Freshwater Biology (2008) 53, 264–277
doi:10.1111/j.1365-2427.2007.01890.x
Differential response of phytoplankton to additions
of nitrogen, phosphorus and iron in Lake Tanganyika
A A I K E D E W E V E R * , K O E N R A A D M U Y L A E R T †, D E N I S L A N G L E T ‡, L A U R E N T A L L E M A N ‡, – ,
J E A N - P I E R R E D E S C Y § , L U C A N D R É ‡, C H R I S T I N E C O C Q U Y T * , * * A N D W I M V Y V E R M A N *
*Laboratory for Protistology and Aquatic Ecology, Department Biology, Ghent University, Ghent, Belgium
†
K.U.Leuven-Campus Kortrijk, Kortrijk, Belgium
‡
Musée Royal de l’Afrique Centrale, Section de Pétrographie-Minéralogie-Géochimie, Tervuren, Belgium
§
Laboratory of Freshwater Ecology, Department of Biology, University of Namur, URBO, Namur, Belgium
–
Département Chimie et Environnement, Ecole des Mines de Douai, Douai, France
**National Botanic Garden of Belgium, Meise, Belgium
SU M M A R Y
1. In order to evaluate limitation of different phytoplankton groups by inorganic nutrients,
multiple nutrient enrichment bioassays using the addition of iron (Fe) and the combined
addition of nitrogen and phosphorus (NP) were carried out in the north and the south of
Lake Tanganyika during the rainy and dry seasons in 2003 and 2004.
2. Nutrient additions resulted in an increase in phytoplankton growth rate relative to
control treatments in all experiments. HPLC pigment data and epifluorescence microscopy
counts indicated differential stimulation of the dominant phytoplankton groups. Iron
additions mainly stimulated prokaryotic picophytoplankton, while enrichments with
nitrogen and phosphorus stimulated green algae and in some cases diatoms. Extended
incubation (3 days) indicated co-limitation of Fe and NP, in particular for picocyanobacteria.
Keywords: iron, Lake Tanganyika, nutrient limitation, phytoplankton, picocyanobacteria
Introduction
Phytoplankton primary productivity in aquatic ecosystems has long been known to be potentially limited
by the availability of the macronutrients nitrogen (N)
and phosphorus (P). Since the early 1990s, evidence
has been accumulating that, in large parts of the
oceans, phytoplankton primary production can also
be limited by the micronutrient iron (Fe) (Martin &
Gordon, 1988; Jickells et al., 2005). Iron limitation is
not restricted to marine environments, and Schelske,
Hooper & Heartl (1962) had already found that
phytoplankton photosynthesis in hardwater calcareous lakes was stimulated by Fe additions. More
Correspondence: Wim Vyverman, Laboratory for Protistology
and Aquatic Ecology, Ghent University, Krijgslaan 281 S8,
9000 Ghent, Belgium. E-mail: [email protected]
264
recently, Fe limitation of phytoplankton communities
has been demonstrated in nutrient bioassays in high
phosphorus, saline prairie lakes (Evans & Prepas,
1997), in acidic Canadian Shield lakes (Auclair, 1995),
and in the Great Lakes Erie (Twiss, Auclair &
Charlton, 2000a) and Superior (Sterner et al., 2004).
While in the ocean limitation by iron usually occurs
only at very low concentrations (<0.2 nM , Johnson,
Gordon & Coale, 1997), bioassays in lakes suggest that
Fe limitation may occur over a much wider range of
concentrations (<2–140 nM ), depending on the bioavailability of Fe (Chang, Kuwabara & Pasilis, 1992;
Auclair, 1995; Imai, Fukushima & Matsushige, 1999;
Twiss et al., 2000a; Sterner et al., 2004). Bioavailability
of Fe in lakes may be restricted by high pH (e.g.
Kuma, Nishioka & Matsunaga, 1996; Evans & Prepas,
1997) or by high concentrations of humic substances
(Imai et al., 1999). High pH and redox potential may
reduce Fe availability by high rates of oxidation of
2007 The Authors, Journal compilation 2007 Blackwell Publishing Ltd
Differential response of phytoplankton
bioavailable ferrous Fe2+ to ferric Fe3+, followed by the
formation of the highly insoluble ferric hydroxide
(Stumm & Morgan, 1996), while the presence of humic
substances enhances organic complexation of Fe3+.
Iron limitation not only affects primary production
of the phytoplankton community as a whole, but may
influence specific groups of phytoplankton differentially. In oceanic environments, large-scale Fe additions have almost always resulted in the development
of diatom blooms (e.g. Cavender-Bares et al., 1999;
Coale et al., 2004, respectively, in the Equatorial and
the Southern Oceans). During Fe enrichment bioassays in lakes, most studies report the stimulation of
one or more specific phytoplankton groups or size
classes. In Lake Erie, pico- (0.2–2 lm) and nanoplankton (2–20 lm) were stimulated by Fe additions (Twiss
et al., 2000a). In a more recent paper, Twiss et al.
(2005) found indications of periodically stimulatory
effects of trace metal enrichment, especially on the
picoplankton size class. In Canadian shield lakes,
picocyanobacteria and mixotrophic chrysophytes
were stimulated by Fe additions (Auclair, 1995). In
studies of eutrophic lakes (Lake Erken in Sweden and
Lake Kasumigaura in Japan), Fe additions were found
to favour colonial cyanobacteria (Imai et al., 1999;
Hyenstrand et al., 2001). This situation, where one
component of the phytoplankton community is stimulated by nutrient addition, may point to community
co-limitation sensu Arrigo (2005). Community
co-limitation is invoked when, for example, two
phytoplankton species or groups are each limited by
a different nutrient.
Lake Tanganyika is the second deepest lake on
earth (max. depth 1470 m) and the third largest by
volume. The lake is considered oligotrophic and has
average epilimnial concentrations of dissolved N and
P, respectively, around 3.6 and 0.5 lM (Plisnier et al.,
1999). The lake is permanently stratified but monsoon
winds cause seasonal erosion of the thermocline,
resulting in increased nutrient concentrations in the
epilimnion (Hecky, Spigel & Coulter, 1991; Langenberg, Sarvala & Roijackers, 2003b). Because of permanent stratification and the lake’s hydrodynamics, no
oxygen is present at depths below 100–150 m in the
north and 200 m in the south (Coulter & Spigel, 1991).
These anoxic conditions have important implications
for Fe availability due to the formation and precipitation of FeS. Hecky et al. (1991) consider vertical
mixing as the main source of P for the epilimnion
265
while cyanobacterial nitrogen fixation would be the
main source of N. Nutrient addition assays with N
and P and particulate nutrient ratios, have indicated a
fairly balanced nutrient budget with a tendency for P
to be more frequently limiting (Järvinen et al., 1999).
Potential (co-)limitation because of Fe has not been
investigated in this lake. In Lake Malawi, however,
a bioassay study indicated that phytoplankton was colimited by Fe, N and P, while similar experiments in
Lake Victoria demonstrated a stimulation of nitrogen
fixation as a result of Fe addition (Guildford et al.,
2003). Potential sources of Fe in Lake Tanganyika
include terrestrial input through aeolian dust, rainfall
and riverine inflow, and internal recycling through
deep water mixing and upwelling. As in Lake
Malawi, the highly laterized catchment of Lake
Tanganyika does not suggest limitation in the Fe
supply, but rather phytoplankton growth limitation
due to the low bioavailability of Fe because of the high
pH (8.4–9.2) and alkalinity (6.36–6.81 meq L)1; Talling
& Talling, 1965) (e.g. Wetzel, 2001).
We investigated the response of different phytoplankton groups to additions of Fe, as well as to the
combined addition of N and P, in in situ incubation
experiments. Experiments were carried out in contrasting seasons in the north and the south of the lake.
We compared the response of different phytoplankton
groups to the nutrient additions using both cell counts
and HPLC pigment analysis.
Methods
Study site
Lake Tanganyika is part of the East African rift valley
and is bordered by Burundi, Tanzania, Zambia and
D. R. Congo. The lake measures roughly 650 by 50 km
and has a maximum depth of 1470 m. During the dry
season (May to October), south-east monsoon winds
result in an erosion of the thermocline and upwelling
of nutrient rich deep water in the southern part of the
lake. When the monsoon winds stop (October), the
thermocline tilts to the north and further oscillates
between S and N (October to April) until maximum
stability is reached at the end of the rainy season
(January to April). These internal oscillations (or
waves) have been suggested to result in nutrient
inputs to the epilimnion of the lake (Coulter & Spigel,
1991; Naithani, Deleersnijder & Plisnier, 2002).
2007 The Authors, Journal compilation 2007 Blackwell Publishing Ltd, Freshwater Biology, 53, 264–277
266
A. De Wever et al.
0
Burundi
siz
3°S
i
A total of five nutrient limitation bioassay experiments were carried out: during the rainy season and
the dry season of 2003 and during the rainy season of
2004 at the northern station off Kigoma (20 March
2003, 26 August 2003 and 9 February 2004) and during
the dry season of 2003 and the rainy season of 2004 at
the southern station off Mpulungu (15 September 2003
and 22 February 2004). All samples were collected
using Hydrobios (Kiel, Germany) or Go-Flo (General
Oceanics, Miami, FL, U.S.A.) bottles (5 or 10 L)
attached to a stainless steel Teflon coated cable wire.
For the experiments, lake water collected at different
depths (0, 10, 20 and 30 m) in the euphotic zone was
pooled in a large container. Prior to transfer into three
replicate 2 L polycarbonate bottles, the water was prescreened through a 28-lm mesh plankton net to
remove large grazers. Large zooplankton is present at
low densities in Lake Tanganyika (about four to nine
individuals per litre, Kurki et al., 1999) and their
inclusion in the incubations might therefore result in
undesired variation between replicates. In addition,
results from concurrent size fractionation experiments
(including unfiltered treatments) do not suggest the
100
4°S
Kigoma
TK1
Tanzania
si
gara
Mala
5°S
6°S
7°S
TK8
D.R.
Congo
8°S
Mpulungu
N
Zambia
29°E
Experimental setup
50
km
Ru
Chlorophyll a (Chl a) concentrations in the epilimnion
range from 0.3 to 3.4 lg L)1 (Descy et al., 2005) and
cyanobacterial picophytoplankton is always a dominant component of the phytoplankton community in
the southern basin, but not in the northern basin
where the larger phytoplankton is better developed.
The community of larger phytoplankton is composed
mainly of chlorophytes and diatoms (Nitzschia spp.).
Large blooms of diazotrophic filamentous cyanobacteria were observed by Symoens (1955); Hecky &
Kling (1981) and Salonen et al. (1999) at the onset of
the rainy season (September to November), but these
cyanobacteria have not been observed as a major
constituent of the phytoplankton more recently
(Cocquyt & Vyverman, 2005; Descy et al., 2005). A
decrease in water column stability is followed by an
increase in picocyanobacteria in the south of the lake
and by an increase in diatoms in the north of the lake
(Descy et al., 2005; De Wever et al., in press). Experiments for this study were carried out at two sites:
Kigoma (0452.51¢S, 2935.75¢E, Tanzania) in the
northern basin and Mpulungu (0845.11¢S 3104.35¢E,
Zambia) in the southern basin (Fig. 1).
30°E
31°E
32°E
33°E
Fig. 1 Lake Tanganyika sampling sites for experiments and
biweekly monitoring (squares) and deep Fe profiles (TK1 and
TK8; dots).
occurrence of strong shifts in phytoplankton community composition due to this filtration (A. De Wever,
unpublished data).
We used a bifactorial experimental design, with the
factors N + P and Fe additions. Treatments were
prepared in triplicate. N, P and Fe were added to
reach final concentrations of 2.5 lM NaH2PO4, 10 lM
NH4Cl and 12 lM FeCl3Æ6H2O in the incubations. The
N and P concentrations used correspond to the
maximum concentrations observed in the epilimnion
of the lake, while Fe was added in excess compared
with in situ concentrations to ensure that the concentration of bioavailable Fe was raised during the
experiments. This level of iron addition is high in
comparison with in situ concentrations. Due to the
high pH, however, it is not expected to result in a
direct rise in bioavailable ferrous iron, but would
rather influence the extent of iron photoreduction
(Wells & Mayer, 1991). The response of the biomass of
the major phytoplankton groups in the lake to
2007 The Authors, Journal compilation 2007 Blackwell Publishing Ltd, Freshwater Biology, 53, 264–277
Differential response of phytoplankton
different nutrient additions was monitored in closed
containers over a period of 3 days.
All containers and equipment coming into contact
with the water used in the experiments was thoroughly soaked in 10% HCl and rinsed three times
with distilled water and once with lake water. The
polycarbonate bottles were incubated in the lake at
5 m depth, which corresponds to c. 50% of surface
irradiance when light absorption by the bottle wall
was taken into account. At this depth, light should
mostly saturate photosynthesis, according to published photosynthesis-irradiance curves for Lake
Tanganyika phytoplankton (Sarvala et al., 1999). Subsamples for pigment analysis (500 mL) and phytoplankton enumeration (100 mL) were collected at the
start of the experiment and after 24 and 72 h of
incubation.
Field monitoring
To obtain background information on the limnological
conditions in the lake at the time of the experiments,
data were collected biweekly at fixed sites close
(<2.5 km) to the locations where water for the bioassays was collected, starting 4 weeks before and
ending 4 weeks after the experiment. On each occasion we recorded Secchi depth and measured
temperature and oxygen profiles using conductivity–
temperature depth instruments to determine the
depth of the thermocline. Samples for the determination of nutrient content were collected at 20-m
intervals down to 100 m, phytoplankton samples
were collected at 20 m. Samples for geochemical
analysis of dissolved Fe and Al were collected
fortnightly from March 2003 to July 2004 at 20 m
depth at the pelagic site in Mpulungu. Additional
sampling for geochemistry analysis was performed
for deep profiles at station TK1 (northern basin) and
TK8 (southern basin) during a north–south transect in
January–February 2004 (Fig. 1).
Laboratory procedures
Dissolved nutrients analysis was performed on prefiltered water samples (Macherey-Nägel GF-5 filter;
Macherey-Nägel, Düren, Germany). Nitrate concentrations were determined spectrophotometrically using
Macherey-Nägel kits (Macherey-Nägel, Düren, Germany). Dissolved reactive phosphate concentration
267
was measured according to standard methods
(Greenberg et al., 1992). Particulate nutrient concentrations in the mixed layer were determined by filtering
3 L of water over a pre-combusted GF-5 filter, and the
filters were stored in the freezer after at least 12 h
drying. Elemental analysis of the seston was carried out
using a Carlo Erba NA1500 elemental analyser (Thermo
Fisher Scientific, Milan, Italy) for carbon (C) and
nitrogen (N) after HCl treatment to remove carbonates.
Particulate phosphorus was measured spectrophotometrically after 30 min digestion at 120 C with potassium persulphate. The obtained nutrient ratios were
compared with the threshold proposed by Healey &
Hendzel (1979) in order to evaluate potential N or P
limitation.
Samples for pigment analysis were directly filtered
over a Whatman GF/F filter (Whatman, Maidstone,
U.K.). The glass fibre filters were wrapped in aluminium foil and stored at )20 C. Pigments were
extracted using 90% acetone using sonication (tip
sonicator, 40 W for 30 s; Sonics & Materials Inc.,
Newtown, CT, U.S.A.). Pigment extracts were filtered
over a 0.2-lm nylon membrane to remove particles.
Pigments were analysed according to the method of
Wright & Jeffrey (1997) using a Gilson HPLC system
(Gilson, Middleton, WI, U.S.A.) equipped with an
Alltima reverse-phase C18 column (Alltech Associates
Inc., Deerfield, IL, U.S.A.; 25 cm · 4.6 mm, 5 lm
particle size). Pigment identity was always verified
by inspecting the absorption spectra using a diode
array detector. The pigment data from the start of the
experiment were processed with the C H E M T A X software (Mackey et al., 1996) to estimate the contribution
of major algal groups to total phytoplankton biomass
using an initial pigment ratio matrix derived from a
published study on Lake Tanganyika (Descy et al.,
2005). The Chl a concentration was used as a measure
for total phytoplankton biomass while zeaxanthin,
lutein and fucoxantin were used as marker pigments
for picocyanobacteria, chlorophytes and diatoms,
respectively. As marker pigments of other phytoplankton groups were below detection limits, the
nanophytoplankton cells counted using epifluorescence microscopy were considered as chlorophytes.
Diatoms were only studied using their marker pigment, as the low cell densities did not allow accurate
counts by epifluorescence microscopy.
Samples for counting phytoplankton were fixed
according to the lugol-formalin-thiosulphate method
2007 The Authors, Journal compilation 2007 Blackwell Publishing Ltd, Freshwater Biology, 53, 264–277
268
A. De Wever et al.
(Sherr & Sherr, 1993). A 100-mL sample was filtered
onto a 0.8-lm pore size polycarbonate membrane
filter. Pico- and nanophytoplankton were counted
using epifluorescence microscopy with green and blue
light illumination. A distinction was made between
prokaryotic picophytoplankton (<2 lm and fluorescing under green light illumination) and eukaryotic
nanophytoplankton (fluorescing under blue light
illumination) (MacIsaac & Stockner, 1993). At least
400 cells of each group were counted. Samples for
inverted microscopy (1 L) were concentrated by 48 h
sedimentation. Microscopical analysis allowed us to
determine dominant phytoplankton groups during
the course of the experiments.
For Fe and Al analysis, 1–4 L of lake water was
transferred on board to 7 L gauged Perspex filtration
units and filtered on 47 mm diameter Poretics polycarbonate membranes with 0.4 lm porosity under
pressure of filtered air (0.4 lm porosity). Filtered
water samples were acidified (0.2%) using ultrapure
bi-distilled nitric acid HNO3 and stored in a fridge in
50 mL acid clean HDPE bottles. In a clean laboratory,
1-mL aliquots of the filtrated fraction were acidified
(2% HNO3). Al27, Fe56 isotopes were analysed using a
magnetic sector inductively coupled plasma mass
spectrometer (HR-ICP-MS; ELEMENT2 Finnigan,
Thermo Scientific, Waltham, MA, U.S.A.). External
references for quality control consisted of a multielement standard solution and an international standard material of river water (SLRS-4). Instrumental
drift effects were corrected by measuring the fluctuations of two internal standards (115In and 209Bi) at
1 lg L)1. Detection limits (3r of the blank values) for
Fe and Al were 10 and 50 nM respectively. All
materials used were of polypropylene and were
previously cleaned in a succession of nitric acid baths
for several hours followed by a rinsing phase in
de-ionized distilled water (Millipore Water Systems,
Millipore, Billerica, MA, U.S.A.).
Data analyses
To allow easy comparison of the water column
stability between sampling occasions, we calculated
the potential energy anomaly (PEA) for the upper
100 m of the water column according to Simpson et al.
(1982). Secchi depth was converted to euphotic depth
(1% of surface irradiance) using a conversion factor
obtained from simultaneous measurements with a
LICOR quantum sensor (LI-COR, Lincoln, NE, U.S.A.
k ¼ 1.57/Secchi depth). Differences in the environmental conditions between subsequent experiments
were tested using one-way A N O V A followed by an
LSD post hoc tests. In case of divergence from
homogeneity of variance as indicated by Levene’s
<0.01, data were log-transformed before performing
the A N O V A analysis.
Differences in cell densities and pigment concentrations between experimental treatments were tested
using two-way A N O V A after applying a Levene’s test
to verify the homogeneity of variances across treatments. In case of significant A N O V A results, a Tukey’s
honest significant difference post hoc test was applied
to test for the significance of differences between
individual treatments and to identify homogeneous
groups of treatments not significantly different from
each other. Fe or N + P were considered as individually limiting if both treatments receiving this amendment were significantly higher than the control
treatment, i.e. Fe was limiting if both the Fe and
FeNP treatments had a higher growth rate than the
control treatment. Fe and N + P were considered to be
co-limiting if growth rate in the FeNP treatments was
higher than in the control treatment and the treatments receiving Fe and N + P.
In order to summarize the experimental response to
the different nutrient amendments over the different
experiments, we used the meta-analysis methods as
described by Downing, Osenberg & Sarnelle (1999),
with
Dr ¼
ln
Nt;E
Nt;C
t
as the effect size measure and unweighted resampling
methods (Adams, Gurevitch & Rosenberg, 1997) using
5000 replicates. This effect size Dr, with Nt,E and Nt,C
the cell density or pigment concentration, respectively, in the experimental and the control treatment
and t time (in days), corresponds to the definition of
nutrient limitation as being the per unit growth rate of
an algal assemblage following the addition of surplus
nutrients.
Results
At both stations, the PEA and surface water temperature (Fig. 2) were higher during the rainy season
2007 The Authors, Journal compilation 2007 Blackwell Publishing Ltd, Freshwater Biology, 53, 264–277
Kig RS 03
Mpu RS 04
20
Kig RS 04
Mpu DS 03
0
Kig DS 03
Differential response of phytoplankton
S 04
Kig R
60
Kig
DS
03
Depth (m)
40
80
100
120
23
24
25
26
Temperature (°C)
27
28
Fig. 2 Temperature profiles measured during the experiments.
The oxycline depth is indicated with a grey dot for the northern
station Kigoma, the oxycline depth for Mpulungu could not be
located on the conductivity–temperature depth profiles reaching
up to 100 m.
than during the dry season (Table 1). The PEA was
lowest during the dry season in the southern station.
Epilimnetic dissolved phosphate concentration varied
between 0.1 and 0.6 lM while nitrate concentrations
varied from 1.6 to 3.7 lM . Phosphate and nitrate
concentration did not significantly differ between the
dry and rainy season. Dissolved silica concentration
was always relatively high (24.6–33.5 lM ). The
particulate nutrient ratios did not show significant
differences among the experiments. C : N ratio
ranged from 7.52 to 8.33, C : P from 140 to 186 and
N : P from 16.8 to 24.3.
The deep Fe and Al profiles during the rainy season
of 2004 clearly indicated an increase in dissolved Fe
(<0.4 lm) concentration with depth below 150 m and
a decrease in dissolved Al concentration with depth
(Fig. 3). Epilimnial concentrations ranged from 7.1 to
25.9 nM for Fe and from 135.7 to 329.5 nM for Al,
while concentrations below 150 m ranged from 81.7 to
250.0 nM for Fe and 59.9 to 83.3 nM for Al. Surface
water monitoring of Fe and Al revealed that concentrations at 20 m ranged from 1.47 to 92.28 and 106 to
649 nM , respectively (Fig. 4). These epilimnetic concentrations of both elements co-varied over time,
suggesting external terrestrial input of both elements.
The increase in Al and Fe concentrations near the end
269
of the rainy season corresponded with the period of
maximal inflow rate of the three main tributary rivers
(Langenberg et al., 2003a).
Chlorophyll a concentration at the time of the
experiments varied between 0.6 and 1.7 lg L)1. The
Chl a concentration was significantly higher at
Mpulungu than at Kigoma (P < 0.001) and was also
significantly higher during the dry season than the
rainy season at both stations (Table 1). The C H E M T A X
analysis of the HPLC pigment data revealed that
picocyanobacteria were relatively more important at
the southern station Mpulungu, especially during the
dry season. Chlorophytes and diatoms had a significantly higher contribution during the dry season at
Kigoma, but not at Mpulungu. The cell density of
picocyanobacteria ranged from 2.71 to 9.88 · 104
cells mL)1 and was highest during the dry season at
the southern station. Nanophytoplankton density was
comparable during all experiments and ranged from
0.98 to 2.36 · 103 cells mL)1. Analysis of samples with
an inverted microscope showed that the larger phytoplankton (‡5 lm) was dominated by chlorophytes
in the rainy season at the northern station, by
Chroococcales (coccoid cyanobacteria) during the rainy
season at the southern station and by the large
colonial diatom Nitzschia asterionelloides O. Müll.
during the dry season at both stations (Cocquyt &
Vyverman, 2005).
Addition of Fe had a significant positive effect on
picocyanobacterial density and zeaxanthin concentration in all experiments, except Mpu RS 04 (Fig. 5).
Even in the latter experiment, there was an effect of Fe
in combination with NP after 3 days. No stimulatory
effects of Fe addition were observed for nanophytoplankton and its corresponding marker pigment
lutein, or for the main accessory pigment in diatoms
(fucoxanthine).
Addition of NP had a significant positive effect on
picocyanobacterial densities in the Kig RS 04 and Mpu
RS 04 experiments after 1 day, whereas no effects on
the zeaxanthin concentrations were observed. Clear
increases in nanophytoplankton densities and lutein
concentration were observed in response to NP
addition at the northern station Kigoma; nanophytoplankton densities were raised in the experiments
during the rainy season and lutein concentrations
showed an increase in all experiments. At the southern station, the response was less apparent, there
being no significant increase in lutein concentration
2007 The Authors, Journal compilation 2007 Blackwell Publishing Ltd, Freshwater Biology, 53, 264–277
RS 2003
25.99 (0.13)
17.81 (0.49)
0.66 (0.16)
0.09 (0.03)
3.71 (4.49)
NA
7.93 (0.50)
171.18 (10.17)
21.69 (2.50)
0.56 (0.05)
0.34 (0.03)
Season
Temperature (C)
PEA (J m)3)
Zm : Zeu
PO4 (lM )
NO3 (lM )
Si (lM )
C : N (mol mol)1)
C : P (mol mol)1)
N : P (mol mol)1)
Chl a (lg L)1)
Picocyanobacteria
(lg Chl a L)1)
Chlorophytes
(lg Chl a L)1)
Diatoms (lg Chl a L)1)
Picocyanobacteria
(104 cells mL)1)
Nanophytoplankton
(103 cells mL)1)
Dominant
phytoplankton
group (‡5 lm)
>>
>>>
>>>
<<
<<
<<
Chlorococcales
1.67 (0.42)
0.04 (0.01)
3.62 (0.74)
0.24 (0.02)
RS 2004
25.71 (0.15)
16.66 (2.29)
1.41 (0.17)
0.61 (0.42)
3.28 (0.78)
25 (2)
8.13 (1.05)
152.18 (31.59)
18.67 (2.47)
0.68 (0.05)
0.40 (0.02)
Diatoms (Nitzschia
asterionelloides)
2.36 (1.28)
0.07 (0.02)
9.88 (0.06)
0.39 (0.02)
DS 2003
24.75 (0.28)
4.97 (3.41)
4.29 (2.38)
0.48 (0.13)
2.92 (0.93)
32 (4)
7.52 (0.27)
153.53 (32.34)
20.33 (3.64)
1.72 (0.01)
1.27 (0.02)
>>>
>>>
>>>
<<<
<<<
>>
Southern station – Mpulungu
Chroococcales
1.92 (0.30)
0.06 (0.05)
2.74 (0.21)
0.36 (0.07)
RS 2004
26.10 (0.16)
18.16 (0.91)
0.81 (0.08)
0.36 (0.06)
3.07 (1.07)
33 (2)
7.74 (0.37)
185.69 (73.18)
24.28 (10.78)
0.97 (0.11)
0.55 (0.02)
Standard deviation is given between brackets. Significant differences between consecutive seasons based on A N O V A -analysis followed by an LSD-test are indicated by one to three
less/greater-than signs, respectively corresponding to P < 0.05, P < 0.01 and P < 0.001.
Diatoms (Nitzschia
asterionelloides)
Chlorococcales
0.12 (0.01)
2.90 (0.66)
0.98 (0.12)
<<
0.05 (0.01)
2.71 (0.31)
0.64 (0.00)
DS 2003
25.30 (0.09)
9.36 (2.08)
0.81 (0.22)
0.55 (0.26)
1.64 (0.93)
28 (1)
8.33 (0.44)
140.15 (11.82)
16.81 (0.87)
1.07 (0.03)
0.31 (0.03)
1.36 (0.24)
<<<
0.17 (0.01)
<<<
>>>
>>>
Northern station – Kigoma
Station
Table 1 Environmental conditions during the experiments showing the mean of four samplings over 0–100 m for temperature, potential energy anomaly (PEA), mixed over
euphotic depth (Zm : Zeu), PO4, NO3 and Si; mean of four samplings over 0–40 m for the particulate nutrient ratios C : N, C : P and N : P; starting conditions in the experiments
for chlorophyll a (Chl a), the breakdown of Chl a into the groups picocyanobacteria, chlorophytes and diatoms (based on C H E M T A X analysis), picocyanobacteria and
nanophytoplankton cell densities (from epifluorescence microscopy) and the dominant phytoplankton group ‡5 lm (as observed using inverted microscopy)
270
A. De Wever et al.
2007 The Authors, Journal compilation 2007 Blackwell Publishing Ltd, Freshwater Biology, 53, 264–277
Differential response of phytoplankton
0
50
Al (nM)
150
200
100
250
300
350
0
100
TK1
TK8
Depth
200
300
400
500
0
50
100
150
Fe (nM)
200
250
300
Fig. 3 Vertical profiles for dissolved Fe (<0.4 lm; dashed line)
and Al (full line) measured at the northern station TK1 and the
southern station TK8 during the rainy season of 2004. The
oxycline depth is indicated with a grey dot.
100
700
Dry season
90
Dry season
600
80
Fe (nM)
60
400
50
300
40
30
Al (nM)
500
70
200
20
100
10
0
0
F
M A M J
J
2003
A
S
O
N D
J
F
M A M
J
J
2004
Date
Fig. 4 Temporal variation in dissolved Fe (<0.4 lm; dashed line)
and Al (solid line) concentrations at 20 m at the southern station
Mpulungu.
after 3 days in the treatments receiving NP addition.
Fucoxanthin concentrations were significantly
increased by NP additions in the Kig DS 03 and Kig
RS 04 experiments.
The increase in cyanobacterial picophytoplankton,
as well as zeaxantin concentration, was higher in the
FeNP treatment than in the Fe and NP treatments in
all experiments after 3 days, suggesting that
picocyanobacteria became co-limited by Fe and NP
during extended incubation. In the Mpu DS 03
271
experiment, even by day 1, cyanobacterial picophytoplankton growth indicated co-limitation by FeNP.
Co-limitation of nanophytoplankton by FeNP was
observed after 3 days in the Kig RS 03, Kig DS 03 and
Mpu RS 04 experiments. Lutein and fucoxanthin
increases were never higher in the FeNP treatments
than in the Fe or NP treatments.
The Chl a concentration was significantly raised in
the control treatments in all experiments, except Mpu
RS 04. Fe addition alone did not result in a significant
increase in Chl a concentration, while NP caused a
significant increase in the Kig DS 03 and Kig RS 04
experiments. Combined addition of Fe and NP
resulted in a significant increase in Chl a concentration in all experiments.
Considering the number of significant effects (Fe-/
NP- or co-limitation) after 1 day for prokaryotic
picocyanobacteria and nanophytoplankton, the counts
showed effects in 11 of 30 cases while pigments
revealed significant effect in only six of 30 cases. After
3 days, counts revealed significant effect in 15 of 30
cases and pigments revealed significant effect in 11 of
30 cases.
The experimental results were summarized using a
meta-analysis. The largest effect sizes (up to 0.76) were
observed for picocyanobacteria and zeaxanthin
(Fig. 6a,b) in the Fe and FeNP treatments, with the
highest response found in treatments receiving all
nutrients. While the effect size after 1 day for zeaxanthin was similar for Fe and FeNP addition, it was
considerably higher in the FeNP treatment after 3 days.
Nanophytoplankton and lutein (Fig. 6c,d) did not
show a significant positive response to Fe addition.
The evidence rather pointed to a stimulatory effect of
NP, with a positive effect on the lutein concentrations
after 3 days and a significant positive effect of FeNP
addition on nanophytoplankton density. No significant
positive effects were observed for fucoxanthin (Fig. 6e)
in any of the treatments. The effect size for Chl a (Fig. 6f)
was positive after 3 days for the treatments receiving
FeNP and NP. Only Fe did not have a positive effect
size, which may be due to the fact that it had a positive
effect on zeaxanthin concentration but a negative effect
on lutein concentration.
Discussion
Five nutrient enrichment bioassays with NP and Fe
were performed during three consecutive seasons at
2007 The Authors, Journal compilation 2007 Blackwell Publishing Ltd, Freshwater Biology, 53, 264–277
272
A. De Wever et al.
Kig RS 03
104 cells mL–1
Prokaryotic
picoplankton
25
Kig DS 03
BC
20
B
D
15
AB
10
5
C
A
B
A
0
1.4
C
0.6
A
A
103 cells mL–1
10 A A
B A
C
C
A
B
B
8
6
6
10
A
A
0
10
B
15
C
10
8
AB
A
25
20
B
15
Fe
FeNP NP
B
40
30
30
A
A
A
10
B
B
1.0
B
0.2
0.8
0.6
0.4
B
A
AB
A
A
A
0.2 B
AB
0.0
B
A
AB
0.0
C
Fe
FeNP NP
C
Fe
Kig RS 03
FeNP NP
FeNP NP
0.6
3
0.4
2
B
A
A
B
C
Fe
0
FeNP NP
0
0.8
0.7
0.6
0.5
0.4
0.3
0.2
A
0.1 AB
0.0
C
B
AB
C
Fe
FeNP NP
A
B
B
5
A
A
A
15
5
0
Fe
FeNP NP
C
Fe
A
A
Fe
FeNP NP
C
Fe
FeNP NP
Mpu RS 04
0.20
0.4
0.15
0.3
0.10
0.2
0.05
0.1
B
A
A
A
0.0
C
Fe
FeNP NP
C
Fe
A A
B
C
Fe
FeNP NP
B
4
3
2
A
5
A A
FeNP NP
Mpu RS 04
B
10
A
A
A
A
0
C
FeNP NP
15
10
5
0
C
0.00
B
A
1
0.02
0.00
Mpu DS 03
10
2
A
0.04
A
20
4
3
A
0.08
B
C
B
A
0.10
0.05
FeNP NP
B
A
0.06
Kig RS 04
20
15
AB
0.00
Fe
AB
A
0.12
B
Mpu DS 03
0.0
C
A
0
B
B
AB
B
B
2
0.10
A
C
4
0.30
FeNP NP
FeNP NP
B
6
0.15
C
AB
Fe
8 AB
0.20
Fe
C
10
0.25
B
A
12
0
0.35
A A BC
A
A
0.2
Mpu RS 04
5
C
Kig DS 03
25
B
FeNP NP
AB
0.0
C
B
AB A AB
1.0
0.2
B
Kig RS 03
6
A
Fe
A A
10
20
0.4
A
B
15
0.8
1
AB
A
0.0
A
0.4
B
20
0.6
C
2 A A
B
1.2
B
4
A
AB
A
Mpu DS 03
BC
AB
B
B
0.6
Kig RS 04
5
B
A A
A
1.0
0
Fe
Kig DS 03
1.0
0.8
C
10 A
1.2
0.3
A
B
5
B
A
0.8
Kig RS 04
40
0
1.4
0.1
B A
A A
50
B
20
B
B
10
2
BC
AB A
A
0
4
Kig DS 03
B
µg L–1
20 A
0
C
50
C
0
0.4
µg L–1
20
4
0.5 A A AB A B
FeNP NP
5 AA A A
µg L–1
30
0
10
0.0
Fe
30
0.2
20
40
30
B
Kig RS 03
Eukaryotic
nanophytoplankton
A A
1.0
B
0.4
C
Lutein
B
B
25
C
50
40
Mpu RS 04
30
D
60
1.5
0.0
Fucoxanthin
Mpu DS 03
B
50
2.0
0.8
0.2
Chlorophyll a
Kig RS 04
60
C
2.5
1.0
µg L–1
Zeaxanthin
1.2
40
35
30
25
20
C
15
B
10 B A
5
0
3.0
0
A
A
1
0
C
Fe
FeNP NP
C
Fe
FeNP NP
Fig. 5 From top to bottom picocyanobacterial densities, zeaxanthin concentrations, nanophytoplankton densities, lutein, fucoxanthin
and chlorophyll a concentrations after 24 h (white bars) and 72 h (shaded bars) during five nutrient addition experiments (Kig,
Kigoma; Mpu, Mpulungu; RS, rainy season; DS, dry season; 03/04, year). C, control treatment receiving no nutrient addition; Fe,
treatment receiving iron (Fe), FeNP, treatment receiving both Fe, nitrogen (N) and phosphorus (P) addition; NP, treatment receiving N
and P addition. Initial densities or concentrations are indicated using a dashed line. Significant P-levels of two-way A N O V A are
included in Table 2. Homogeneous groups based on Tukey-tests are indicated with A, B, etc., with AB not significantly different from
both A and B. No letters were used in case no significant differences between groups were observed during A N O V A or post hoc tests.
2007 The Authors, Journal compilation 2007 Blackwell Publishing Ltd, Freshwater Biology, 53, 264–277
Differential response of phytoplankton
Fe
Fe t24
Fe t72
FeNP
FeNP t24
FeNP t72
NP
NP t24
NP t72
(a)
(b)
Fe
Fe t24
Fe t72
FeNP
FeNP t24
FeNP t72
NP
NP t24
NP t72
(c)
(d)
Fe
Fe t24
Fe t72
FeNP
FeNP t24
FeNP t72
NP
NP t24
NP t72
(e)
(f)
–0.3 0
0.3
0.6 0.9 1.2 –0.3 0
0.3
0.6
0.9
1.2
Effect size Δr
Fig. 6 Effects of nutrient addition on phytoplankton growth,
as measured by the effect size Dr (see text for details) for (a)
picocyanobacteria, (b) zeaxanthin, (c) eukaryotic nanophytoplankton, (d) lutein, (e) fucoxanthin and (f) chlorophyll a. Error
bars represent 95% confidence intervals of Dr based on the resampling procedures with 5000 iterations. Analyses were performed using all data, data obtained after 1 day (t24) and data
obtained after 3 days (t72) and combined in one graph.
Table 2 Results of two-way
Picocyano-bacteria
Zeaxanthin
Nanophyto-plankton
Lutein
Fucoxanthin
Chl a
t24
t72
t24
t72
t24
t72
t24
t72
t24
t72
t24
t72
ANOVA
273
two stations in Lake Tanganyika. When set against
monitoring data at these two sites (Cocquyt &
Vyverman, 2005; Descy et al., 2005; De Wever et al.,
In press), the timing of our experiments includes
contrasting periods in the lake with different mixing
intensities and differences in the phytoplankton
community.
Both cell counts and pigment data indicated that,
phytoplankton communities were always stimulated
by the addition of at least one nutrient. Furthermore we
demonstrated that different algal groups respond
differentially to different nutrients; Fe additions
mainly stimulated prokaryotic picoplankton, while
NP stimulated green algae and, in some cases, diatoms.
In addition, we observed a number of interesting
differences between the results obtained after 1 and
3 days incubation. Meta-analysis results integrating all
five experiments suggest initial Fe limitation of zeaxanthine concentration after 1 day shifting towards colimitation with NP after 3 days. The fact that Fe
limitation changed into co-limitation could be due to
a decrease in ambient macronutrient concentrations
during incubation, which resulted in limitation by N
or/and P. Co-limitation by Fe and P for example was
observed during nutrient enrichment bioassays in Lake
Erie, where addition of these nutrients was thought to
facilitate nitrate uptake (North et al., 2007).
There were a larger number of significant responses
to nutrients based on cell counts rather than pigment
concentrations. This may underlie the different
responses obtained after 1 and 3 days, and can
showing significant effects and their corresponding P-level
Kig RS 03
Kig DS 03
Kig RS 04
Mpu DS 03
MpuRS04
FeNP < 0.001
FeNP < 0.001
/
FeNP < 0.001
NP 0.001
FeNP 0.001
Fe 0.004
NP < 0.001
Fe 0.003 NP < 0.001
FeNP 0.010
/
FeNP < 0.001
FeNP 0.023
FeNP < 0.001
Fe 0.004
FeNP < 0.001
Fe 0.036
FeNP < 0.001
Fe 0.002
NP 0.001*
NP < 0.001
NP < 0.001
FeNP 0.024*
NP < 0.001
Fe 0.002/NP < 0.001
FeNP < 0.001*
Fe < 0.001
FeNP < 0.001
FeNP 0.002
FeNP 0.044
FeNP 0.025
NP < 0.001
NP 0.011
FeNP 0.003
FeNP 0.024
FeNP < 0.001
FeNP < 0.001
FeNP < 0.001
Fe < 0.001
FeNP < 0.001
/
/
/
NP 0.003
/
FeNP 0.036
/*
FeNP < 0.001
NP < 0.001
FeNP 0.011
NP 0.007
Fe 0.008/NP 0.015*
FeNP 0.009
Fe < 0.001 NP 0.001
/
NP 0.012
FeNP 0.015
/
/
Fe 0.020/NP 0.020
Results supported by Tukey tests are shown in bold. The P-levels of individual factors (Fe and NP) are not shown in this table in case
the interaction term (FeNP) was significant, as these should be evaluated based on post hoc tests as shown in Fig. 5.
*Deviating homogeneity of variance Levene’s <0.01.
2007 The Authors, Journal compilation 2007 Blackwell Publishing Ltd, Freshwater Biology, 53, 264–277
274
A. De Wever et al.
probably be ascribed to initial decreases in the cellular
pigment concentration during incubation, which may
mask increases in biomass. Such changes in pigment
content can arise as a result of the altered average
light conditions experienced by the cells upon incubation or due to changes in the phytoplankton growth
rate (Kirk, 1994; McManus, 1995; Henriksen et al.,
2002). In general, however, the pigment data were in
agreement with the cell count data.
Iron limitation during our experiments was
observed for picocyanobacteria, but not for chlorophytes and diatoms. This agrees with the experimental observations of Parparova & Yacobi (1998) in
monoalgal cultures that cyanobacteria were most
sensitive to a reduction in Fe bioavailability. Similarly,
Brand (1991) found that picocyanobacteria had the
highest Fe requirements from a wide range of marine
phytoplankters. An effect of Fe on picocyanobacteria
or the picophytoplankton size class (£2 lm) during
nutrient bioassay experiments has previously been
observed in other lakes, as in Lake Erie (Twiss,
Auclair & Charlton, 2000b; Twiss et al., 2000a) and
in acidic Canadian Shield lakes (Auclair, 1995). The
importance of Fe as a limiting nutrient for picocyanobacteria could be linked to the high Fe requirements
of their photosynthetic apparatus. Raven, Evans &
Korb (1999) pointed out that the photosynthetic
apparatus of picocyanobacteria has a high proportion
of the Fe-rich photosystem I relative to the Fe-poor
photosystem II. In addition, Fe requirements of
picocyanobacteria in unialgal cultures were found
to be higher under low-light conditions (Kudo &
Harrison, 1997).
In contrast to picocyanobacteria, chlorophytes
appeared to be limited primarily by N and/or P rather
than by Fe. Stimulation of diatom growth by NP was
also observed in treatments receiving macronutrient
addition at the northern station. This limitation by
macronutrients was also indicated by the particulate
nutrient ratios, which suggest moderate P limitation
becoming more severe during the rainy season.
Despite the fact that significant responses to nutrient additions were observed for separate algal groups
in all experiments after 1 day of incubation, a
response in total Chl a was usually only observed
after 3 days and often only in the treatments receiving
all nutrients (N, P and Fe). A lack of response of total
Chl a to the addition of a single nutrient is not unusual
in a situation where community co-limitation (sensu
Arrigo, 2005) applies. That is, where one component
of the phytoplankton community is stimulated by one
resource and another component by a different
resource. Since picocyanobacteria, chlorophytes and
diatoms were often stimulated by different nutrients,
the addition of one only will not necessarily result in
noticeable changes in Chl a concentration [e.g. in case
of a positive response in zeaxanthin (picocyanobacteria) in combination with a decrease in lutein (chlorophytes)]. Decreases in the Chl a content of the algae
during incubation (as discussed earlier) may have
contributed to the slow response of total Chl a to
nutrient additions in the experiments. The increase in
Chl a concentration in the treatment receiving all
nutrients (Fe, N and P) agrees with nutrient enrichment bioassays in Lake Malawi, which also revealed a
strong stimulatory effect when Fe was added together
with N, P and Si (Guildford et al., 2003).
The most remarkable observation during these
nutrient enrichment bioassays was undoubtedly the
high sensitivity of picocyanobacteria to Fe addition.
This suggests that Fe might be an important factor
driving seasonal or spatial variation in picocyanbacterial production in the Lake Tanganyika. In situ
concentrations of dissolved (fraction <0.4 lm) Fe
showed a maximum at the end of the rainy season.
Both the close correlation between dissolved Fe and
Al concentrations, and the fact that the timing of the
Fe maxima coincided with the maximum river inflow
(Langenberg et al., 2003a), suggest an external source.
Data on Fe concentrations in the larger rivers discharging into the lake are needed to verify this
hypothesis. It is unclear, however, whether the measured variability in Fe concentrations in the lake can
influence picocyanobacterial production, especially
because the concentration of Fe added in our experiments was much higher than the maximal dissolved
Fe concentrations measured in the lake. In lakes
where cyanobacterial growth was limited by Fe,
however, it was proposed that the bioavailability of
Fe was limited due to rapid oxidation of ferrous to
ferric iron at high pH or due to complexation of Fe
with humic acids (Kuma et al., 1996; Evans & Prepas,
1997; Imai et al., 1999). This could also be the case in
Lake Tanganyika, where Fe may be biologically
unavailable due to the high pH (8.4 to 9.2). Other
factors affecting Fe bioavailability include irradiance
and the presence of organic ligands, which affect
photoreduction rates (e.g. Rijkenberg et al., 2006) as
2007 The Authors, Journal compilation 2007 Blackwell Publishing Ltd, Freshwater Biology, 53, 264–277
Differential response of phytoplankton
does pH (Miller et al., 1995). Decreases in pH associated with upwelling during the dry season (Plisnier
et al., 1999) could therefore have an important influence on the Fe bioavailability. Although the dry
season upwelling coincided with an increase in
picocyanobacteria, the current experimental data are
inconclusive about the role of Fe in their success.
Descy et al. (2005) suggested that the high ratio of
mixing depth to photic depth during periods of low
water column stability at Mpulungu would favour
small phytoplankton like picocyanobaceria as these
are efficient light harvesters. Both mechanisms may be
at work but their relative importance needs further
experimentation.
In conclusion, the differential response of phytoplankton groups observed during nutrient enrichment
bioassays suggest frequent multi-nutrient co-limitation in the phytoplankton community of Lake Tanganyika. In particular, our results highlight the high
sensitivity of picocyanobacteria to iron. There is
currently a debate on whether primary productivity
of Lake Tanganyika is decreasing due to increasing
water column stability and air temperatures (O’Reilly
et al., 2003; Verburg, Hecky & Kling, 2003; Sarvala
et al., 2006; Stenuite et al., 2007). Insights into the
ecology of one of the major primary producers of the
lake, the picocyanobacteria, are of key importance in
understanding these changes. Further studies detailing the dynamics of in situ Fe bioavailability, and their
effects on the phytoplankton community, could therefore significantly improve our understanding of this
unique system.
Acknowledgments
This study was carried out in the framework of the
CLIMLAKE project (EV/2) (Climate Variability as
Recorded in Lake Tanganyika), which is supported
by the Federal Science Policy Office, Belgium.
A.D.W. received support from the Institute for the
promotion of Innovation trough Science and Technology in Flanders (IWT). K.M. is a postdoctoral
fellow of the Flemish Fund for Scientific Research.
We thank M. Lionard for assistance during
pigment analyses, R. Dasseville, R. Wakakumbe, G.
Kasumbe, C. Sichamba for assistance during sampling for experiments. Dr H. Phiri, DOF, Mpulungu
and Dr D. Chitamwebwa, TAFIRI, Kigoma for
organizing bi-weekly sampling.
275
References
Adams D.C., Gurevitch J. & Rosenberg M.S. (1997)
Resampling tests for meta-analysis of ecological data.
Ecology, 78, 1277–1283.
Arrigo K.R. (2005) Marine microorganisms and global
nutrient cycles. Nature, 437, 349–355.
Auclair J.C. (1995) Implications of increased UV-B induced
photoreduction – iron(II) enrichment stimulates picocyanobacterial growth and the microbial food-web in
clear-water acidic Canadian Shield lakes. Canadian
Journal of Fisheries and Aquatic Sciences, 52, 1782–1788.
Brand L.E. (1991) Minimum iron requirements of marinephytoplankton and the implications for the biogeochemical control of new production. Limnology and
Oceanography, 36, 1756–1771.
Cavender-Bares K.K., Mann E.L., Chisholm S.W., Ondrusek M.E. & Bidigare R.R. (1999) Differential response
of equatorial Pacific phytoplankton to iron fertilization.
Limnology and Oceanography, 44, 237–246.
Chang C.C.Y., Kuwabara J.S. & Pasilis S.P. (1992)
Phosphate and iron limitation of phytoplankton biomass in Lake Tahoe. Canadian Journal of Fisheries and
Aquatic Sciences, 49, 1206–1215.
Coale K.H., Johnson K.S., Chavez F.P. et al. (2004) Southern
ocean iron enrichment experiment: Carbon cycling in
high- and low-Si waters. Science, 304, 408–414.
Cocquyt C. & Vyverman W. (2005) Phytoplankton in
Lake Tanganyika: a comparison of community composition and biomass off Kigoma with previous
studies 27 years ago. Journal of Great Lakes Research,
31, 535–546.
Coulter G.W. & Spigel R.H. (1991) Hydrodynamics. In:
Lake Tanganyika and its Life. (Ed. G.W. Coulter), pp.
49–75. Oxford University Press, New York.
De Wever A., Muylaert K., Cocquyt C., Van Wichelen J.,
Plisnier P.D. & Vyverman W. (in press) Seasonal and
spatial variability in the abundance of auto- and
heterotrophic plankton in Lake Tanganyika. Fundamental and Applied Limnology (Archiv für Hydrobiologie).
Descy J.-P., Hardy M.-A., Sténuite S., Pirlot S., Leporcq
B., Kimirei I., Sekadende B., Mwaitega S.R. & Sinyenza
D. (2005) Phytoplankton pigments and community
composition in Lake Tanganyika. Freshwater Biology,
50, 668–684.
Downing J.A., Osenberg C.W. & Sarnelle O. (1999) Metaanalysis of marine nutrient-enrichment experiments:
variation in the magnitude of nutrient limitation.
Ecology, 80, 1157–1167.
Evans J.C. & Prepas E.E. (1997) Relative importance of
iron and molybdenum in restricting phytoplankton
biomass in high phosphorus saline lakes. Limnology
and Oceanography, 42, 461–472.
2007 The Authors, Journal compilation 2007 Blackwell Publishing Ltd, Freshwater Biology, 53, 264–277
276
A. De Wever et al.
Greenberg A.E., American Public Health Association,
American Water Works Association & Water Environment Federation (1992) Standard Methods for the Examination of Water and Wastewater. American public health
association, Washington, DC.
Guildford S.J., Hecky R.E., Taylor W.D., Mugidde R. &
Bootsma H.A. (2003) Nutrient enrichment experiments
in tropical Great Lakes Malawi/Nyasa and Victoria.
Journal of Great Lakes Research, 29, 89–106.
Healey F.P. & Hendzel L.L. (1979) Indicators of phosphorus and nitrogen deficiency in 5 algae in culture.
Journal of the Fisheries Research Board of Canada, 36,
1364–1369.
Hecky R.E. & Kling H.J. (1981) The phytoplankton and
proto-zooplankton of the euphotic zone of Lake Tanganyika – species composition, biomass, chlorophyll
content, and spatio-temporal distribution. Limnology
and Oceanography, 26, 548–564.
Hecky R.E., Spigel R.H. & Coulter G.W. (1991) The
nutrient regime. In: Lake Tanganyika and its Life (Ed.
G.W. Coulter), pp. 76–89. Oxford University Press,
New York.
Henriksen P., Riemann B., Kaas H., Sorensen H.M. &
Sorensen H.L. (2002) Effects of nutrient-limitation and
irradiance on marine phytoplankton pigments. Journal
of Plankton Research, 24, 835–858.
Hyenstrand P., Rydin E., Gunnerhed M., Linder J. &
Blomqvist P. (2001) Response of the cyanobacterium
Gloeotrichia echinulata to iron and boron additions – an
experiment from Lake Erken. Freshwater Biology, 46,
735–741.
Imai A., Fukushima T. & Matsushige K. (1999) Effects of
iron limitation and aquatic humic substances on the
growth of Microcystis aeruginosa. Canadian Journal of
Fisheries and Aquatic Sciences, 56, 1929–1937.
Järvinen M., Salonen K., Sarvala J., Vuorio K. & Virtanen
A. (1999) The stoichiometry of particulate nutrients in
Lake Tanganyika – implications for nutrient limitation
of phytoplankton. Hydrobiologia, 407, 81–88.
Jickells T.D., An Z.S., Andersen K.K. et al. (2005) Global
iron connections between desert dust, ocean biogeochemistry, and climate. Science, 308, 67–71.
Johnson K.S., Gordon R.M. & Coale K.H. (1997) What
controls dissolved iron concentrations in the world
ocean? Marine Chemistry, 57, 137–161.
Kirk J.T.O. (1994) Light and Photosynthesis in Aquatic
Ecosystems. Cambridge University Press, Cambridge.
Kudo I. & Harrison P.J. (1997) Effect of iron nutrition on
the marine cyanobacterium Synechococcus grown on
different N sources and irradiances. Journal of Phycology, 33, 232–240.
Kuma K., Nishioka J. & Matsunaga K. (1996) Controls on
iron(III) hydroxide solubility in seawater: the influence
of pH and natural organic chelators. Limnology and
Oceanography, 41, 396–407.
Kurki H., Vuorinen I., Bosma E. & Bwebwa D. (1999)
Spatial and temporal changes in copepod zooplankton
communities of Lake Tanganyika. Hydrobiologia, 407,
105–114.
Langenberg V.T., Sarvala J. & Roijackers R. (2003b) Effect
of wind induced water movements on nutrients,
chlorophyll-a, and primary production in Lake Tanganyika. Aquatic Ecosystem Health & Management, 6,
279–288.
Langenberg V.T., Nyamushahu S., Roijackers R. &
Koelmans A.A. (2003a) External nutrient sources for
Lake Tanganyika. Journal of Great Lakes Research, 29,
169–180.
MacIsaac E.A. & Stockner J.G. (1993) Enumeration of
phototrophic picoplankton by autofluorescence
microscopy. In: Handbook of Methods in Aquatic Microbial Ecology (Eds P.F. Kemp, B.F. Sherr, E.B. Sherr & J.J.
Cole), pp. 187–197. Lewis Publishers, Boca Raton, FL,
U.S.A.
Mackey M.D., Mackey D.J., Higgins H.W. & Wright S.W.
(1996) Chemtax – a program for estimating class
abundances from chemical markers: application to
HPLC measurements of phytoplankton. Marine Ecology-Progress Series, 144, 265–283.
Martin J.H. & Gordon R.M. (1988) Northeast Pacific iron
distributions in relation to phytoplankton productivity. Deep-Sea Research Part I-Oceanographic Research
Papers, 35, 177–196.
McManus G.B. (1995) Phytoplankton abundance and
pigment changes during simulated in situ dilution
experiments in estuarine waters: possible artifacts
caused by algal light adaptation. Journal of Plankton
Research, 17, 1705–1716.
Miller W.L., King D.W., Lin J. & Kester D.R. (1995)
Photochemical redox cycling of iron in coastal seawater. Marine Chemistry, 50, 63–77.
Naithani J., Deleersnijder E. & Plisnier P.-D. (2002) Origin
of intraseasonal variability in Lake Tanganyika.
Geophysical Research Letters, 29, doi: 10.1029/
2002GL015843.
North R.L., Guildford S.J., Smith R.E.H., Havens S.M.
& Twiss M.R. (2007) Evidence for phosphorus,
nitrogen, and iron colimitation of phytoplankton
communities in Lake Erie. Limnology and Oceanography, 52, 315–328.
O’Reilly C.M., Alin S.R., Plisnier P.-D., Cohen A.S. &
Mckee B.A. (2003) Climate change decreases aquatic
ecosystem productivity of Lake Tanganyika, Africa.
Nature, 424, 766–768.
Parparova R. & Yacobi Y.Z. (1998) Chelatable iron in the
sub-tropical Lake Kinneret: its seasonal variation
2007 The Authors, Journal compilation 2007 Blackwell Publishing Ltd, Freshwater Biology, 53, 264–277
Differential response of phytoplankton
and impact on carbon uptake by natural algal assemblages and monoalgal cultures. Aquatic Sciences, 60,
157–168.
Plisnier P.-D., Chitamwebwa D., Mwape L., Tshibangu
K., Langenberg V. & Coenen E. (1999) Limnological
annual cycle inferred from physical-chemical fluctuations at three stations of Lake Tanganyika. Hydrobiologia, 407, 45–58.
Raven J.A., Evans M.C.W. & Korb R.E. (1999) The role of
trace metals in photosynthetic electron transport in O-2evolving organisms. Photosynthesis Research, 60, 111–149.
Rijkenberg M.J.A., Gerringa L.J.A., Carolus V.E.,
Velzeboer I. & De Baar H.J.W. (2006) Enhancement
and inhibition of iron photoreduction by individual
ligands in open ocean seawater. Geochimica Et
Cosmochimica Acta, 70, 2790–2805.
Salonen K., Sarvala J., Järvinen M., Langenberg V.,
Nuottajärvi M., Vuorio K. & Chitamwebwa D.B.R.
(1999) Phytoplankton in Lake Tanganyika – vertical
and horizontal distribution of in vivo fluorescence.
Hydrobiologia, 407, 89–103.
Sarvala J., Langenberg V., Salonen K. et al. (2006) Fish
catches from Lake Tanganyika mainly reflect changes
in fishery practices, not climate. Verhandlungen der
Internationalen Vereinigung für theoretische und angewandte Limnologie, 29, 1182–1188.
Sarvala J., Salonen K., Jarvinen M. et al. (1999) Trophic
structure of Lake Tanganyika: carbon flows in the
pelagic food web. Hydrobiologia, 407, 149–173.
Schelske C.L., Hooper F.F. & Heartl E.J. (1962) Responses
of a maerl lake to chelated iron and fertilizer. Ecology,
43, 646–653.
Sherr E.B. & Sherr B.F. (1993) Preservation and storage of
samples for enumeration of heterotrophic protists. In:
Handbook of Methods in Aquatic Microbial Ecology (Eds
P.F. Kemp, B.F. Sherr, E.B. Sherr & J.J. Cole), pp. 207–
212. Lewis Publishers, Boca Raton, FL, U.S.A.
Simpson J.H., Tett P.B., Argoteespinoza M.L., Edwards
A., Jones K.J. & Savidge G. (1982) Mixing and phytoplankton growth around an island in a stratified sea.
Continental Shelf Research, 1, 15–31.
Stenuite S., Pirlot S., Hardy M.-A., Sarmento H., Tarbe
A.-L., Leporcq B. & Descy J.-P. (2007) Phytoplankton
production and growth rate in Lake Tanganyika:
evidence of a decline in primary productivity in recent
decades. Freshwater Biology, doi: 10.1111/j.1365–
2427.2007.01829.x.
277
Sterner R.W., Smutka T.M., Mckay R.M.L., Xiaoming Q.,
Brown E.T. & Sherrell R.M. (2004) Phosphorus and
trace metal limitation of algae and bacteria in lake
superior. Limnology and Oceanography, 49, 495–507.
Stumm W. & Morgan J.J. (1996) Aquatic Chemistry.
Chemical Equilibria and Rates in Natural Waters. John
Wiley & Sons, New York.
Symoens J.J. (1955) Observation d’une fleur d’eau à
cyanophycées au lac Tanganyika. Folia Scientifica Africae Centralis, 1, 17.
Talling J.F. & Talling I.B. (1965) The chemical composition of African lake waters. Internationale Revue der
gesamten Hydrobiologie, 50, 421–463.
Twiss M.R., Auclair J.C. & Charlton M.N. (2000a) An
investigation into iron-stimulated phytoplankton productivity in epipelagic Lake Erie during thermal
stratification using trace metal clean techniques.
Canadian Journal of Fisheries and Aquatic Sciences, 57,
86–95.
Twiss M.R., Auclair J.C. & Charlton M.N. (2000b) An
investigation into iron-stimulated phytoplankton productivity in epipelagic Lake Erie during thermal
stratification using trace metal clean techniques (vol
57, pg 86, 2000). Canadian Journal of Fisheries and Aquatic
Sciences, 57, 870.
Twiss M.R., Gouvea S.P., Bourbonniere R.A., Mckay
R.M.L. & Wilhelm S.W. (2005) Field investigations of
trace metal effects on Lake Erie phytoplankton productivity. Journal of Great Lakes Research, 31, 168–179.
Verburg P., Hecky R.E. & Kling H. (2003) Ecological
consequences of a century of warming in Lake Tanganyika. Science, 301, 505–507.
Wells M.L. & Mayer L.M. (1991) The photoconversion of
colloidal iron oxyhydroxides in seawater. Deep-Sea
Research Part I-Oceanographic Research Papers, 38, 1379–
1395.
Wetzel R.G. (2001) Limnology, Lake and River Ecosystems.
Academic Press, San Diego, CA.
Wright S.W. & Jeffrey S.W. (1997) High-resolution HPLC
system for chlorophylls and carotenoids of marine
phytoplankton. In: Monographs on Oceanographic Methodology 10. Phytoplankton Pigments in Oceanography:
Guidelines to Modern Methods (Eds S.W. Jeffrey, R.F.C.
Mantoura & S.W. Wright), p. 327. UNESCO publishing, France.
(Manuscript accepted 11 September 2007)
2007 The Authors, Journal compilation 2007 Blackwell Publishing Ltd, Freshwater Biology, 53, 264–277