, and Shelf Science 55: 385-402.
Redfield, A. C. 1934. On the proportions of organic derivatives in sea water and their
relation to the composition of plankton. In: James Johnstone Memorial Volume.
(pp. 176-192) University Press, Liverpool, UK.
Statistical Analysis System (SAS) Institute. 2004. SAS users guide. SAS Institute, Inc.
Cary, North Carolina.
Smith, V. H., G. D. Tilman, and J. C. Nekola. 1999. Eutrophication: impacts of excess
nutrient inputs on freshwater, marine, and terrestrial ecosystems. Environmental
Pollution 100: 179-196.
Stewart-Oaten, A. 1996. Problems in the analysis of environmental monitoring data. In
Detecting ecological impacts: concepts and applications in coastal habitats, pp.
109-131. Ed. By R. J. Schmitt, and C. W. Osenberg. Academic Press, San Diego,
USA. 401 pp.
Southwest Florida Water Management District (SWFWMD). 2003. Crystal River/ Kings
Bay Summary Report. Southwest Florida Water Management District, Brooksville,
FL. pp. 33.
Terrell, J. B., and D. E. Canfield, Jr. 1996. Evaluation of the effects of nutrient removal
and the "Storm of the Century" on submersed vegetation in Kings Bay Crystal
River, Florida. Journal of Lake and Reservoir Management 12(3): 394-403.
Tett, P., L. Gilpin, H. Svendsen, C. P. Erlandsson, U. Larsson, S. Kratzer, E. Fouilland,
C. Janzen, J. Lee, C. Grenz, A. Newton, J. G. Ferreira, T. Fernandes, and S. Scory.
2003. Eutrophication and some European waters of restricted exchange.
Continental Shelf Research 23: 1635-1671.
United States Fish and Wildlife Service (USFWS). 2005. Crystal River National Wildlife
Refuge. http://crystalriver.fws.gov/ Last assessed August 2005.
Vanni, M. J., and J. Temte. 1990. Seasonal patterns of grazing and nutrient limitation of
phytoplankton in a eutrophic lake. Limnology and Oceanography 35(3): 697-709.
Yobbi, D. K., and L. A. Knochenmus. 1989. Effects of river discharge and high-tide stage
on salinity intrusion in the Weeki Wachee, Crystal, and Withlacoochee river
estuaries, southwest Florida. U. S. Geological Survey, Tallahassee, Florida. Report
I am originally from St. Louis, MO. I moved to Florida to get my B.S. in marine
science at Eckerd College in St. Petersburg, FL. For my undergraduate thesis, I studied
the historic spatial and temporal changes in submerged aquatic vegetation in Tampa Bay,
FL, using aerial photography. After graduating, I spent a year working in the chemistry
lab at the Florida Wildlife Research Institute in St. Petersburg, FL. I then came to the
University of Florida to earn my M.S. under the advisement of Dr. Tom Frazer at the
Department of Fisheries and Aquatic Sciences.