Effects of UV on carbon assimilation of phytoplankton 
in a mixed water column
Jan K?hler 1,*, Mechthild Schmitt2, Hartwig Krumbeck3, Maria Kapfer 3, 
Elena Litchman4, ** and Patrick J. Neale 4
1 Institute of Freshwater Ecology and Inland Fisheries, M?ggelseedamm 301, D-12587 Berlin,
Germany
2 Centre of Environmental Research, Permoserstr. 15, D-04318 Leipzig, Germany
3 Technical University Cottbus, Seestr. 6, D-15526 Bad Saarow, Germany
4 Smithsonian Environmental Research Center, 647 Contees Wharf Rd, PO Box 28, Edgewater,
MD 21037
** Present address EAWAG, Kastanienbaum, Switzerland
Key words: Primary production, ultraviolet radiation, light inhibition, turbulent
mixing, model.
ABSTRACT
Carbon assimilation is usually measured at fairly constant light intensities. Under natural condi-
tions, however, planktonic algae are moved through the water column and experience light of fluc-
tuating intensity and spectral composition. They may cope with strong UV for a short residence in
the upper water layer. In order to estimate the effects of UV on primary production of phyto-
plankton under conditions of turbulent mixing, we compared carbon assimilation and exudation
of algae incubated in UV-transparent quartz and in UV-absorbing glass bottles which were moved
through different water layers. Computer-controlled elevators were used to simulate mixing
depths between 2 and 14 meters. Compared to the glass bottles, particulate C assimilation in the
quartz bottles was reduced by 20?30% at mixing depths between 2 and 10 m. There was no sig-
nificant difference between both types of incubation bottles at a mixing depth of 14 m. Exudation
was enhanced by UV near the water surface (mixing depth up to 4 m) but not in the deep-mixed
samples. Our results indicate serious damage of planktonic algae by UV even under conditions of
vertical mixing if the euphotic zone exceeds the mixing depth. Depression was low for circulation
through the whole euphotic zone and may disappear at even deeper mixing. Our results indicate
lower photoinhibition per UV dosage at fluctuating than at constant light intensities. A model pre-
dicting inhibition as function of weighted irradiance spectra was adapted to describe wavelength
dependent photoinhibition occurring at different mixing depths. The model results agreed very
well with the inhibition rates measured under fluctuating light. These preliminary results are used
to discuss the importance of UV on photosynthesis of planktonic algae in aquatic environments of
different mixing depths and stabilities of stratification.
Aquat.sci.63 (2001) 294?309
1015-1621/01/030294-16 $ 1.50+0.20/0
? Birkh?user Verlag, Basel, 2001 Aquatic Sciences
* Corresponding author, e-mail: Koehler@igb-berlin.de
UV effects in mixed water columns 295
Introduction
During the last decades, water quality of many alpine and pre-alpine lakes has been
improved and arrived at an oligotrophic state again (e.g., B?rgi et al., 1999). As a
consequence, the depth of light transmission increased and enhanced ultraviolet
(UV) radiation may have greater effects on aquatic organisms. 
Many studies have been done on the impact of increased UV (280?400 nm) 
on phytoplankton. Particular attention has focused on the UV-B (280?320 nm)
which is selectively increased by stratospheric ozone depletion. Among other
effects, UV radiation damages the photosynthetic apparatus as well as the DNA
(e.g., Mitchell and Karentz, 1993; Buma et al., 1997; Vincent and Neale, 2000). UV
inhibition is a time-dependent phenomenon. Repair mechanisms may counterbal-
ance damage if the cell is exposed to UV for sufficiently short periods. Light of
longer wavelengths may activate such repair. Therefore, effects of UV depend on
duration and intensity of exposure to UV as well as on the ratio between PAR and
UV radiation. 
Carbon assimilation of phytoplankton is conventionally measured under con-
stant light intensities in the laboratory or at fixed depths and thus under fairly con-
stant light in situ (e.g., Gala and Giesy, 1991). Under natural conditions, plankton-
ic algae are subjected to turbulent water movement. Vertically transported algae
experience light of fluctuating intensity and spectral composition due to the expo-
nential, wavelength-dependent decline of light intensity with depth. Under mixing
conditions, planktonic algae receive strong light only for short periods and may use
their stay in deeper water layers to repair any damage done near the water surface.
As a consequence, the severe photoinhibition often measured during conventional
incubations near the water surface may be considered an artifact rather than a
process occurring under natural conditions. 
Most models of phytoplankton photosynthesis do not include time dependence
in the variation of photosynthetic parameters. In particular, inhibition of photosyn-
thesis by UV has been modelled as function of biologically weighted irradiance
(Cullen et al., 1992; Neale et al., 1994). This model represents the steady-state rate
of photosynthesis achieved when damage is counterbalanced by repair (Lesser et
al., 1994). Other modelling and measuring approaches have included time-depen-
dent responses to UV exposure as would occur under mixing conditions. The inter-
action between UV inhibition and vertical mixing in the Southern Ocean was mod-
elled using a simple relationship of inhibition to cumulative exposure (Neale et al.,
1998). Complex modelling approaches could combine models of photosynthesis
which incorporate known kinetics of photoinhibition and recovery (e.g., Pahl-Wostl
and Imboden, 1990) with hydrodynamic models describing the vertical movement
of algal cells in the mixed layer. Few attempts have been made to measure primary
production under simulated light fluctuations (circulator ? glass tube with pump,
Jewson and Wood, 1975; laboratory, Marra, 1978; rotating wheel, Benndorf and
Werner, 1980; lift, Nixdorf and Behrendt, 1991). These few studies on carbon as-
similation under conditions of vertical mixing used glass incubators and thus exclud-
ed ultraviolet radiation. Even fewer attempts have been made to consider UV dam-
age on planktonic organisms under fluctuating dose rates. Helbling et al. (1994) sim-
ulated the light regime of the mixed layer by incubating natural phytoplankton with
a rotating filter set. Zagarese et al. (1998a, 1998b) simulated light fluctuation in situ,
using a rotating wheel, to investigate the mortality of UV on zooplankton. 
This paper describes first measurements of carbon assimilation and exudation in
bottles receiving high and low exposure to short wavelength (l < 350 nm) UV which
were moved through different depths by a computer-controlled lift. The lift simu-
lated circulation cells of the Langmuir type with diameters between 2 and 
14 m. We tested the hypotheses that the degree of UV damage declines with
increased mixing depth and that damaging effects are negligible at sufficiently deep
mixing. The experimental procedure was supplemented by a model calculation of
the wavelength dependent photoinhibitory effect of UV-radiation.
Methods 
Experiments were performed in Lake Lucerne on 13 and 15 September, 1999 (assess-
ment of basic parameters described in Bossard et al., 2001, this issue). On the first
day, the same water sample was used as described by Neale et al. (2001a, this issue).
On 15 September, samples were drawn from 1, 3, 5, and 7 m depth, about 200 m away
from the landing of EAWAG station, Kastanienbaum. The mixed water samples
were subsampled into Duran and quartz bottles (about 120 ml) and inoculated with
NaH14CO3 (final activity about 0.39 mCi ml-1). A dark bottle control was wrapped
into aluminium foil. Each 2?3 bottles were horizontally positioned in a special hold-
ing device and fixed at the rope of the lifts. An early version of these elevators was
described by Behrendt (1989). The different lifts moved the bottles from water sur-
face to 3.9 m and 14 m depth on 13 September and to 2.0, 3.9 and 10 m depth dur-
ing the second experiment, respectively. The lifts simulated circular paths (lowest
vertical speeds at the upper and lower ends) with durations of one circle of 4 min
(0?2 m), 8 min (0?3.9 m) and 20 min (0?10 and 0?14 m), respectively. Exposures
lasted from 12:15 h till 16:15 h at the first day and from 10:25 h till 14:25 h (CEST) at
the second one. The lifts were fixed on a beam between the shadeless side of a boat
and a small buoy at the sampling position of the second day. The Duran bottles used
absorbed more than 95% radiation with wavelengths shorter than 315 nm and
about 50% of the radiation at 340 nm. The quartz bottles absorbed about 10% of
UV-B and about 50% of radiation at 200 nm (B?hlmann et al., 1987) (Figure 1A).
The effect of bottle transmission on biologically effective irradiance (using the bio-
logical weighting function of Neale et al., 2001a, this issue) is shown in Figure 1B.
Compared to quartz bottles, the Duran bottles absorb most of the biologically effec-
tive UV-B and a significant fraction of the biologically effective UV-A below 
350 nm. At the surface, about 60% of the decrease in effective irradiance in Duran
bottles is due to protection from UV-A and 40% due to protection from UV-B.
After the incubation, bottles were put into a dark box and transferred to the lab-
oratory. Total (TO14C), particulate (PO14C) and dissolved (DO14C) organic carbon
were distinguished. 7 ml subsamples were filtered onto polycarbonate filters (pore
size 0.2 ?m), to separate particulate and dissolved organic 14C. The filtrates and 7 ml
of the untreated sample for TO14C-measurements were acidified with HCl and bub-
bled with air for at least 30 min. Filters were placed on HCl-soaked filter paper, air
dried and dissolved with Soluene (Packard). All samples were measured with stan-
296 K?hler et al.
UV effects in mixed water columns 297
dard liquid scintillation techniques. At least two replicate determinations were
done from each bottle. 
Chlorophyll a analyses were performed by filtering 1?1.5 l of the mixed sample
through a GF/F filter, followed by extraction of the pigments in 6 ml ethanol and
HPLC measurement. 
Light intensities were calculated in 10 s intervals from the incubation depth, the
measured attenuation of PAR, UV-A and UV-B and the surface light intensities.
Irradiance, physical water parameters and nutrients were measured by EAWAG
laboratory, Kastanienbaum, using a LICOR-sensor for PAR and a MACAM-sen-
sor for UV-A und UV-B. Broad band attenuation coefficients were calculated from
a standard irradiance profile, measured on 13 September.
Figure 1. A Transmission properties of the Duran and quartz bottles used for incubation experi-
ments. (Quartz bottle data are redrawn from Fig. 1 of B?hlmann et al. (1987)). B Surface spectral
irradiance for local solar noon on September 13, 1999 at Lake Lucerne (see Neale et al., 2001b,
this issue) as weighted with the biological weighting function for UV inhibition of photosynthesis
by Lake Lucerne phytoplankton and after application of spectral transmission (Fig. 1A) of quartz
(broken line) and Duran (solid line) bottles. Vertical line marks the spectral boundary between
UV-B and UV-A
A
B
The wavelength dependent inhibition of photosynthesis by UV radiation was
calculated according to the BWF-PI model of Cullen et al. (1992). The model re-
presents UV effects as a function of weighted irradiance, and thus ignores dynami-
cal effects (Cullen and Neale, 1997). Instantaneous rates were calculated in inter-
vals of 20 s using estimated spectral irradiance. Surface spectral irradiance was mea-
sured with a SR18 UV-B spectroradiometer with 2 nm resolution (13 September)
and Macam broad-band UV sensors (13, 15 September). Measurements by these
sensors were used to adjust the output of a spectral radiative transfer model to
obtain full spectral irradiance 290?400 nm (details in Neale et al., 2001b, this issue).
Spectral attenuation coefficients (Neale et al., 2001a, this issue) were applied to
estimate in situ spectral irradiance as a function of depth and time. Measurement
and calculation of the biological weighting function was carried out by Neale et al.
(2001a, this issue). Application of biological weighting factors to in situ spectral
irradiance resulted in E*inh, the biologically effective fluence rate for inhibition of
photosynthesis (dimensionless). Results are expressed either as photosynthesis 
relative to the rate in the absence of inhibition (1/[1 +E*inh]) or as P/Pmax. P is the
modelled actual rate of photosynthesis depending on the intensity of the photo-
synthetic active radiation EPAR, the inhibitory effect of UV-radiation (expressed as
1/[1 +E*inh]) and the photosynthetic performance of the sample. Pmax is the potential
maximum rate of photosynthesis in absence of inhibition. The overall equation for
P is P/Pmax = [1 ? exp(?EPAR/Es)] * 1/[1 +E*inh], where Es is the characteristic irradi-
ance for light saturation of photosynthesis (286 mmol photons m?2 s?1, see Neale et
al., a this issue). Thus, 1/[1 +E*Inh] describes the inhibition of photosynthesis due to
the effect of UV-radiation, and P/Pmax describes the actual rate of photosynthesis
depending on PAR, inhibition by UV-radiation and physiological condition of the
sample. For both inhibition parameters, the relative inhibition was calculated as
response in quartz bottles (with UV) as percentage of response in Duran bottles 
(filtered UV).
Results
1. Environmental conditions
The general limnology of Lake Lucerne as well as the conditions during our experi-
ment have been described by Bossard et al. (2001, this issue). The vertical profiles of
temperature and chlorophyll fluorescence (Fig. 2) indicate a seasonal thermocline at
about 6 m with a weak temperature gradient extending to the surface. Maximum
phytoplankton biomass was found in the metalimnetic zone about 6?10 m deep.
Mean intensities of photosynthetically active radiation (PAR) of 1256 (13 Sept.) and
1157 mE m?2 s?1 (15 Sept.) were measured at the water surface. The attenuation coef-
ficients of PAR, UV-A and UV-B were calculated as 0.31 m?1, 0.58 m?1 and 1.3 m?1,
respectively. The euphotic zone (zeu) reached about 14 m but 99% of UV radiation
was attenuated at 7.9 m (UV-A) and at 3.5 m (UV-B). Intensities of radiation
received by vertically moved phytoplankton samples are illustrated in Figure 3. The
algae were exposed to UV-B intensities oscillating between 7.4?100% of surface
intensities at the upper lift (0?2 m) and between 0.6?100% from 0?3.9 m. The 
298 K?hler et al.
UV effects in mixed water columns 299
deep-mixed samples spent more than half of each circulation period in depths with
UV-B intensities below 1 mW cm?2 (see Table 1).
2. Carbon assimilation
The measured carbon assimilation rates are shown in Figure 4. They differed
between the simulated mixing depths according to the received light intensities.
Carbon assimilation was lowest at deepest mixing (0?14 m) and highest at medium
mixing depths. Assimilation rates were similar in bottles moved from 0?3.9 m and
Figure 2. Vertical profiles of chlorophyll fluorescence (a) and temperature (b) at Lake Lucerne on
September 15, 1999
a
b
300 K?hler et al.
Figure 3. Time course of intensity of PAR (a, b), UV-A (c, d) and UV-B (e, f) available to phyto-
plankton moved from 0?3.9 m (a, c, e) and from 0?14 m (b, d, f)
UV effects in mixed water columns 301
from 0?10 m. Carbon assimilation of phytoplankton kept in the upper water layer
(0?2 m) was inhibited by strong light. 
Rates of carbon assimilation were lower in quartz bottles (with UV) than in
Duran bottles which partially excluded UV (l < 350 nm). The reduction varied in
dependence of mixing depth. Up to a mixing depth of 10 m, UV depressed the total
carbon assimilation rates by 10?30%. In the deepest mixed sample (0?14 m depth),
assimilation rates in quartz bottles achieved 93% of the incubations with low UV.
Particulate 14C assimilation decreased by 20?25% in UV-bottles at mixing depths
up to 10 m. It was not significantly reduced at deepest mixing (0?14 m). Exudation
was enhanced by UV at 3.9 m mixing depth (120?134% of Duran bottles) and
declined during deeper mixing by about 29% (see Fig. 5). 
3. Inhibition model
The calculated relative production (P/Pmax) was highest in both quartz and Duran
bottles at depths of about 2 m (Fig. 6). Phytoplankton attained only 70% (quartz)
or 75% (Duran bottle) of the maximum production even in the optimum layer. The
differences between both bottle types were highest near the water surface and neg-
ligible below about 4 m.
The biological weighting function for L. Lucerne of Neale et al. (2001a, this issue)
was also used to calculate the time course of inhibition during vertical transport
through the underwater light field. Figure 7 illustrates the time courses of inhibition
factor E*inh and relative production P/Pmax during circular movement between water
surface and 14 m depth. The circular path of the bottles caused extended residence
periods near the water surface and at depth with rapid transitions in between. For
much of the cycle there was no UV, but photosynthesis was light limited. This was
followed by a relatively long episode near the surface with strong inhibition. As a
Table 1. Time and depth integrated doses and intensities of photosynthetically active radiation
(PAR), UV-A and UV-B received by vertically moved samples for 4:05 h (Sept. 13) and 4:00 h
(Sept. 15) incubation time
September 13 September 15
0?3.9 m 0?14 m 0?2 m 0?3.9 m 0?10 m
PAR [E m?2] 1.02 0.51 1.16 0.93 0.56
PAR [mE m?2 s?1] 692 345 809 643 386
UV-A [J cm?2] 10.6 4.96 13.6 9.65 5.31
UV-A [mW cm?2] 720 337 944 670 368
Relative time with
UV-A < 1 mW cm?2 0.0% 53.3% 0.0% 0.0% 42.6%
UV-B [J cm?2] 1.00 0.52 1.35 0.89 0.53
UV-B [mW cm?2] 68 35 94 62 37
Relative time with 
UV-B < 1 mW cm?2 40.6% 72.0% 0.0% 40.7% 67.5%
302 K?hler et al.
Figure 4. Average and standard deviation of total carbon assimilation (a), exudation (b) and par-
ticulate carbon assimilation (c) with high (quartz) and low (Duran) UV on Sept. 13 and 15, 1999
a
b
c
result, the impact of UV in the quartz bottles was stronger on P/Pmax (which includes
the effect of light limitation) than on 1/[1 +E*inh] when mixing depth exceeded 4 m
(Table 2).
The parameters of inhibition were integrated over time to allow for comparison
with measured data. The mean effective fluence rate for inhibition (E*inh) ranged
from 0.10?0.31 in Duran and from 0.15?0.5 in quartz bottles. The ratio of modelled
actual to potential photosynthesis (P/Pmax) declined with increasing mixing depth
UV effects in mixed water columns 303
and was higher in Duran than in quartz bottles (Table 2). The difference between
quartz and Duran bottles in computed photosynthesis ranged from 7.4?12.8%. 
Discussion
Photoinhibition by UVR was observed in the ocean (Helbling et al., 1994; Buma et
al., 1996) as well as in freshwater systems (Gala and Giesy, 1991; Kinzie et al., 1998).
The depth and stability of the vertical mixed layer was pointed out as an important
Table 2. Parameters of inhibition calculated with the BWF-PI model for phytoplankton of Lake
Lucerne incubated in quartz and Duran bottles. Light conditions and mode of sample movement
as on September 13 and 15, 1999
Depth range 0?2 m 0?3.9 m 0?3.9 m 0?10 m 0?14 m
Date Sept. 15 Sept. 13 Sept. 15 Sept. 15 Sept. 13
1/(1+E*Inh) Quartz 0.66 0.74 0.76 0.86 0.87
Duran 0.76 0.80 0.83 0.90 0.91
Q/D 87.5% 91.5% 92.1% 95.9% 96.3%
P/Pmax Quartz 0.61 0.58 0.62 0.43 0.32
Duran 0.70 0.64 0.68 0.47 0.35
Q/D 87.2% 90.3% 91.2% 92.6% 91.5%
Figure 5. Effects of UV on total C assimilation, exudation and particulate C assimilation of phy-
toplankton moved through different mixing depths (samples with low UV = 100%, the asterisks
indicate incubation at September 15, no marker = September 13, 1999). The modelled inhibition
parameter P/Pmax is given for comparison
304 K?hler et al.
regulating mechanism for photoinhibition of phytoplankton (Helbling et al., 1994;
Neale et al., 1998). 
In Lake Lucerne, Bossard et al. (2001, this issue) compared carbon assimilation
in bottles with and without UV at fixed depths between 0 and 5 m on 13 September,
1999. In this experiment, inclusion of UV-A + B caused an average decline in C
assimilation of about 25%. Depression was most severe in samples kept near the
water surface. However, results can not be directly compared with circulated bottles
because the fixed depth incubation used different spectral treatments (UV-A + B
transparent quartz and UV opaque plastic). We therefore applied the BWF-PI model
to compare production in fixed depth vs. circulated samples that are exposed to the
same average weighted irradiance. According to the model, the average E*Inh was
0.46 for the mid-depth (0?3.9 m) incubations on 13 September. A fixed depth incu-
bation at 1.2 m would have also experienced an E*Inh of 0.46 and relative inhibition
(1?1/[1+E*Inh]) of 32% compared to 26% calculated for the 0?3.9 m circulated
bottle. The 6% reduction in inhibition is not the result of a dynamic response 
(which is absent from the BWF-PI model) but instead from the non-linear (hyper-
bolic) relationship of photosynthetic response to UV (Lesser et al., 1994; Neale,
2000). Circulated bottles receive most of their UV exposure from high surface irra-
diance which is less efficient (per unit exposure) in inducing inhibition than moder-
ate irradiance. Likewise, a fixed depth incubation would receive an exposure of 0.23
at 2.1 m and exhibit a 18% inhibition compared to the 13% estimated for 0?14 m
circulated bottle, which received similar exposure. The actual inhibition at 1.2 m and
2.1 m during static incubation (estimated by interpolation from the standard profile
presented by Bossard et al., 2001, this issue) was 40% and 24%, respectively. This 
is a stronger response than was estimated by the BWF-PI model under the same 
Figure 6. Vertical profile of calculated relative production (P/Pmax) in quartz bottles with high UV
(solid line) and in Duran bottles with low UV (dotted line)
UV effects in mixed water columns 305
irradiance conditions, but the difference between model and observed is within
experiment error (Neale et al., 2001a, this issue). However, we can not exclude the
possibility that there may have been a cumulative effect of UV over 4 h static ex-
posure, a response that is not included in the BWF-PI model.
These results indicate that several mechanisms can cause more severe damage
by constant UV compared to fluctuating UV of the same dosage. The non-linear
relationship between inhibition and cumulative UV exposure in the Weddell-Scotia
Confluence (Neale et al., 1998) resulted in differing impacts of UV in modelled sta-
tic vs. circulating water columns depending on mixing depth. Depression of inte-
grated primary production was enhanced in shallow mixed layers but reduced in
deep mixed layers. The latter result is similar to our results for Lake Lucerne.
Kinzie et al. (1998) noticed a higher induction of UVR-absorbing mycosporine-like
amino acids in benthic communities (which were exposed to rather constant light
intensities) than in planktonic organisms (which received fluctuating light). The
importance of fluctuating light pulses of PAR was also pointed out by Flameling and
Kromkamp (1997). They demonstrated for cultures of Scenedesmus protuberans
that efficiency of photosynthesis decreases during (at the short-term) constant light
conditions but remained high at fluctuating light of the same dosage. Cullen and
Figure 7. Time course of calculated parameters of UV-inhibition in bottles with high UV (quartz,
solid line) and with low UV (Duran, dotted line) moved at a circular path between surface and 
14 m depth, September 15, 1999. Top: biologically effective fluence rate for inhibition of photo-
synthesis E*inh, bottom: relative production P/Pmax
306 K?hler et al.
Lesser (1991) speculated that a certain time period (in their experiments with a
marine diatom ca. 30 min) is necessary to balance UV damage and repair mecha-
nisms. They also stressed the importance of intensity and duration of UV exposure.
Short exposures to high intensities of UV could cause higher damage to phyto-
plankton than long exposures to low intensities. 
On the other hand, the BWF-PI model was based on measurements of photo-
synthesis made under constant polychromatic UV-PAR treatments during 1 h expo-
sures. The model does not include the dynamics of damage and repair under fluc-
tuating light. The nearly perfect agreement between inhibition rates measured
under fluctuating light and modelled rates (Fig. 5) seems to indicate surprisingly low
influence of dynamic adaptation. This could occur if the rate of dynamic adjustment
(time period to attain steady-state) was much faster in the Lake Lucerne assem-
blage than the 30 min observed by Cullen and Lesser (1991), also if any lag in inhi-
bition is matched by a corresponding lag in recovery.
Bottles excluding UV were not incubated at fixed depths below 5 m, so we do
not know the depth limit of UV effects under constant light conditions. The model
calculated similar photosynthesis in quartz and in Duran bottles at depths below 
4 m (Figure 6). At this depth, about 99% of incident UV-B were attenuated. In 
the mixing approach, UV-B inhibition was non-negligible up to a mixing depth of 
10 m. Gala and Giesy (1991) observed, that UV-inhibition of primary production
was restricted to the first 6 metres of Lake Michigan, which was 1% of UV-intensi-
ty at the lake surface. 
The UV dosage received by suspended algae depends on the ratio of transmis-
sion depth to mixing depth. In Lake Lucerne, the steep thermal gradient shown in
Figure 2 indicated a 6?7 m mixing depth during our experiments. Most phyto-
plankton was concentrated below the mixed layer at a depth of 6?10 m. This me-
talimnetic phytoplankton received about 1?2% of the UV-A and 0.01% of the UV-
B intensity of the water surface. It was much less affected by UV than the epilim-
netic phytoplankton which was subjected to about 45% of the incident UV-A and
30% of the UV-B during circulation in the upper 7 m. Metalimnetic phytoplankton
thus efficiently avoided UV damage as long as it was not mixed into the epilimnet-
ic layer. Any erosion of the metalimnion by increased wind speeds would cause sud-
denly increased UV exposure of the re-mixed algae. 
Intensity and depth of turbulent mixing of Lake Lucerne were not studied dur-
ing our experiments. Our lifts simulated circular paths of planktonic algae which
typically occur in Langmuir cells. Langmuir circulation is caused by wind of medi-
um speed in the upper layer of lakes and ocean. The speed of rotation is governed
by the wind speed and the mixing depth and may range from 10?60 min per turn
under conditions of Lake Lucerne (see Buranathanitt et al., 1982). This roughly met
the rotation frequency used in our experiments. 
Low phosphorus concentrations (TP < 1 mg l?1) could have further influenced the
sensitivity of phytoplankton. Karentz et al. (1994) speculated that phytoplankton
may be more affected by UV when cells are deficient in nutrients. 
A clear influence of UVR on photosynthetic extracellular release (PER) could
be detected. Especially in the upper mixed layers (0?4 m) an enhancement of
20?30% of PER in quartz bottles in comparison to Duran bottles indicates cell
damage or a change of DO14C-composition. Buma et al. (1996) noticed a change in
intracellular composition of phytoplankton during UV-exposure, especially an
increase of proteins under low and moderate levels of UVR, probably associated
with the biosynthesis of UV stress proteins. This might change PER too. Pausz and
Herndl (1999) noticed a decrease of PER by UV in a diatom culture in the same
range as total primary production is reduced, but no changes in percentage of PER
to total amount of fixed carbon. They further observed a changed uptake of PER
by bacteria under UV-exposure, indicating a qualitative change of PER. Kieber et
al. (1989) mentioned an increased release of labile DOC as a result of UV radiation,
probably stimulating bacterioplankton and shifting carbon flow in aquatic food
webs.
Our preliminary results indicate serious depression of carbon assimilation of
phytoplankton due to UVR even in well-mixed environments. Depression was low
at circulation through the whole euphotic zone and may disappear at even deeper
mixing. There was a tendency of lower photoinhibition per UV dosage at fluctuat-
ing than at constant light intensities. Response of phytoplankton to UV obviously
depends not only on intensity or dosage of radiation but also on its dynamics. The
latter is a function of depth, speed and type of mixing as well as of optical proper-
ties of the water. In order to understand UV effects on phytoplankton in well-mixed
environments, we have to study hydrodynamical conditions as well as kinetics of
photoinhibition and recovery in more detail. 
ACKNOWLEDGEMENTS
This study was carried out during the 7th International GAP Workshop, held on 9?17 Sept. 1999 in
Z?rich, Switzerland, and supported by the Swiss National Science Foundation (SNF), the Swiss
Academies of Natural and Technical Sciences (SANW and SATW), the Swiss Society of Hydrol-
ogy and Limnology (SGHL), by EAWAG, Z?rich Water Supply and the University of Z?rich, as
well as by Hoffmann-La Roche, Lonza, Novartis, Canberra Packard S.A, Millipore AG and Faust
Laborbedarf AG.
We would like to thank Peter Bossard for the good working conditions at Kastanienbaum and
the support of our field work. Daniel Steiner measured basic limnological parameters, global radi-
ation and pigment concentrations. Many thanks to him for his help in the radioisotope lab and dur-
ing field work.
REFERENCES
Behrendt, H., 1989. Devices for turbulence simulation during in-situ measurements of biological
processes in aquatic ecosystems. II. A bottle lift for simulation of vertical movement in natur-
al waters. Acta hydrophys. 33: 9?14.
Benndorf, J. and E. Werner, 1980. Berechnung der Wachstumsrate des Phytoplanktons in durch-
mischten Wasserk?rpern. Limnologica (Berlin) 12: 23?31.
Bossard, P., S. Gammeter, C. Lehmann, F. Schanz, R. Bachofen, H.R. B?rgi, D. Steiner and U.
Zimmermann, 2001. Limnological description of the Lakes Z?rich, Lucerne, and Cadagno.
Aquat. Sci. 63/3: 225?249.
B?hlmann, B., P. Bossard and U. Uehlinger, 1987. The influence of longwave ultraviolet radiation
(UV-A) on the photosynthetic activity (14C-assimilation) of phytoplankton. J. Plankton Res. 9:
935?943.
Buma, A.G.J., H.J. Zemmelink, K. Sjollema, W.W.C. Gieskes, 1996. UVB radiation modifies pro-
tein and photosynthetic pigment content, volume and ultrastructure of marine diatoms. Mar.
Ecol. Prog. Ser. 142: 47?54. 
Buranathanitt, T., D.J. Cockrell, P.H. John, 1982. Some effects of Langmuir circulation on the qual-
ity of water resource systems. Ecol. Modell. 15: 49?74.
UV effects in mixed water columns 307
B?rgi, H.R., C. Heller, S. Gaebel, N. Mookerji, and J.V. Ward, 1999. Strength of coupling between
phyto- and zooplankton in Lake Lucerne (Switzerland) during phosphorus abatement subse-
quent to a weak eutrophication. J. Plankton Res. 21: 485?507. 
Cullen, J.J. and M. Lesser, 1991. Inhibition of photosynthesis by ultraviolet radiation as a function
of dose and dosage rate: results for a marine diatom. Marine Biology 111: 183?190.
Cullen, J.J. and P.J. Neale, 1997. Biological weighting functions for describing the effects of ultra-
violet radiation on aquatic systems, pp. 97?118. In: D.-P. H?der [ed.] Effects of ozone deple-
tion on aquatic ecosystems. R. G. Landes.
Cullen, J.J., P.J. Neale and M.P. Lesser, 1992. Biological weighting function for the inhibition of
phytoplankton photosynthesis by ultraviolet radiation. Science 258: 646?650.
Flameling, I.A. and J. Kromkamp, 1997. Photoacclimation of Scenedesmus protuberans (Chlo-
rophyceae) to fluctuating irradiances simulating vertical mixing. J. Plankton Res. 19: 1011?
1024.
Gala, W.R. and J.P. Giesy, 1991. Effects of ultraviolet radiation on the primary production of nat-
ural phytoplankton assemblages in Lake Michigan. Ecotox. Environm. Safety 22: 345?361.
Helbling, E.W., V.E. Villafane and O. Holm-Hansen, 1994. Effects of ultraviolet radiation on
Antarctic marine phytoplankton photosynthesis with particular attention to the influence of
mixing. Ultraviolet radiation in Antarctica: Measurements and biological effects. Antarctic
research series 62: 207?227.
Jewson, D.H. and R.B. Wood, 1975. Some effects on integral photosynthesis of artificial circula-
tion of phytoplankton through light gradients. Verh. Intern. Verein. Limnol. 19: 1037?1044.
Karentz, D., R.B. Bothwell, A. Coffin, G.J. Hanson, G.J. Herndl, S.S. Kilham, M.P. Lesser, R.E.
Moeller, D.P. Morris, P.J. Neale, R.W. Sanders, C.S. Weiler and R.G. Wetzel, 1994. Impact of
UV-B radiation on pelagic freshwater ecosystems: Report of working group on bacteria and
phytoplankton. Arch. Hydrobiol. Beih. 43: 31?69.
Kieber, D.J., J. McDaniel and K. Mopper, 1989. Photochemical source of biological substrates in
seawater: implications for carbon cycling. Nature 341: 637?639.
Kinzie, R., A.T. Banaszak and M. Lesser, 1998. Effects of UV-radiation on primary productivity in
a high altitude tropical lake. Hydrobiologia 385: 23?32.
Lesser, M.P., J.J. Cullen and P.J. Neale, 1994. Carbon uptake in a marine diatom during acute expo-
sure to ultraviolet B radiation: relative importance of damage and repair. J. Phycol. 30:
183?192.
Marra, J., 1978. Phytoplankton photosynthetic response to vertical movement in a mixed layer.
Marine Biology 46: 203?208.
Mitchell, D.L. and D. Karentz, 1993. The induction and repair of DNA photodamage in the envi-
ronment. In: Young, A.R., L. Bj?rn, J. Moan and W. Nultsch (ed): Plenum Press: Environ-
mental UV photobiology. 245?377. 
Neale, P.J., E. Litchman, C. Sobrino, C. Callieri, G. Morabito, V. Montecino, Y. Huot, P. Bossard,
C. Lehmann and D. Steiner, 2001a. Quantifying the response of phytoplankton photosynthe-
sis to ultraviolet radiation: Biological weighting functions versus in situ measurements in two
Swiss lakes. Aquat. Sci. 63/3: 265?285.
Neale, P.J., P. Bossard and Y. Huot, 2001b. Incident and in situ irradiance in Lakes Cadagno and
Lucerne: A comparison of methods and models. Aquat. Sci. 63/3: 250?264.
Neale, P.J., 2000. Spectral weighting functions for quantifying the effects of ultraviolet radiation in
marine ecosystems, pp. 73?100. In: S.J. de Mora, S. Demers and M. Vernet [eds.], The effects
of UV radiation on marine ecosystems. Cambridge Univ. Press.
Neale, P.J., M.P. Lesser and J.J. Cullen, 1994. Effects of ultraviolet radiation on the photosynthe-
sis of phytoplankton in the vicinity of McMurdo Station (78?S), pp. 125?142. In: C.S. Weiler
and P.A. Penhale [eds.], Ultraviolet Radiation in Antarctica: Measurements and Biological
Effects. Am. Geophysical Union.
Neale, P.J., R.F. Davis and J.J. Cullen, 1998. Interactive effects of ozone depletion and vertical mix-
ing on photosynthesis of Antarctic phytoplankton. Nature 392: 585?589.
Nixdorf, B. and H. Behrendt, 1991. Discrepancies between O2- and 14C-method resulting from
measurements of primary production under consideration of vertical mixing in a shallow
eutrophic lake. Verh. Intern. Verein. Limnol. 24: 1268?1271.
Pahl-Wostl, C. and D.M. Imboden, 1990. DYPHORA ? a dynamic model for the rate of photo-
synthesis of algae. J. Plankton Res. 12: 1207?1221.
308 K?hler et al.
Pausz, C. and G.J. Herndl, 1999. Role of ultraviolet radiation on phytoplankton extracellular
release and its subsequent utilization by marine bacterioplankton. Aquat. Microb. Ecol. 18:
85?93.
Vincent, W.F. and P.J. Neale, 2000. Mechanisms of UV damage in aquatic organisms, pp. 149?176.
In: S.J. de Mora, S. Demers and M. Vernet [eds.], The effects of UV radiation on marine
ecosystems. Cambridge Univ. Press.
Zagarese, H., W. Cravero, P. Gonzalez and F. Pedrozo, 1998a. Copepod mortality induced by fluc-
tuating levels of natural ultraviolet radiation simulating vertical water mixing. Limnol.
Oceanogr. 43: 169?174.
Zagarese, H., B. Tartarotti, W. Cravero and P. Gonzalez, 1998b. UV damage in shallow lakes: the
implications of water mixing. J. Plankton Res. 20: 1423?1433.
Received 3 March 2000;
revised manuscript accepted 22 September 2000.
UV effects in mixed water columns 309