ArticlePDF Available

Phytoplankton growth and the interaction of light and temperature: A synthesis at the species and community level

Authors:

Abstract and Figures

Temperature strongly affects phytoplankton growth rates, but its effect on communities and ecosystem processes is debated. Because phytoplankton are often limited by light, temperature should change community structure if it affects the traits that determine competition for light. Furthermore, the aggregate response of phytoplankton communities to temperature will depend on how changes in community structure scale up to bulk rates. Here, we synthesize experiments on 57 phytoplankton species to analyze how the growth-irradiance relationship changes with temperature. We find that light-limited growth, light-saturated growth, and the optimal irradiance for growth are all highly sensitive to temperature. Within a species, these traits are co-adapted to similar temperature optima, but light-limitation reduces a species' temperature optimum by ∼5°C, which may be an adaptation to how light and temperature covary with depth or reflect underlying physiological correlations. Importantly, the maximum achievable growth rate increases with temperature under light saturation, but not under strong light limitation. This implies that light limitation diminishes the temperature sensitivity of bulk phytoplankton growth, even though community structure will be temperature-sensitive. Using a database of primary production incubations, we show that this prediction is consistent with estimates of bulk phytoplankton growth across gradients of temperature and irradiance in the ocean. These results indicate that interactions between temperature and resource limitation will be fundamental for explaining how phytoplankton communities and biogeochemical processes vary across temperature gradients and respond to global change.
Content may be subject to copyright.
Phytoplankton growth and the interaction of light and temperature:
A synthesis at the species and community level
Kyle F. Edwards,*
1
Mridul K. Thomas,
2
Christopher A. Klausmeier,
3
Elena Litchman
4
1
Department of Oceanography, University of Hawaii, Honolulu, Hawaii
2
Department of Aquatic Ecology, Eawag: Swiss Federal Institute of Aquatic Science and Technology, Zurich, Switzerland
3
Department of Plant Biology, Kellogg Biological Station, Michigan State University, Hickory Corners, Michigan
4
Department of Zoology, Kellogg Biological Station, Michigan State University, Hickory Corners, Michigan
Abstract
Temperature strongly affects phytoplankton growth rates, but its effect on communities and ecosystem
processes is debated. Because phytoplankton are often limited by light, temperature should change commu-
nity structure if it affects the traits that determine competition for light. Furthermore, the aggregate response
of phytoplankton communities to temperature will depend on how changes in community structure scale up
to bulk rates. Here, we synthesize experiments on 57 phytoplankton species to analyze how the growth-
irradiance relationship changes with temperature. We find that light-limited growth, light-saturated growth,
and the optimal irradiance for growth are all highly sensitive to temperature. Within a species, these traits
are co-adapted to similar temperature optima, but light-limitation reduces a species’ temperature optimum
by 58C, which may be an adaptation to how light and temperature covary with depth or reflect underlying
physiological correlations. Importantly, the maximum achievable growth rate increases with temperature
under light saturation, but not under strong light limitation. This implies that light limitation diminishes
the temperature sensitivity of bulk phytoplankton growth, even though community structure will be
temperature-sensitive. Using a database of primary production incubations, we show that this prediction is
consistent with estimates of bulk phytoplankton growth across gradients of temperature and irradiance in
the ocean. These results indicate that interactions between temperature and resource limitation will be fun-
damental for explaining how phytoplankton communities and biogeochemical processes vary across temper-
ature gradients and respond to global change.
Global warming has underscored the need to understand
how temperature affects organisms, populations, commun-
ities, and ecosystems. Predicting the ecological effects of
temperature is difficult, in part because populations are typi-
cally limited by competition or predation, and so to predict
growth or abundance we need to know how temperature
modulates the multiple physiological processes that underlie
species interactions (Vasseur and McCann 2005; Kordas et al.
2011; O’Connor et al. 2011). For example, simple scaling
relationships for the temperature-dependence of ecosystem
processes may only apply when resources are not limiting
(Xu et al. 2004; L
opez-Urrutia and Mor
an 2007; De Castro
and Gaedke 2008). Although resource limitation and other
processes complicate the role of temperature, there still may
be general rules for how temperature modulates physiology
and species interactions, and quantifying such rules will
enhance our ability to explain ecosystem responses to tem-
perature gradients (Dell et al. 2014). Because light and nu-
trient limitation strongly affect primary producers, it is
essential to characterize any general patterns for how tem-
perature interacts with limitation by these resources.
In this study, we synthesize monoculture experiments
that characterize the interactive effects of light and tempera-
ture on phytoplankton growth. Phytoplankton contribute
nearly half of global primary production, are the base of the
food web in aquatic environments, and play a critical role in
the feedbacks of the global carbon cycle to anthropogenic
forcing (Falkowski et al. 1998; Field et al. 1998). Phytoplank-
ton are very sensitive to environmental change (Doney et al.
2012), and both temperature and irradiance are among the
key environmental drivers whose distribution is predicted to
continue changing in the future (De Stasio et al. 1996; Boyd
et al. 2015). The temperature and irradiance that
*Correspondence: kfe@hawaii.edu
Additional Supporting Information may be found in the online version of this
article.
1232
LIMNOLOGY
and
OCEANOGRAPHY Limnol. Oceanogr. 61, 2016, 1232–1244
V
C2016 Association for the Sciences of Limnology and Oceanography
doi: 10.1002/lno.10282
phytoplankton experience tend to be positively correlated,
because solar radiation increases water temperature, and
increased temperature drives stratification and shoals the
mixed layer, thus increasing average irradiance experienced
by phytoplankton. Nonetheless, phytoplankton occur over a
wide range of temperature-irradiance combinations, includ-
ing low irradiance at the deep chlorophyll maximum or
below in warm waters (Fennel and Boss 2003; Cullen 2015),
and saturating irradiance in shallow mixed layers in cold
waters (e.g., due to meltwater in the summer in polar
regions; Lancelot et al. 1993). Therefore, understanding the
processes that control individual growth, community struc-
ture, and primary production requires us to understand
how light and temperature interact, i.e., how temperature
modulates the growth-irradiance relationship and how light
modulates the growth-temperature relationship. Whether
temperature has important direct effects on phytoplankton
growth and community size structure in the ocean is cur-
rently debated (Mor
an et al. 2010; Mara~
n
on et al. 2012,
2014; Regaudie-de-Goux and Duarte 2012), and conflicting
results may in part be driven by interactions between tem-
perature and resource limitation.
The independent effects of irradiance and temperature on
phytoplankton growth have been intensively studied and are
well-characterized. Growth increases nearly linearly at low
irradiance, saturates at some optimal irradiance for growth,
and then declines due to photoinhibition (Langdon 1988;
Talmy et al. 2013; Edwards et al. 2015). Interspecific differen-
ces in this relationship are thought to be due to differences
in pigment content, respiratory costs, cell size, and pathways
for photoprotection and repair of photodamage (Langdon
1988; Six et al. 2007). Temperature responses are also unimo-
dal, typically with a left-skew such that growth increases
exponentially or linearly from low temperature, and declines
more rapidly above the optimum (Eppley 1972; Montagnes
et al. 2003; Thomas et al. 2012). Interspecific differences in
this relationship are thought to be due to differences in pro-
tein structure (particularly the stability of enzymes, which is
related to specificity and reaction rates), lipid composition of
cell membranes, and chaperone protein production (Clarke
2003; Kingsolver 2009). For both irradiance and temperature
responses, differences between genotypes or species meas-
ured in the lab have been correlated with differences in dis-
tributions across depths, seasons, or latitudes (Rodr
ıguez
et al. 2005; Johnson et al. 2006; Thomas et al. 2012, 2016;
Edwards et al. 2013a,b).
The interactive effects of temperature and irradiance on
phytoplankton have been studied in many experiments (e.g.,
Dauta 1982; Verity 1982; Palmisano et al. 1987), but there is
currently no clear consensus for how growth-irradiance rela-
tionships change with temperature, or how thermal optima
change with irradiance. It is often expected that light-
limited photosynthesis and growth will be less sensitive to
temperature than light-saturated rates, due to limitation of
photosynthesis by photon absorption at low irradiance, but
contradictory results have been observed (Raven and Geider
1988; Davison 1991; Nicklisch et al. 2008). Phytoplankton
growth is often modeled as an exponential function of tem-
perature under all resource conditions (Blackford et al. 2004;
Taucher and Oschlies 2011), which assumes that resource-
saturated growth has the same monotonic temperature sensi-
tivity as light- or nutrient-limited growth. In contrast, the
initial slope of the chlorophyll-specific photosynthesis-irradi-
ance curve is sometimes modeled as temperature-insensitive,
while the maximum rate of photosynthesis is given an expo-
nential temperature dependence (Geider et al. 1998; Moore
et al. 2002). Importantly, the way in which temperature
effects are modeled has large effects on projections of global
primary production under climate change (Sarmiento et al.
2004; Taucher and Oschlies 2011).
Many pressing ecological questions require us to “scale
up” from community diversity and dynamics to aggregate
ecosystem processes. For example, to understand the role of
the biosphere in the global carbon cycle we need to know
how complex communities respond to multiple environ-
mental factors, and how community structure determines
aggregate processes like primary production or carbon export
to the deep ocean (Duffy and Stachowicz 2006; Boyd et al.
2015; Worden et al. 2015). Responses to temperature are an
area where the difference between individual and aggregate
outcomes are significant: even though individual species
exhibit unimodal responses to temperature (Thomas et al.
2012; Dell et al. 2014), bulk ecosystem rates typically change
monotonically with temperature, i.e., they do not decline
above some optimum. This difference between species and
community response was explained in an influential paper
by Eppley (1972), which compiled measurements of phyto-
plankton growth rate as a function of temperature. Although
individual species exhibited unimodal responses to tempera-
ture, the highest observed growth rates across species as a
function of temperature increased exponentially. He charac-
terized this with an exponential curve, l50.59 310
0.02753T
,
where lis specific growth rate (d
21
) and Tis temperature
(8C). This curve is equivalent to a Q
10
of 1.88, i.e., growth
increases by a factor of 1.88 when temperature increases by
108C (a recent update using more data by Bissinger et al.
(2008) found an essentially identical exponent, l50.81 3
10
0.02743T
,orQ
10
51.88). This result implies that species will
replace one another along a temperature gradient via compe-
tition, with the result that phytoplankton whole-community
growth rate increases monotonically with temperature, if the
maximum possible growth rate is higher for species adapted
to higher temperatures (Eppley 1972; Norberg 2004; Bis-
singer et al. 2008). We will refer to the predicted whole-
community curve, derived from the upper envelope of the
single-species curves, as a trait envelope (Fig. 1A). Impor-
tantly, it is possible that the shape or slope of this envelope
changes as a function of resource limitation (Fig. 1B).
Edwards et al. Light–temperature interactions
1233
The diversity of experimental results and modeling
approaches highlights the need for a synthesis of data on
the light–temperature interaction, which can address several
important questions: Are there general rules for how light
and temperature interact to determine growth, and how con-
sistent are these patterns across a diversity of species? How
might community structure be affected by the temperature
responses of traits that determine competition? Can the tem-
perature scaling of aggregate ecosystem processes be pro-
jected from trait variation across species adapted to different
conditions? To address these questions, here we quantify
how the parameters of the growth-irradiance curve change
with temperature, and how the optimal temperature for
growth changes with irradiance, for 57 marine and fresh-
water phytoplankton species. We quantify whether different
light utilization traits have different temperature sensitiv-
ities, and whether different traits of the same species are co-
adapted to the same temperature. Finally, we use a compila-
tion of field-based primary production incubations to ask
whether the effects of light and temperature on aggregate
growth of natural phytoplankton communities is similar to
what is predicted from lab-based trait measurements of spe-
cies adapted to different temperature and irradiance regimes.
Methods
Data compilation
Previously we compiled from the literature comprehensive
datasets of phytoplankton temperature traits (Thomas et al.
2012, 2016) and light utilization traits (Schwaderer et al.
2011; Edwards et al. 2015). While compiling these studies we
gathered a subset of experiments that measured growth rate
of a single phytoplankton isolate across a factorial manipula-
tion of temperature and irradiance. In the current analysis,
we include only those experiments where at least four irradi-
ance levels and four temperature levels were used. The
median range of temperatures used in an experiment was
178C, the narrowest range was 78C and the broadest was
308C. In all experiments, nutrients were not strongly limit-
ing, and the cultures were acclimated to irradiance and tem-
perature treatments before growth rate was measured. These
criteria yielded 59 experiments from 29 publications on 57
unique species (31 freshwater and 26 marine; Supporting
Information Tables A1, A2, A3). Taxonomically the species
include 19 diatoms, 18 chlorophytes, 6 cyanobacteria, 3 hap-
tophytes, 3 cryptophytes, 2 dinoflagellates, 2 desmids, and 2
chrysophytes. For all analyses, we pool marine and fresh-
water species to simplify presentation of the results. In our
dataset marine species tend to have lower temperature
optima on average, but preliminary analyses showed that
the interaction between temperature and irradiance, which
is the focus of this study, did not vary between these groups.
Supplementary plots show the major results coded by fresh-
water or marine origin (Supporting Information Figs. A1–
A3). Scripts for all statistical models used in the analysis are
included as Supporting Information.
Growth-irradiance and growth-temperature curves
To characterize how light utilization changes with tem-
perature, for each experiment we fit a growth-irradiance
curve to the measurements from each temperature level. We
used the following curve:
lIðÞ5lmaxI
lmax
aI2
opt
I21122lmax
aIopt

I1lmax
a
(1)
where lis the specific growth rate (d
21
) as a function of the
photon flux density (I,lmol photons m
22
s
21
), l
max
is the
maximum growth rate achieved at I
opt
, the optimal irradi-
ance, and ais the initial slope of the curve. In other words,
Fig. 1. Scaling from individual to whole-community temperature responses. (A) Under saturating irradiance, individual species/genotypes exhibit
unimodal responses to temperature (solid lines), but competitive species sorting leads to an exponential response of the aggregate growth rate across
a temperature gradient (dashed line). (B) Under light limitation, maximal growth may not increase at higher temperatures; aggregate community
rates may then be insensitive to temperature even if individual species are sensitive. Note that the axes for (A) and (B) are on different scales.
Edwards et al. Light–temperature interactions
1234
as I!0, lIðÞ!aI. This curve was derived by Eilers and Pee-
ters (1988) from a dynamic model of photoinhibition of
photosynthesis, and it allows us to compare across species
the relative performance under limiting irradiance (a), rela-
tive performance under saturating irradiance (l
max
), and the
irradiance above which photoinhibition reduces growth
(I
opt
). As reviewed previously (Edwards et al. 2015), theory
and experiments show that these parameters determine com-
petitive ability and coexistence under a variety of irradiance
regimes. For example, for species with equal loss rates, a
should determine competitive ability under chronically low
irradiance (such as that experienced at a deep chlorophyll
maximum). In mixed layers, a,l
max
, and I
opt
may all be
important for competitive outcomes, depending on incident
irradiance and the depth of mixing (Huisman and Weissing
1994; Huisman et al. 1999; Gerla et al. 2011).
Equation 1 does not include a parameter for maintenance
respiration, i.e., the growth rate when irradiance is zero.
However, maintenance respiration is typically low for phyto-
plankton (0.02 d
21
, Geider and Osborne 1989), and nega-
tive growth rates were only observed in four growth-
irradiance relationships (out of 327 total). In addition, 95
experiments used four irradiance levels, and we did not wish
to estimate a four-parameter curve from four observations.
Therefore, we used the three-parameter curve and removed
the four relationships with negative growth rates from the
analysis. The fitted curves are given in Supporting Informa-
tion S1. As described previously (Edwards et al. 2015), we
used Eq. 1 instead of the curve of Platt et al. (1980) because
the two curves have a very similar shape, but Eq. 1 is para-
meterized in terms of the traits we wish to compare across
species, and often fits the data slightly better.
To compare optimum temperature for growth as a func-
tion of irradiance, we fit the following curve:
lTðÞ512T2z
W=2

aebT (2)
where zis the midpoint of the growth curve, Wis the width
of the unimodal response to temperature, and aand b
jointly determine the overall height, steepness, and skewness
of the curve (Norberg 2004; Thomas et al. 2012). We esti-
mated the optimum temperature (T
opt
) from the fitted curve
by numerical optimization. For subsequent analysis, we use
only those growth-temperature relationships where the esti-
mated optimum is at least 58C from the highest experimen-
tal temperature.
Response of light utilization traits to temperature
Exploration of the growth-irradiance curves showed that
all three traits (a,l
max
,I
opt
) exhibit substantial variation
with temperature for nearly all species (Supporting Informa-
tion S2). For each trait about half of the relationships are
unimodal, and most of the remainder are monotonically
increasing, while a few are monotonically decreasing or
essentially flat. To characterize the typical shape of the tem-
perature responses and compare the sensitivity to tempera-
ture across the three traits, we took two approaches. The first
approach was to quantify how steeply the trait values rise
and fall with temperature, by breaking each curve into rising
and falling portions. The second approach was to character-
ize the mean shape of the curve using a nonparametric
smoother.
To characterize the rising portion of the curve, for each
experiment we selected the trait values measured at or below
the temperature of the maximum trait value; we only
included experiments for which there were at least two val-
ues at temperatures below the maximum, yielding a total of
at least three values. To quantify the typical shape of the ris-
ing curve, we fit a generalized additive mixed model
(GAMM) where the trait value was a non-parametric smooth
function of temperature, and a random effect for species was
included to account for the fact that species differ in their
mean trait (across temperatures). Because species have differ-
ent temperature optima, we standardized the temperatures
so that all species had their trait maximum at the same posi-
tion on the temperature axis (set to 0). We also fit a linear
mixed model with log(trait) as the response, which is equiva-
lent to assuming that the trait increases exponentially with
temperature. Using the fitted slope we calculated a Q
10
coef-
ficient for the trait. We then repeated this whole procedure
for the falling portion of the curve, again only using experi-
ments where there were at least two trait values at tempera-
tures above the temperature of the maximum trait value.
The second approach was to characterize the typical shape
of the whole temperature response curve for each trait. For
this analysis, we used only those species for which the maxi-
mum trait value was not at the highest or lowest tempera-
ture (i.e., the relationship appears unimodal). Again we used
a GAMM with a random effect for species, and we standar-
dized the temperature axis so that all species had their maxi-
mum trait value at the same position (set to 0). For all of the
above analyses, we only used I
opt
estimates when the esti-
mated I
opt
was less than the maximum irradiance used in
the experiment; otherwise there was not sufficient data to
estimate I
opt
. When comparing I
opt
across temperatures, we
only used experiments where I
opt
could be estimated for at
least four temperatures.
Response of temperature optima to irradiance
To characterize how the optimal growth temperature
changes with irradiance, we fit a GAMM with T
opt
as the
response variable, a smoother for the effect of irradiance,
and a random effect for species to account for differences
between species in mean T
opt
across irradiances.
Comparison of temperature optima across traits
To test whether different traits have similar temperature
optima, we compared the temperature of the maximal trait
values across species. We will refer to these respectively as
Edwards et al. Light–temperature interactions
1235
Ta
opt ,Tl
opt , and TI
opt . It should be noted that a higher aor l
max
are always beneficial, all else equal, while a higher I
opt
will
reduce photoinhibition but also reduce growth at lower irradi-
ances. Nonetheless, as shown below TI
opt is correlated with
Ta
opt and Tl
opt, suggesting that species exhibit higher I
opt
at
temperatures to which they are best adapted. We performed
standardized major axis regression (SMA; Warton et al. 2006)
for Ta
opt vs. Tl
opt,TI
opt vs. Tl
opt,andTa
opt vs. TI
opt. For these analy-
ses, we only compared temperature optima when at least one
of the optima was not at the highest or lowest temperature
measured. Our rationale is that if two traits both peak at the
highest temperature in the experiment (or the lowest temper-
ature), there is not sufficient information to ask whether these
traits have similar optima. However, if at least one trait shows
a unimodal relationship to temperature, then we can ask
whether the two traits peak at the same temperature or not.
Trait envelopes
To understand how the interaction between irradiance and
temperature will affect whole-community growth, we quanti-
fied trait envelopes (Fig. 1) for a,l
max
,andI
opt
as a function
of temperature, by performing quantile regression on the trait
data from all species. For each trait we fit a model where the
90
th
percentile of log(trait) is a linear function of temperature;
this is equivalent to assuming that the 90
th
percentile
increases exponentially with temperature. We used the 90
th
percentile because we are interested in the upper envelope of
trait variation, but higher percentiles tend to have low statisti-
cal confidence. For comparison we also fit an ordinary least
squares regression. In preliminary analyses, we also fit non-
parametric curves to the 90
th
percentile of the data (using
GAMLSS, generalized additive models for location, scale, and
shape; Rigby and Stasinopoulos 2005) and to the mean of the
data (using a generalized additive model, GAM). However, we
found that the relationships only deviated from linear at
extreme temperatures for which there was less data, and so
we present the linear fits here for simplicity.
Light–temperature interactions in field incubations
To compare the patterns found in the trait envelope anal-
yses to whole-community growth in natural systems, we
used the extensive compilation of 24,000 marine primary
production observations compiled by Behrenfeld and
colleagues (http://www.science.oregonstate.edu/ocean.
productivity/field.data.c14.readme.php). This compilation
contains
14
C uptake measurements, from field incubations of
2–24 h duration (only a small percentage are <6 h), taken
over depth profiles (range 0–175 m) at >1600 stations across
a wide range of productivity and latitude (Behrenfeld and Fal-
kowski 1997). In addition to daily carbon fixation, the dataset
includes chlorophyll concentration, surface PAR, incubation
PAR, latitude, longitude, date, and sea surface temperature.
This information can be used to calculate chlorophyll-specific
daily primary production. If the chlorophyll-to-carbon ratio
(Chl:C) of the phytoplankton were known (it is not), then
phytoplankton growth rate could be approximated as the
carbon-specific rate of daily carbon fixation (Eppley 1972;
Mara~
n
on 2005). For our purposes, we are more interested in
how the interaction between irradiance and temperature
causes relative changes in growth than the absolute magni-
tude of growth. Therefore, we took two approaches to the
issue of Chl:C, and the fact that Chl:C may change with irra-
diance and temperature (Cloern et al. 1995). (1) Convert Chl-
specific production to specific growth rate, using a Chl:
C ratio of 0.01, which is an intermediate value based on
estimates for ocean phytoplankton (Behrenfeld et al. 2005);
(2) assume that Chl:C varies according to the model of Beh-
renfeld et al. (2005), which is an empirical model of how
Chl:C changes with irradiance and temperature, based
on remote sensing of bulk chlorophyll and phytoplankton
biomass. The model is Chl:C 5Chl:C
min
1(Chl:C
max
Chl:C
min
)e
23I
,withChl:C
min
50.017 – 0.00045T,and
Chl:C
max
50.015 10.00005e
0.215T
, and where Tis temperature
(8C) and Iis daily irradiance (mol quanta m
22
h
21
). Thus,
Chl:C declines exponentially with irradiance, and the range
of potential Chl:C increases with temperature. Although our
approach yields only a rough estimate of phytoplankton
growth rate, the large number of observations over a wide
range of irradiance and temperature conditions allows us to
ask whether the trait envelope predictions from our monocul-
ture compilation are consistent with light–temperature inter-
actions in natural communities.
Because we are interested in effects of irradiance and tem-
perature, we excluded observations where nutrients were
likely to be strongly limiting. We excluded values where
nitrate concentration is expected to be <0.5 lmol L
21
based
on the World Ocean Atlas (Garcia et al. 2014), and we
excluded values from >608Nor>408S, which are likely to
be iron-limited (Moore et al. 2013). Finally, because only sea
surface temperature is reported in the dataset, we only used
values from the mixed layer, based on a climatology of
mixed layer depth (de Boyer Mont
egut et al. 2007), and we
used SST to approximate temperature for all samples within
the mixed layer at each station. These criteria resulted in
2090 observations for analysis.
To quantify how temperature and irradiance interact to
affect our proxy of phytoplankton growth, we fit linear
regressions of log(growth) vs. temperature for low-light sam-
ples (<20 lmol photons m
22
s
21
) and for sufficient-light
samples (between 100 lmol photons m
22
s
21
and 200 lmol
photons m
22
s
21
). We also fit a generalized additive model
with a two-dimensional smoother for the interactive effect
of irradiance and temperature.
Results
For all three growth-irradiance traits (a,l
max
,I
opt
), nearly
all species exhibit substantial variation with temperature
(Supporting Information Fig. A1). About half of the
Edwards et al. Light–temperature interactions
1236
relationships were unimodal, and most of the remainder are
monotonically increasing, while a few are monotonically
decreasing or essentially flat. Both the rising and falling por-
tions of the curve could be approximated by an exponential
relationship, for all three traits (i.e., a linear relationship
between log(trait) and temperature; Fig. 2). Although an expo-
nential relationship is certainly a simplification, because the
full unimodal relationships are flatter near the optimum (Fig.
2C,F,I), the near-exponential rising and falling portions are
useful to characterize the sensitivity of these traits to tempera-
ture. Combining the data across species, the estimated Q
10
values for the rising curves are 2.64, 1.90, and 1.73 for a,
l
max
,andI
opt
, respectively (95% confidence intervals for Q
10
are [2.17, 3.19], [1.77, 2.03], and [1.62, 1.85], respectively).
The estimated Q
10
values for proportional decrease along the
falling curves are 2.26, 2.38, and 1.51 for a,l
max
,andI
opt
,
respectively (95% confidence intervals for Q
10
are [1.90, 2.69],
[1.67, 3.42], and [1.32, 1.72], respectively).
Comparison of the temperatures at which growth-
irradiance traits reach their maximum values shows that the
three traits tend to be co-adapted to similar temperatures
(Figs. 3A–C, 2C,F,I). For the temperature of peak avs. the
temperature of peak l
max
, SMA regression has an intercept of
25.52 (95% CI: [211.2,0.17]), a slope of 1.06 (95% CI:
Fig. 2. Rising portion (A, D, G), falling portion (B, E, H), and full curve (C, F, I) for the three growth-irradiance traits as a function of temperature.
The rising and falling portions are fit both as linear regressions with log(trait) as the response (dashed lines) and generalized additive models (solid
lines with 95% confidence bands). The plotted points are corrected, using the fitted GAMM, to remove differences between species in the mean trait
value across temperatures. The x-axis uses temperature values that have been standardized, such that each species reaches its maximum trait value at
08, as described in Methods.
Edwards et al. Light–temperature interactions
1237
[0.86,1.32]), and R
2
50.48. For the temperature of peak I
opt
vs. the temperature of peak l
max
, SMA regression has an
intercept of 23.08 (95% CI: [28.6,2.43]), a slope of 0.96
(95% CI: [0.77,1.2]), and R
2
50.71. For the temperature of
peak avs. the temperature of peak I
opt
, SMA regression has
an intercept of 23.12 (95% CI: [29.5,3.2]), a slope of 1.06
(95% CI: [0.83, 1.36]), and R
2
50.59. Therefore, the slopes of
these relationships are not significantly different from 1,
while the intercept for avs. l
max
is likely lower than 0, indi-
cating that atends to peak at a lower temperature than l
max
.
It may also be the case that I
opt
tends to peak at a lower tem-
perature than l
max
, because 11 values in Fig. 3B are below
the 1:1 line, while only 2 are above the 1:1 line.
An analysis of how the optimal growth temperature (T
opt
)
changes with irradiance (Fig. 3D) is consistent with the differ-
ences in Fig. 3A,B. The value of T
opt
increases by about 48C
from the lowest irradiance to about 100 lmol photons
m
22
s
21
, and then decreases by 1–28C at the highest irradiance.
Acomparisonofavalues for all species across temperatures
shows that the upper limit on adoes not change with tempera-
ture (Fig. 4A). The regression slope for the 90
th
percentile of
log
10
avs. temperature is 20.0024 (95% CI: [20.0066, 0.015]).
Fig. 3. (A–C) Comparison across species of temperature optima for aand l
max
,I
opt
and l
max
, and aand I
opt
, respectively. (D) Optimal temperature
for growth as a function of irradiance. The y-axis in this plot is relative T
opt
, which substracts the mean value of T
opt
for each species, to better visualize
how T
opt
changes with irradiance for all species. The fitted smoother is from a generalized additive model.
Edwards et al. Light–temperature interactions
1238
Likewise, an OLS regression through the middle of the data has
a weakly increasing slope not different from zero (0.0068, 95%
CI: [20.00033, 0.014]). In contrast, the upper limit and mean
of l
max
increase with temperature (Fig. 4B). The 90
th
percentile
slope for log
10
l
max
vs. temperature is 0.015 (95% CI: [0.011,
0.019]), which corresponds to a Q
10
of 1.42 (95% CI: [1.28,
1.56]), and the OLS slope is also 0.015 (95% CI: [0.011, 0.018]).
Finally, the upper limit on I
opt
alsotendstoincreasewithtem-
perature (Fig. 4C). The 90
th
percentile slope for log
10
I
opt
vs.
temperature is 0.012 (95% CI: [0.0052, 0.014]), which corre-
sponds to a Q
10
of 1.33 (95% CI: [1.13, 1.39]), and the OLS
slope is 0.0088 (95% CI: [0.0042, 0.013]).
An analysis of field estimates of primary production shows
that the temperature sensitivity of growth depends on irradi-
ance. When irradiance is 20 lmol photons m
22
s
21
,Chl-
specific C uptake shows no trend with temperature (Fig. 4D;
regression slope of log
10
(growth) vs. temperature 50.004, 95%
CI: [20.008, 0.016]). In contrast, when irradiance is between
100 lmol photons m
22
s
21
and 200 lmol photons m
22
s
21
,
Chl-specific C uptake increases with temperature (Fig. 4E;
regression slope of log
10
(growth) vs. temperature 50.031, 95%
CI: [0.023, 0.038]; equivalent Q
10
52.04, 95% CI: [1.71, 2.42]).
Finally, in the field data the optimal irradiance for growth
tends to increase with temperature, changing from about 100
lmol photons m
22
s
21
to 200 lmol photons m
22
s
21
when
moving from the lowest to highest temperatures. This is most
readily seen by fitting a 2D smoother for the effect of irradiance
and temperature (Fig. 4F).
The patterns in Fig. 4D–F use a constant Chl:C of 0.01 to
convert from Chl-specific to C-specific values. If a variable
Chl:C is used instead, derived from the remote sensing
model of Behrenfeld et al. (2005), the results are very similar
(Supporting Information Fig. A4).
Discussion
Our compilation shows a substantial effect of tempera-
ture on the growth-irradiance relationship. Each of the
Fig. 4. (A–C) Trait envelopes for the three growth-irradiance traits across temperatures. The solid line is the quantile regression fit to the 90
th
percen-
tile of all trait data for all species, the dashed line is an ordinary least squares fit to the mean of the data. (D) Whole community growth rate vs. tem-
perature, at irradiances below 20 lmol photons m
22
s
21
. Specific growth rate is approximated using Chl-specific primary production (
14
C uptake),
with a fixed Chl: C of 0.01. (E) Whole community growth vs. temperature, at irradiances between 100 lmol photons m
22
s
21
and 200 lmol photons
m
22
s
21
.(F) Two-dimensional smoother from a GAM fit to whole community growth as a function of temperature and irradiance.
Edwards et al. Light–temperature interactions
1239
growth-irradiance parameters exhibits a unimodal tempera-
ture response that has a fairly consistent shape across spe-
cies, with different species possessing different temperature
optima (Figs. 1, 2). The temperature sensitivities of the traits
are comparable, with each trait exhibiting a mean Q
10
between 1.5 and 2.6 on both the rising and falling portions
of the response. For the rising portion of the curve, ahas a
Q
10
that is greater than the often-used values of 1.88 or 2
(95% CI: [2.17, 3.19]), while l
max
has a Q
10
that overlaps
these values (95% CI: [1.77, 2.03]), and I
opt
has a lower Q
10
(95% CI: [1.62, 1.85]). This means that for individual speces,
the light-limited growth rate is at least as sensitive to tem-
perature as the light-saturated growth rate, and susceptibility
to photoinhibition also changes significantly with tempera-
ture. For any particular species the three traits tend to have
similar temperature optima, but on average the optimum for
ais 58C lower than the optimum for l
max
, and the opti-
mum for I
opt
lies between these. Likewise, the optimal tem-
perature for growth increases by about 48C from the lowest
irradiance to about 100 lmol photons m
22
s
21
, and then
decreases by 1–28at the highest irradiance.
Temperature will change community structure in part by
altering the values of traits that determine resource (e.g.,
light) competition. The responses of a,I
opt
and l
max
to tem-
perature imply that competitive ability under either chronic
light limitation such as at a DCM (where aapproximates
competitive ability) or fluctuating light limitation such as in
a mixed layer (where a,l
max
, and I
opt
may contribute to
competitive ability) will be temperature-sensitive, leading to
distinct temperature niches under competition for light. It
may also be the case that a transition from saturating to limit-
ing irradiance causes species’ temperature niches to shift
toward cooler temperatures, i.e., it decreases their ability to tol-
erate higher temperatures. This shift could be adaptive,
because temperature and irradiance are positively correlated
over depth, and over time in seasonal environments. In addi-
tion, Thomas et al. (2012) found that T
opt
for marine isolates
tends to be 48C higher than mean SST at the isolation loca-
tion, which may be due to widespread (co)limitation of growth
by irradiance, which reduces T
opt
below that measured under
sufficient irradiance. Nutrient limitation may have a similar
effect on species’ thermal optima (Thomas et al. unpubl.).
These predictions can be tested in field and lab experiments by
quantifying how community composition changes in response
to factorial combinations of light and temperature.
Ecosystem processes such as primary production and ele-
ment cycling depend on aggregate community responses to
environmental forcing. Predicting aggregate responses to tem-
perature (or other factors) is more challenging than predicting
species- or genotype-level responses, because aggregate pat-
terns emerge from the outcome of complex interactions
among diverse actors. If a single trait determines competitive
outcomes along an environmental gradient, then in theory
the upper envelope of that trait will quantify how aggregate
function changes along the gradient (Norberg 2004). The
compiled monoculture data shows that neither the upper
envelope nor the mean of achanges across temperatures
(Fig. 4A). In contrast, the upper envelope and mean of l
max
both increase exponentially with temperature, consistent with
previous findings (Eppley 1972; Bissinger et al. 2008), and I
opt
also increases with temperature (Fig. 4B,C). From these pat-
terns, we can predict that whole-community growth should
be temperature-insensitive under strong light limitation but
temperature-sensitive under saturating light, and intermediate
light limitation will yield a dampened temperature sensitivity.
In addition, the optimal irradiance should increase with
temperature. It is interesting that afor individual species is
quite temperature-sensitive, while the upper envelope for ais
not, which suggests substantial species turnover with little
ecosystem-level effect across a temperature gradient, at the
lowest irradiances (Fig. 1B). The analysis of primary produc-
tion data is consistent with the predictions from trait enve-
lopes. Whole-community growth (approximated as Chl-
specific C uptake) is temperature-insensitive under low irradi-
ance, but increases exponentially with temperature under
moderate or high irradiance, and the optimal irradiance
increases modestly with temperature (Fig. 4D–F). Although
14
C incubation data is a rough proxy of the actual bulk
growth rate, it is encouraging that predictions from the lab-
measured trait data are consistent with field patterns.
One difference between the lab and field patterns is the
slope of the growth response under saturating irradiance,
which has a Q
10
of 1.42 for the trait envelope but a Q
10
of
2.04 in the field data. This may be due to insufficient sam-
pling of species adapted to high temperatures in the trait
compilation or biases in the field data that change with tem-
perature; using the 99
th
percentile of the data instead of the
90
th
percentile yields a similar Q
10
of 1.44. Prior compila-
tions of maximal growth rates using a larger number of spe-
cies have found a Q
10
of 1.88 (Eppley 1972; Bissinger et al.
2008), which is closer to the slope observed in the field data
and supports the upper envelope for l
max
as a predictor of
light-saturated growth rate. Studies applying the “metabolic
theory of ecology” have argued that primary production has
a temperature sensitivity largely determined by the activa-
tion energy of Rubisco (0.32 eV), which is thought to limit
the light-saturated rate of photosynthesis (Allen et al. 2005;
L
opez-Urrutia et al. 2006). This activation energy corre-
sponds approximately to a Q
10
of 1.64, which is intermediate
between the lab and field sensitivities under saturating light
in our study. Our results indicate that the metabolic theory
approach is likely not suitable for phytoplankton when light
limits growth. Nonetheless, it is possible that under suffi-
cient light the temperature response of phytoplankton
growth is driven by Rubisco. However, it is not clear how to
reconcile this enzyme-specific view with unimodal species-
level responses, which scale up to produce monotonic
Edwards et al. Light–temperature interactions
1240
aggregate responses. Furthermore, this framework does not
account for responses to excess irradiance.
Prior studies have found conflicting results for the effect of
temperature on aggregate phytoplankton growth or specific
primary production. Chen et al. (2012) found that growth rate
in marine dilution experiments increases exponentially with
temperature, and Regaudie-de-Gioux and Duarte (2012) found
the same using a compilation of Chl-specific gross primary
production in the open ocean. In contrast, Mara~
n
on et al.
(2014) found that resource supply across ocean regions
explains phytoplankton growth with little direct role for tem-
perature, and De Castro and Gaedke (2008) found that sea-
sonal variation in Chl-specific photosynthesis in Lake
Constance was unrelated to temperature. Our results suggest
that conflicting patterns can be reconciled by accounting for
resource supply, with light limitation diminishing the
temperature-sensitivity of bulk growth. Nutrient limitation
may have a similar dampening effect, as seen in mesocosm
experiments (Staehr and Sand-Jensen 2006; O’Connor et al.
2009), although we currently lack sufficient culture studies to
make predictions based on trait envelopes for nutrient compe-
tition. Due to the ubiquity of resource limitation (or co-limita-
tion) in marine and fresh waters, it will be essential to better
quantify how temperature and resources interact, and how
temperature also modulates grazers and pathogens (e.g., Chen
et al. 2012). It will also be important to better quantify tem-
perature effects on photosynthesis vs. respiration, particularly
under low irradiance. The data compiled here did not permit
an analysis of respiratory costs, but if respiration has a greater
temperature sensitivity than light-limited photosynthesis,
this could have important effects on patterns of net primary
production (e.g., L
opez-Urrutia et al. 2006).
Based on our data synthesis, we make some recommenda-
tions below for modeling phytoplankton growth at the spe-
cies and community levels. Models of phytoplankton growth
often use a simple exponential term to account for the effect
of temperature (Blackford et al. 2004; Taucher and Oschlies
2011). If the phytoplankton variable is intended to represent
bulk phytoplankton, or aggregate growth of a diverse func-
tional group, then such a monotonic temperature effect is
appropriate, but only under sufficient irradiance. Therefore,
it would be appropriate to use a functional form where the
maximum growth rate has an exponential or Arrhenius-type
temperature dependence, but the initial slope of the irradi-
ance response is temperature-insensitive. An interaction
between irradiance and temperature is a feature of the photo-
acclimation models of Geider et al. (1997, 1998), which have
been used in a variety of biogeochemical models (e.g., Moore
et al. 2002; Stock et al. 2014). In these models, the maximum
Chl-specific rate of photosynthesis (PC
m) increases with tem-
perature but the initial slope of Chl-specific photosynthesis
vs. irradiance (a
chl
) is insensitive to temperature. The Chl:C
ratio (h) also depends on temperature via a dependence on
PC
m. Under these assumptions, specific growth rate is essen-
tially temperature-insensitive at 1 lmol photons m
22
s
21
,
weakly temperature-sensitive at 10 lmol photons m
22
s
21
,
and strongly temperature sensitive at 100 lmol photons
m
22
s
21
(Supporting Information Fig. A5). Therefore, use of
this model to represent aggregate growth is largely consistent
with observed light-temperature interactions, although the
temperature-sensitivity may arise at too low of an irradiance,
and, in addition, the model does not account for photoinhi-
bition. In contrast to monotonic temperature effects, models
of individual phytoplankton populations, or models that
include a diversity of phytoplankton species (e.g., Dutkiewicz
et al. 2013), would be made most realistic by incorporating
unimodal temperature functions for growth-irradiance param-
eters. A diversity of species adapted to different temperature
regimes can be implemented by making the parameters fol-
low the trait envelopes in Fig. 4. An example of a growth-
irradiance model with temperature-dependent parameters is
given in Supporting Information S3.
Through effects on stratification, global warming is
expected to decrease nutrient supply to the euphotic zone
while alleviating light limitation (Sarmiento et al. 2004). The
role of the direct effects of temperature on plankton, and
how these interact with resource limitation, are less clear.
The results presented here suggest that the interaction of
temperature and irradiance is substantial, with consequences
for the niches of individual species, the structure of com-
munities, and key ecosystem rates. Important next steps
include testing these patterns in the field, integrating these
interactions with effects of CO
2
, nutrients, and grazers, and
incorporating the empirical patterns in ecosystem models.
References
Allen, A. P., J. F. Gillooly, and J. H. Brown. 2005. Linking
the global carbon cycle to individual metabolism. Funct.
Ecol. 19: 202–213. doi:10.1111/j.1365-2435.2005.00952.x
Behrenfeld, M., and P. G. Falkowski. 1997. A consumer’s
guide to phytoplankton primary productivity models.
Limnol. Oceanogr. 42: 1479–1491. doi:10.1029/
2004GB002299
Behrenfeld, M. J., E. Boss, D. A. Siegel, and D. M. Shea. 2005.
Carbon-based ocean productivity and phytoplankton
physiology from space. Global Biogeochem. Cycles 19:
GB1006. doi:10.1029/2004GB002299
Bissinger, J. E., D. J. S. Montagnes, and J. Sharples. 2008. Pre-
dicting marine phytoplankton maximum growth rates
from temperature: Improving on the Eppley curve using
quantile regression. Limnol. Oceanogr. 53: 487–493. doi:
10.4319/lo.2008.53.2.0487
Blackford, J. C., J. I. Allen, and F. J. Gilbert. 2004. Ecosystem
dynamics at six contrasting sites: A generic modelling study.
J. Mar. Syst. 52:191215.doi:10.1016/j.jmarsys.2004.02.004
Boyd, P., S. Lennartz, D. Glover, and S. Doney. 2015. Biologi-
cal ramifications of climate-change-mediated oceanic
Edwards et al. Light–temperature interactions
1241
multi-stressors. Nat. Clim. Chang. 5: 71–79. doi:10.1038/
NCLIMATE2441
Chen, B., M. R. Landry, B. Huang, and H. Liu. 2012. Does
warming enhance the effect of microzooplankton grazing
on marine phytoplankton in the ocean? Limnol. Ocean-
ogr. 57: 519–526. doi:10.4319/lo.2012.57.2.0519
Clarke, A. 2003. Costs and consequences of evolutionary
temperature adaptation. Trends Ecol. Evol. 18: 573–581.
doi:10.1016/j.tree.2003.08.007
Cloern, J., C. Grenz, and L. Vidergar-Lucas. 1995. An empiri-
cal model of the phytoplankton chlorophyll: Carbon
ratio-the conversion factor between productivity and
growth rate. Limnol. Oceanogr. 40: 1313–1321. doi:
10.4319/lo.1995.40.7.1313
Cullen, J. J. 2015. Subsurface chlorophyll maximum layers:
Enduring enigma or mystery solved? Ann. Rev. Mar. Sci.
7: 207–239. doi:10.1146/annurev-marine-010213-135111
Dauta, A. 1982. Conditions de d
eveloppement du phyto-
plancton:
Etude comparative du comportement de huit
esp
eces en culture. I. D
et
ermination des parametres de
croissance en fonction de la lumi
ere et de la temp
erature.
Ann. Limnol. 18: 217–262. doi:10.1051/limn/1982005
Davison, I. 1991. Environmental effects on algal photosyn-
thesis: Temperature. J. Phycol. 27: 2–8. doi:10.1111/
j.0022-3646.1991.00002.x
de Boyer Mont
egut, C., J. Mignot, A. Lazar, and S. Cravatte.
2007. Control of salinity on the mixed layer depth in the
world ocean: 1. General description. J. Geophys. Res. 112:
C06011. doi:10.1029/2006JC003953
De Castro, F., and U. Gaedke. 2008. The metabolism of lake
plankton does not support the metabolic theory of ecol-
ogy. Oikos 117: 1218–1226. doi:10.1111/j.2008.0030-
1299.16547.x
DeStasio,B.,D.K.Hill,J.M.Kleinhans,N.P.Nibbelink,
and J. J. Magnuson. 1996. Potential effects of glo-
bal climate change on small north-temperate lakes:
Physics, fish, and plankton. Limnol. Oceanogr. 41:
1136–1149. doi:10.4319/lo.1996.41.5.1136
Dell, A. I., S. Pawar, and V. M. Savage. 2014. Temperature
dependence of trophic interactions are driven by asymme-
try of species responses and foraging strategy. J. Anim.
Ecol. 83: 70–84. doi:10.1111/1365-2656.12081
Doney, S. C., and others. 2012. Climate change impacts on
marine ecosystems. Ann. Rev. Mar. Sci. 4: 11–37. doi:
10.1146/annurev-marine-041911-111611
Duffy, J. E., and J. J. Stachowicz. 2006. Why biodiversity is
important to oceanography: Potential roles of genetic,
species, and trophic diversity in pelagic ecosystem proc-
esses. Mar. Ecol. Prog. Ser. 311: 179–189. doi:10.3354/
meps311179
Dutkiewicz, S., J. R. Scott, and M. J. Follows. 2013. Winners
and losers: Ecological and biogeochemical changes in a
warming ocean. Global Biogeochem. Cycles 27: 463–477.
doi:10.1002/gbc.20042
Edwards, K. F., E. Litchman, and C. A. Klausmeier. 2013a.
Functional traits explain phytoplankton community
structure and seasonal dynamics in a marine ecosystem.
Ecol. Lett. 16: 56–63. doi:10.1111/ele.12012
Edwards, K. F., E. Litchman, and C. A. Klausmeier. 2013b.
Functional traits explain phytoplankton responses to
environmental gradients across lakes of the United States.
Ecology 94: 1626–1635. doi:10.1890/12-1459.1
Edwards, K. F., M. K. Thomas, C. A. Klausmeier, and E.
Litchman. 2015. Light and growth in marine phytoplank-
ton: Allometric, taxonomic, and environmental variation.
Limnol. Oceanogr. 60: 540–552. doi:10.1002/lno.10033
Eilers, P. H. C., and J. C. H. Peeters. 1988. A model for the
relationship between light intensity and the rate of pho-
tosynthesis in phytoplankton. Ecol. Modell. 42: 199–215.
doi:10.1016/0304-3800(88)90057-9
Eppley, R. 1972. Temperature and phytoplankton growth in
the sea. Fish. Bull. 70: 1063–1085.
Falkowski, P., R. Barber, and V. Smetacek. 1998. Biogeochem-
ical controls and feedbacks on ocean primary production.
Science 281: 2002206. doi:10.1126/science.281.5374.200
Fennel, K., and E. Boss. 2003. Subsurface maxima of phyto-
plankton and chlorophyll: Steady-state solutions from a
simple model. Limnol. Oceanogr. 48: 1521–1534. doi:
10.4319/lo.2003.48.4.1521
Field, C., M. Behrenfeld, J. Randerson, and P. Falkowski.
1998. Primary production of the biosphere: Integrating
terrestrial and oceanic components. Science 281: 237–
240. doi:10.1126/science.281.5374.237
Geider, R. J., and B. A. Osborne. 1989. Respiration and
microalgal growth: A review of the quantitative relation-
ship between dark respiration and growth. New Phytol.
112: 327–341. doi:10.1111/j.1469-8137.1989.tb00321.x
Geider, R. J., H. L. MacIntyre, and T. M. Kana. 1997.
Dynamic model of phytoplankton growth and acclima-
tion: Responses of the balanced growth rate and the chlo-
rophyll a:carbon ratio to light, nutrient-limitation and
temperature. Mar. Ecol. Prog. Ser. 148: 187–200. doi:
10.3354/meps148187
Geider, R., H. L. MacIntyre, and T. Kan. 1998. A dynamic
regulatory model of phytoplanktonic acclimation to light,
nutrients, and temperature. Limnol. Oceanogr. 43: 679–
694. doi:10.4319/lo.1998.43.4.0679
Gerla, D. J., W. M. Mooij, and J. Huisman. 2011. Photoinhi-
bition and the assembly of light-limited phytoplankton
communities. Oikos 120: 359–368. doi:10.1111/j.1600-
0706.2010.18573.x
Huisman, J., and F. J. Weissing. 1994. Light-limited growth
and competition in well-mixed aquatic environments: An
elementary model. Ecology 75: 507–520. doi:10.2307/
1939554
Huisman, J., R. R. Jonker, C. Zonneveld, and F. Weissing.
1999. Competition for light between phytoplankton spe-
cies: Experimental tests of mechanistic theory. Ecology
Edwards et al. Light–temperature interactions
1242
80: 211–222. doi:10.1890/0012-9658(1999)080[0211:
CFLBPS]2.0.CO;2
Johnson, Z. I., E. M. S. Woodward, and S. W. Chisholm.
2006. Niche partitioning among Prochlorococcus ecotypes
along ocean-scale environmental gradients. Science 311:
1737–1740. doi:10.1126/science.1118052
Kingsolver, J. G. 2009. The well-temperatured biologist.
(American Society of Naturalists Presidential Address).
Am. Nat. 174: 755–768. doi:10.1086/648310
Kordas, R. L., C. D. G. Harley, and M. I. O’Connor. 2011.
Community ecology in a warming world: The influence
of temperature on interspecific interactions in marine sys-
tems. J. Exp. Mar. Biol. Ecol. 400: 218–226. doi:10.1016/
j.jembe.2011.02.029
Lancelot, C., and others. 1993. Factors controlling phyto-
plankton ice-edge blooms in the marginal ice-zone of the
northwestern Weddell Sea during sea ice retreat 1988:
Field observations and mathematical modelling. Polar
Biol. 13: 377–387. doi:10.1007/BF01681979
Langdon, C. 1988. On the causes of interspecific differences
in the growth-irradiance relationship for phytoplankton.
II. A general review. J. Plankton Res. 10: 1291–1312. doi:
10.1093/plankt/10.6.1291
L
opez-Urrutia, A., E. San Martin, R. P. Harris, and X.
Irigoien. 2006. Scaling the metabolic balance of the
oceans. Proc. Natl. Acad. Sci. USA 103: 8739–8744. doi:
10.1073/pnas.0601137103
L
opez-Urrutia,A.,andX.A.G.Mor
an. 2007. Resource limita-
tion of bacterial production distorts the temperature depend-
ence of oceanic carbon cycling. Ecology 88: 817–822. doi:
10.1890/06-1641
Mara~
n
on, E. 2005. Phytoplankton growth rates in the Atlan-
tic subtropical gyres. Limnol. Oceanogr. 50: 299–310. doi:
10.4319/lo.2005.50.1.0299
Mara~
n
on, E., P. Cerme~
no, M. Latasa, and R. D. Tadonl
ek
e.
2012. Temperature, resources, and phytoplankton size
structure in the ocean. Limnol. Oceanogr. 57: 1266–1278.
doi:10.4319/lo.2012.57.5.1266
Mara~
n
on, E., P. Cerme~
no, M. Huete-Ortega, D. C. L
opez-
Sandoval, B. Mouri~
no-Carballido, and T. Rodr
ıguez-
Ramos. 2014. Resource supply overrides temperature as a
controlling factor of marine phytoplankton growth. PLoS
One 9: e99312. doi:10.1371/journal.pone.0099312
Montagnes, D. J. S., S. A. Kimmance, and D. Atkinson. 2003.
Using Q10: Can growth rates increase linearly with tem-
perature? Aquat. Microb. Ecol. 32: 307–313. doi:10.3354/
ame032307
Moore, J., S. Doney, and J. Kleypas. 2002. An intermediate
complexity marine ecosystem model for the global
domain. Deep-Sea Res. II 49: 403–462.
Moore, C. M., and others. 2013. Processes and patterns of
oceanic nutrient limitation. Nat. Geosci. 6: 701–710. doi:
10.1038/ngeo1765
Mor
an,X.A.G.,
A. L
opez-Urrutia, A. Calvo-D
ıaz, and W.
K. W. Li. 2010. Increasing importance of small phyto-
plankton in a warmer ocean. Glob. Chang. Biol. 16:
1137–1144. doi:10.1111/j.1365-2486.2009.01960.x
Nicklisch, A., T. Shatwell, and J. Kohler. 2008. Analysis and
modelling of the interactive effects of temperature and
light on phytoplankton growth and relevance for the
spring bloom. J. Plankton Res. 30: 75–91. doi:10.1093/
plankt/fbm099
Norberg, J. 2004. Biodiversity and ecosystem functioning: A
complex adaptive systems approach. Limnol. Oceanogr.
49: 1269–1277. doi:10.4319/lo.2004.49.4_part_2.1269
O’Connor, M. I., M. F. Piehler, D. M. Leech, A. Anton, and
J. F. Bruno. 2009. Warming and resource availability shift
food web structure and metabolism. PLoS biology 7:
e1000178.
O’Connor, M. I., B. Gilbert, and C. J. Brown. 2011. Theoretical
predictions for how temperature affects the dynamics of
interacting herbivores and plants. Am. Nat. 178:626638.
doi:10.1086/662171
Palmisano, A. C., J. B. SooHoo, and C. W. Sullivan.
1987. Effects of four environmental variables on
photosynthesis-irradiance relationships in Antarctic sea-
ice microalgae. Mar. Biol. 306: 2992306. doi:10.1007/
BF00392944
Platt, T., C. L. Gallegos, and W. G. Harrison. 1980. Photoin-
hibition of photosynthesis in natural assemblages of
marine phytoplankton. J. Mar. Res. 38: 103–111.
Raven, J., and R. Geider. 1988. Temperature and algal
growth. New Phytol. 110: 441–461. doi:10.1111/j.1469-
8137.1988.tb00282.x
Regaudie-de-Gioux, A., and C. M. Duarte. 2012. Temperature
dependence of planktonic metabolism in the ocean.
Global Biogeochem. Cycles 26: GB1015. doi:10.1029/
2010GB003907
Rigby, R., and D. Stasinopoulos. 2005. Generalized additive
models for location, scale, and shape. J Roy Statist Soc.
Series C 54: 507–554.
Rodr
ıguez, F., E. Derelle, L. Guillou, F. Le Gall, D. Vaulot,
and H. Moreau. 2005. Ecotype diversity in the marine
picoeukaryote Ostreococcus (Chlorophyta, Prasinophyceae).
Environ. Microbiol. 7: 853–859. doi:10.1111/j.1462-
2920.2005.00758.x
Sarmiento, J. L., R. Slater, R. Barber, L. Bopp, S. C. Doney, a.
C. Hirst, J. Kleypas, R. Matear, U. Mikolajewicz, P.
Monfray, V. Soldatov, S. a. Spall, and R. Stouffer. 2004.
Response of ocean ecosystems to climate warming. Global
Biogeochem Cy 18.
Schwaderer, A. S., K. Yoshiyama, P. de Tezanos Pinto, N.
G. Swenson, C. A. Klausmeier, and E. Litchman. 2011.
Eco-evolutionary differences in light utilization traits
and distributions of freshwater phytoplankton. Limnol.
Oceanogr. 56:589598.doi:10.4319/lo.2011.56.2.0589
Edwards et al. Light–temperature interactions
1243
Six, C., Z. V. Finkel, A. J. Irwin, and D. A. Campbell. 2007.
Light variability illuminates niche-partitioning among
marine Picocyanobacteria. PLoS One 2: e1341. doi:
10.1371/journal.pone.0001341
Staehr, P. A., and K. Sand-Jensen. 2006. Seasonal changes
in temperature and nutrient control of photosynthesis,
respiration and growth of natural phytoplankton com-
munities. Freshw. Biol. 51: 249–262. doi:10.1111/j.1365-
2427.2005.01490.x
Stock, C. A., J. P. Dunne, and J. G. John. 2014. Global-scale
carbon and energy flows through the marine planktonic
food web: An analysis with a coupled physical–biological
model. Prog. Oceanogr. 120: 1–28. doi:10.1016/
j.pocean.2013.07.001
Talmy, D., J. Blackford, N. J. Hardman-Mountford, A. J.
Dumbrell, and R. J. Geider. 2013. An optimality model of pho-
toadaptation in contrasting aquatic light regimes. Limnol.
Oceanogr. 58: 1802–1818. doi:10.4319/lo.2013.58.5.1802
Taucher, J., and A. Oschlies. 2011. Can we predict the direc-
tion of marine primary production change under global
warming? Geophys. Res. Lett. 38: L02603. doi:10.1029/
2010GL045934
Thomas, M. K., C. T. Kremer, C. A. Klausmeier, and E.
Litchman. 2012. A global pattern of thermal adaptation
in marine phytoplankton. Science 338: 1085–1088. doi:
10.1126/science.1224836
Thomas, M. K., C. T. Kremer, and E. Litchman. 2016. Envi-
ronment and evolutionary history determine the global
biogeography of phytoplankton temperature traits. Glob.
Ecol. Biogeogr. 25: 75–86.
Vasseur, D. A., and K. S. McCann. 2005. A mechanistic
approach for modeling temperature-dependent consumer-
resource dynamics. Am. Nat. 166: 184–198.
Verity, P. 1982. Effects of temperature, irradiance, and day-
length on the marine diatom Leptocylindrus danicus Cleve.
IV. Growth. J. Exp. Mar. Biol. Ecol. 60: 209–222. doi:
10.1016/0022-0981(82)90160-5
Warton, D. I., I. J. Wright, D. S. Falster, and M. Westoby.
2006. Bivariate line-fitting methods for allometry. Biol.
Rev. Camb. Philos. Soc. 81: 259–291. doi:10.1017/
S1464793106007007
Worden, A. Z., M. J. Follows, S. J. Giovannoni, S. Wilken, A.
E. Zimmerman, and P. J. Keeling. 2015. Rethinking the
marine carbon cycle: Factoring in the multifarious life-
styles of microbes. Science 347: 735.
Xu, L., D. D. Baldocchi, and J. Tang. 2004. How soil mois-
ture, rain pulses, and growth alter the response of ecosys-
tem respiration to temperature. Global Biogeochem.
Cycles 18: GB4002. doi:10.1029/2004GB002281
Acknowledgments
We thank two anonymous reviewers for helpful comments. This
research was in part supported by the NSF grants OCE-0928819 and
CBET-1134215 to EL and CAK and DEB 0845932 to EL. This is Kellogg
Biological Station contribution #1915.
Submitted 6 November 2015
Revised 3 February 2016
Accepted 8 February 2016
Associate editor: Heidi Sosik
Edwards et al. Light–temperature interactions
1244
... where μ max is the maximum growth rate (d À1 ) achieved at I opt , I is the irradiance (μmol photons m À2 s À1 ), and α is the initial slope of the growth-irradiance curve (Edwards et al. 2016). Key traits that characterize diatom growth responses to light are the α, μ max , and I opt . ...
... where μ(T) is the temperature-dependent growth rate (day À1 ), z is the midpoint of the growth curve, ω is the width of the unimodal response to temperature, and a and b determine the height, steepness, and skewness of the curve (Edwards et al. 2016). Key temperature-related traits are the optimum temperature for growth, T opt , minimum T min , and maximum T max temperatures that have nonnegative growth and bracket the thermal niche of an organism and the growth rate at T opt . ...
... Meta-analyses of existing data can also help assess the interdependencies of traits. Edwards et al. (2016) showed that light-related traits depend on temperature, and vice versa, temperature traits depend on irradiance. For example, the optimum temperature for growth T opt is a unimodal function of irradiance, with the highest T opt at about 100 μmol quanta m À2 s À1 and declining at low and high irradiances. ...
Chapter
Full-text available
... Furthermore, changes in light availability and nutrient loading may modify plant nutrient content, the carbon (C):nutrient [nitrogen (N) and phosphorus (P)] composition, and growth rates (Gu et al., 2018;Velthuis et al., 2017;Zhang et al., 2020). However, during winter, when temperatures are low, algae growth may be temperature limited (Edwards et al., 2016), and eutrophication may thus have less impact during these early stages of submerged macrophyte establishment. ...
... Although shading reduced growth of P. crispus, biomass buildup of the other primary producers was promoted. It could be expected that reduced light availability may decrease primary producer biomass (Edwards et al., 2016;Karlsson et al., 2009). ...
Article
Full-text available
Submerged macrophytes play a key role in maintaining a clear‐water phase and promoting biodiversity in shallow aquatic ecosystems. Since their abundance has declined globally due to anthropogenic activities, it is important to include them in aquatic ecosystem restoration programs. Macrophytes establishment in early spring is crucial for the subsequent growth of other warm‐adapted macrophytes. However, factors affecting this early establishment of submerged macrophytes have not been fully explored yet. Here, we conducted an outdoor experiment from winter to early spring using the submerged macrophytes Potamogeton crispus and Vallisneria spinulosa to study the effects of shading, nutrient loading, snail herbivory (Radix swinhoei), and their interactions on the early growth and stoichiometric characteristics of macrophytes. The results show that the effects strongly depend on macrophyte species. Biomass and number of shoots of P. crispus decreased, and internode length increased during low light conditions, but were not affected by nutrient loading. P. crispus shoot biomass and number showed hump‐shaped responses to increased snail biomass under full light. In contrast, the biomass of the plant linearly decreased with snail biomass under low light. This indicates an interaction of light with snail herbivory. Since snails prefer grazing on periphyton over macrophytes, a low density of snails promoted growth of P. crispus by removing periphyton competition, while herbivory on the macrophyte increased during a high density of snails. The growth of V. spinulosa was not affected by any of the factors, probably because of growth limitation by low temperature. Our study demonstrates that the interaction of light with snail herbivory may affect establishment and growth of submerged macrophytes in early spring. Macrophyte restoration projects may thus benefit from lowering water levels to increase light availability and making smart use of cold‐adapted herbivores to reduce light competition with periphyton. Our study demonstrates that the interactions of light with snail herbivory, but not nutrient loading, may affect establishment and growth of submerged macrophytes in early spring. Our study thus has implications for recovery of submerged macrophytes in deteriorative water bodies.
... Phytoplankton growth rates increase with light and temperature up to optimal values [14], and decrease thereafterslowly for light, very rapidly for temperature i.e. they are oppositely skewed unimodal functions [15][16][17]. The optimal temperature and light conditions for P. rubescens are poorly constrained based on existing experiments, but Oberhaus et al. (2007) measured growth rates at 15 & 25°C and showed that it was faster at 25°C. ...
Article
Full-text available
Chytrids are important drivers of aquatic ecosystems as phytoplankton parasites. The interaction between these parasites and their hosts are shaped by abiotic factors such as temperature and light. Here, we performed a full-factorial experiment to study how temperature and light interact to affect the dynamics of the bloom-forming toxic cyanobacterium Planktothrix rubescens and its chytrid parasite. We used a dynamic host-parasite model to explore how temperature and light affect long term dynamics. At low temperatures, chytrids do not survive. Higher light and temperature levels stimulated both phytoplankton and chytrid growth, with complex effects on their dynamics. Model exploration indicates that increasing temperature and light shifts equilibrium outcomes from P. rubescens persisting alone to stable coexistence and then to limit cycles. This provides an alternative biological explanation for why P. rubescens is mainly found in the relatively cold and dark lake metalimnion - it may enable avoidance of its parasite. Our study emphasizes the importance of investigating how abiotic factors interact with biotic interactions to drive complex outcomes. ISME Communications; https://doi.
... In contrast to cell volume, the data support the assumption that growth rates increase with temperature across taxa and the E a values are in the expected range (0.2-1.2 eV, average $ 0.6 eV: (Gillooly et al. 2001;Brown et al. 2004;Chen and Laws 2017). Unfortunately, compared to phytoplankton (Thomas et al. 2012;Edwards et al. 2015Edwards et al. , 2016, the available experimental data are scarce and do not allow for inferring thermal responses from the organisms' temperature reaction norm and growth rates at the respective species-specific temperature optima (Supporting Information Table S2). Furthermore, the dataset is limited regarding temperature responses of marine ciliates at both ends of the temperature range (Fig. 2). ...
Article
Full-text available
On the global average, the temperature increase in the ocean is lower than in lakes. Moreover, most freshwater organisms must cope with wider temperature fluctuations than marine organisms. Knowing if the organisms' thermal sensitivity differs in the two realms is crucial for predicting the respective climate-related changes at community and ecosystem levels. We investigated the thermal sensitivity of planktonic ciliates, which are of tremendous significance for biogeochemical cycling in all aquatic ecosystems. Marine and freshwater ciliates differ in their thermal performance; therefore, system-specific activation energies should be applied in models predicting ciliate responses to altered temperatures. This work may serve as a model study for other taxa and be of interest to many marine and freshwater ecologists. Abstract Predicting the performance of aquatic organisms in a future warmer climate depends critically on understanding how current temperature regimes affect the organisms' growth rates. Using a meta-analysis for the published experimental data, we calculated the activation energy (E a) to parameterize the thermal sensitivity of marine and freshwater ciliates, major players in marine and freshwater food webs. We hypothesized that their growth rates increase with temperature but that ciliates dwelling in the immense, thermally stable ocean are closely adapted to their ambient temperature and have lower E a than ciliates living in smaller, thermally more variable freshwater environments. The E a was in the range known from other taxa but significantly lower for marine ciliates (0.390 AE 0.105 eV) than for freshwater ciliates (0.633 AE 0.060 eV), supporting our hypothesis. Accordingly, models aiming to predict the ciliate response to increasing water temperature should apply the environment-specific activation energies provided in this study.
... On a global scale, the temperature is considered an important explanatory variable for the rate of oxygen consumption, phytoplankton nutrient uptake (López-Urrutia and Morán 2015), metabolism, growth rates, survival, distribution, biomass (Chen et al. 2017;Okbah et al. 2017), and community structure (Edwards et al. 2016;Zalat et al. 2021). In the present study, the high temperature limited the growth of Euglenophyceae and Conjugatophyceae, while it enhanced the Dinophyceae and Mediophyceae. ...
Article
Full-text available
El-Mex Bay is a semi-sheltered coastal embayment located west of Alexandria, the Egyptian Mediterranean Sea. It is considered as a hot spot area receiving industrial, agricultural, and domestic wastes from the adjacent Lake Mariut through El-Umum Drain. To evaluate the water quality of the bay, phytoplankton and macroalgae as biological tools were investigated seasonally concerning physicochemical parameters from 2018 to 2019. The results indicated that the Bay harbored a diversified phytoplankton community (162 species), comprising 99 freshwater forms, 38 marine, 22 euryhaline, and only 3 brackish forms. The total average phytoplankton abundance was 419,414 unit/L. Few species were responsible for the main bulk of phytoplankton namely Merismopedia tenuissima , Cylindrotheca closterium , Cyclotella sp., Skeletonema costatum, Scenedesmus quadricauda , Scenedesmus bijugus , and Tetradesmus dimorphus. During the study period, El-Mex Bay was affected by the presence of 13 harmful and /or toxic algal species, among which the mediophyte species Skeletonema costatum was responsible for the peak occurring during autumn (2018). The results of the recorded nutrients indicated that the Bay is highly loaded, where NH 4, NO 2, NO 3 , PO 4 , and SiO 4 attained 141.68 µM, 25.61 µM, 151.16 µM, 10.73 µM, and 232.86 µM, respectively. The macroalgal flora was represented only by the two opportunistic species ( Ulva fasciata and Ulva intestinalis ). Both species could survive in freshwater conditions exhibiting very broad salinity tolerance (6.51–38.41‰) and a high level of nitrogenous compounds. Those results revealed that El-Mex Bay suffers from pollution and deterioration of water quality.
... Phytoplankton, as the primary production, is fundamental in the ocean, and its communities, characterized by a high degree of taxonomic diversity, are closely related to marine ecosystems and biogeochemical function, especially for the health of marine environments and global carbon cycling, which have been increasingly studied [1][2][3][4][5][6]. Hence, it is of substantial interest to efficiently and effectively derive information on phytoplankton biomass and community composition at the regional and, even, global scales. ...
Article
Full-text available
Phytoplankton communities, which can be easily observed by optical sensors deployed on various types of platforms over diverse temporal and spatial scales, are crucial to marine ecosystems and biogeochemical cycles, and accurate pigment concentrations make it possible to effectively derive information from them. To date, there is no practical approach, however, to retrieving concentrations of detailed pigments from phytoplankton absorption coefficients () with acceptable accuracy and robustness in global oceans. In this study, a novel method, which is a stepwise regression method improved by early stopping (the ES-SR method) based on the derivative of hyperspectral , was proposed to retrieve pigment concentrations. This method was developed from an extensive global dataset collected from layers at different depths and contains phytoplankton pigment concentrations and . In the case of the logarithm, strong correlations were found between phytoplankton pigment concentrations and the absolute values of the second derivative ()/the fourth derivative () of . According to these correlations, the ES-SR method is effective in obtaining the characteristic wavelengths of phytoplankton pigments for pigment concentration inversion. Compared with the Gaussian decomposition method and principal component regression method, which are based on the derivatives, the ES-SR method implemented on is the optimum approach with the greatest accuracy for each phytoplankton pigment. More than half of the determination coefficient values () for all pigments, which were retrieved by performing the ES-SR method on , exceeded 0.7. The values retrieved for all pigments fit well to the one-to-one line with acceptable root mean square error (: 0.146–0.508) and median absolute percentage error (: 8.2–28.5%) values. Furthermore, the poor correlations between the deviations from the values retrieved by the ES-SR method and impact factors related to pigment composition and cell size class show that this method has advantageous robustness. Therefore, the ES-SR method has the potential to effectively monitor phytoplankton community information from hyperspectral optical data in global oceans.
... The seawater in both labs was made from Instant Ocean, but the University at Buffalo water comes from the Aquarium of Niagara, in tanks with animals and may have higher nitrogen content. For many photosynthetic organisms, the optimal growth temperature depends on the light and nutrient environment (Edwards et al., 2016), which may explain the discrepancy between studies. This also highlights the potential importance of light and nutrient levels in conducting selection experiments in the laboratory and how they may change in natural environments or inside host tissues (Grottoli et al., 2021). ...
Article
Full-text available
In many cases, understanding species’ responses to climate change requires understanding variation among individuals in response to such change. For species with strong symbiotic relationships, such as many coral reef species, genetic variation in symbiont responses to temperature may affect the response to increased ocean temperatures. To assess variation among symbiont genotypes, we examined the population dynamics and physiological responses of genotypes of Breviolum antillogorgium in response to increased temperature. We found broad temperature tolerance across genotypes, with all genotypes showing positive growth at 26, 30, and 32°C. Genotypes differed in the magnitude of the response of growth rate and carrying capacity to increasing temperature, suggesting that natural selection could favor different genotypes at different temperatures. However, the historical temperature at which genotypes were reared (26 or 30°C) was not a good predictor of contemporary temperature response. We found increased photosynthetic rates and decreased respiration rates with increasing contemporary temperature, and differences in physiology among genotypes, but found no significant differences in the response of these traits to temperature among genotypes. In species with such broad thermal tolerance, selection experiments on symbionts outside of the host may not yield results sufficient for evolutionary rescue from climate change. Our work presents the results of an experiment that examines individual variation in symbionts, isolated and grown at different temperatures for hundreds of generations, from an octocoral species. We exposed replicates of different genotypes to increasing temperature and measured their population dynamics and physiology. We found broad thermal tolerance of this symbiont species, but also found significant variation among genotypes, although this was unrelated to the historical temperature at which they were grown.
... There is a growing concern around the globe to achieve better understanding of dynamics of phytoplankton community because they are proven as significant indicators of both climate change (Edwards and Richardson 2004;Edwards et al. 2016) and the effects caused by anthropogenic influence in the marine environment e.g., the addition of nutrients (Devlin et al. 2007). Coastal water nutrients determine the structure of phytoplankton community as it influences the growth in various ways. ...
Article
Full-text available
Monthly samples were collected from 2014 to 2016 to elucidate seasonal and interannual dynamics of phytoplankton community structure, diversity, species composition and abundance in relation to physicochemical parameters. Totally, 141 phytoplankton species were identified mostly from diatoms belonging to 106 genera, dinoflagellates and blue green algae representing 29 and 6 genera, respectively. The most persistent and predominant diatoms and dinoflagellates were Asterionellopsis spp. Chaetoceros spp. Coscinodiscus spp. Nitzschia spp. Thalassionema nitzschioides, Guinardia spp. Skeletonema costatum, Ditylum brightwelli, Ceratium spp.and Dinophysis caudata.Gradual increase in phytoplankton abundance was noted in relation to nutrient enhancement from 2014 to 2016. Species diversity and composition in 2014 and 2015 found to be higher than 2016, however, the population density started elevating from 2015 to 2016 with shift in species composition. Variation in physicochemical parameters were confirmed in Principal component analysis (PCA) with distinct distribution pattern among years. Cluster and Non-metric multidimensional scaling ordination (NMDS) analyses identified four groups based on phytoplankton abundance. Diatoms and dinoflagellates that grew abundantly during 2016 formed two distinct groups in NMDS (group 1 &2) whereas; diatoms that distributed in 2014 and 2015 showed 2 separate groups (Group3&4). Canonical correspondence analysis (CCA) revealed the most influencing parameters as temperature, nitrite, silicate, and inorganic phosphate (IP) which was responsible for phytoplankton community structure changes among years especially in 2015 and 2016. The present study demonstrated that annual changes in nutrient and physical parameters can significantly alter dominations and community structure pattern of phytoplankton.
Article
In late summer 2019 and 2020 bottom waters in southern Cape Cod Bay (CCB) became depleted of dissolved oxygen (DO), with documented benthic mortality in both years. Hypoxic conditions formed in relatively shallow water where the strong seasonal thermocline intersected the sea floor, both limiting vertical mixing and concentrating biological oxygen demand (BOD) over a very thin bottom boundary layer. In both 2019 and 2020, anomalously high sub-surface phytoplankton blooms were observed, and the biomass from these blooms provided the fuel to deplete sub-pycnocline waters of DO. The increased chlorophyll fluorescence was accompanied by a corresponding decrease in sub-pycnocline nutrients, suggesting that prior to 2019 physical conditions were unfavorable for the utilization of these deep nutrients by the late-summer phytoplankton community. It is hypothesized that significant alteration of physical conditions in CCB during late summer, which is the result of regional climate change, has favored the recent increase in sub-surface phytoplankton production. These changes include rapidly warming waters and significant shifts in summer wind direction, both of which impact the intensity and vertical distribution of thermal stratification and vertical mixing within the water column. These changes in water column structure are not only more susceptible to hypoxia but also have significant implications for phytoplankton dynamics, potentially allowing for intense late-summer blooms of Karenia mikimotoi, a species new to the area. K. mikimotoi had not been detected in CCB or adjacent waters prior to 2017; however, increasing cell densities have been reported in subsequent years, consistent with a rapidly changing ecosystem.
Article
Global warming and eutrophication result in rising temperature and declining underwater light, respectively, which affect the shift of the phytoplankton community in spring. However, knowledge of how temperature and light synergistically impact phytoplankton community shifts and cyanobacterial dominance is limited. In this study, we performed a long-term data analysis and an outdoor mesocosm experiment to detect the synergistic effect of temperature and light on shift of phytoplankton community and dominance of bloom-forming cyanobacteria in Lake Taihu, China. The results showed that cyanobacterial biomass was boosted alone and jointly by increased temperature and decreased light levels (sunshine hours and light intensity), and the interaction might be more important than temperature or light levels independently. Chlorophyta biomass was driven by the joint effect of temperature and light levels. Bacillariophyta biomass was mainly affected by light levels, and decreased with declining light levels. Our results emphasize that the interactions of temperature and light have an important impact on the shift of the phytoplankton community in spring. Increasing temperature and declining underwater light boosted the flourishing of cyanobacteria, especially Microcystis, and were adverse to the development of diatoms in spring. Our findings contribute to an increased understanding of the effects of temperature and light on phytoplankton composition shifts and the development of cyanobacterial dominance in spring.
Article
Full-text available
Prochlorococcus and Synechococcus picocyanobacteria are dominant contributors to marine primary production over large areas of the ocean. Phytoplankton cells are entrained in the water column and are thus often exposed to rapid changes in irradiance within the upper mixed layer of the ocean. An upward fluctuation in irradiance can result in photosystem II photoinactivation exceeding counteracting repair rates through protein turnover, thereby leading to net photoinhibition of primary productivity, and potentially cell death. Here we show that the effective cross-section for photosystem II photoinactivation is conserved across the picocyanobacteria, but that their photosystem II repair capacity and protein-specific photosystem II light capture are negatively correlated and vary widely across the strains. The differences in repair rate correspond to the light and nutrient conditions that characterize the site of origin of the Prochlorococcus and Synechococcus isolates, and determine the upward fluctuation in irradiance they can tolerate, indicating that photoinhibition due to transient highlight exposure influences their distribution in the ocean.
Article
Full-text available
Aim: Ecological and evolutionary forces shape the functional traits of species within and across environments, generating biogeographical patterns in traits. We aimed to: (1) determine the extent to which temperature traits of phytoplankton are adapted to their local environment, and (2) detect and explain differences in patterns of adaptation between functional groups (reflecting evolutionary history) and across ecosystems (freshwater versus marine). Location: We used laboratory-measured data on phytoplankton strains isolated from marine (76° N to 75° S) and freshwater ecosystems (80° N to 78° S). Methods We studied variation in five temperature traits: optimum temperature for growth (Topt), maximum and minimum persistence temperature (Tmax, Tmin), temperature niche width and maximum growth rate, estimated in 439 strains from over 200 species. We tested whether these traits change along environmental temperature gradients (across latitude and ecosystems) and also investigated differences in trait–environment relationships related to evolutionary history (functional group identity). We used mixed models to evaluate our hypotheses while accounting for intraspecific variation. Results: We identified three patterns caused by adaptation and community assembly: (1) Topt, Tmax and Tmin decline sharply with latitude; (2) Topt, Tmax and Tmin are similar across all functional groups at the equator, where temperature variation is low; and (3) Topt and Tmax are higher in freshwater locations than marine locations at similar latitudes. Additionally, evolutionary history explained substantial variation in all traits: functional groups differ strongly in their niche widths and maximum growth rates, as well as their Topt, Tmax, and Tmin relationships with latitude. Main conclusions: Globally, phytoplankton temperature traits are well adapted to local conditions, changing across ecosystems and latitude. Functional groups differ strongly in their patterns of adaptation: traits are similar in hot tropical environments, but diverge at temperate latitudes. These differences reflect two possible evolutionary constraints: cyanobacterial inability to adapt to low temperatures and differences in nutrient requirements between groups.
Article
Temperature provides a powerful theme for exploring environmental adaptation at all levels of biological organization, from molecular kinetics to organismal fitness to global biogeography. First, the thermodynamic properties that underlie biochemical kinetics and protein stability determine the overall thermal sensitivity of rate processes. Consequently, a single quantitative framework can assess variation in thermal sensitivity of ectotherms in terms of single amino acid substitutions, quantitative genetics, and interspecific differences. Thermodynamic considerations predict that higher optimal temperatures will result in greater maximal fitness at the optimum, a pattern seen both in interspecific comparisons and in within‐population genotypic variation. Second, the temperature‐size rule (increased developmental temperature causes decreased adult body size) is a common pattern of phenotypic plasticity in ectotherms. Mechanistic models can correctly predict the rule in some taxa, but lab and field studies show that rapid evolution can weaken or even break the rule. Third, phenotypic and evolutionary models for thermal sensitivity can be combined to explore potential fitness consequences of climate warming for terrestrial ectotherms. Recent analyses suggest that climate change will have greater negative fitness consequences for tropical than for temperate ectotherms, because many tropical species have relatively narrow thermal breadths and smaller thermal safety margins.
Article
We tested if the metabolic theory of ecology (MTE) correctly predicts plankton metabolism in a temperate lake, based on a long-term (about 15 years), high-frequency dataset of body size, abundance and production, using two different techniques: least squares regression and maximum likelihood. For phytoplankton, the general fit was relatively poor (r(2) = 0.53). The assumption of the MTE on temperature dependence of metabolism was not supported, and the assumed value of 3/4 of the allometric exponent was barely within 95% confidence limits. For some of the models, the value of b was significantly higher than 3/4. When radiation was included as an additional predictor, it improved the model considerably (r(2) = 0.67). Including grazing by zooplankton reduced the model residuals during the summer period, when grazing is a dominant factor. The allometric exponent had virtually no effect for phytoplankton, due to little variability in average individual size. Zooplankton production, on the other hand, was better predicted by MTE, showing stronger effects of temperature and body size, the average of which varied by a factor of more than a hundred. However, the best-fitting value of the allometric exponent for zooplankton was 0.85, and significantly higher than the 3/4 predicted by the theory. The ratio of observed production to biomass for the entire plankton community declined linearly with the body size (in log-log) with a slope corresponding to a value of b = 0.85. We conclude that the MTE has little predictive power for the metabolism of lacustrine plankton, in particular for phytoplankton, and especially at the scale of variability of this study, and that this could be improved by incorporating radiation into the model.
Article
Climate change is altering oceanic conditions in a complex manner, and the concurrent amendment of multiple properties will modify environmental stress for primary producers. So far, global modelling studies have focused largely on how alteration of individual properties will affect marine life. Here, we use global modelling simulations in conjunction with rotated factor analysis to express model projections in terms of regional trends in concomitant changes to biologically influential multi-stressors. Factor analysis demonstrates that regionally distinct patterns of complex oceanic change are evident globally. Preliminary regional assessments using published evidence of phytoplankton responses to complex change reveal a wide range of future responses to interactive multi-stressors with <20-300% shifts in phytoplankton physiological rates, and many unexplored potential interactions. In a future ocean, provinces will encounter different permutations of change that will probably alter the dominance of key phytoplankton groups and modify regional productivity, ecosystem structure and biogeochemistry. Consideration of regionally distinct multi-stressor patterns can help guide laboratory and field studies as well as the interpretation of interactive multi-stressors in global models.
Article
Focuses on 1) the short-term effect of temperature on photosynthetic metabolism, ie the photosynthetic response when an alga is suddenly exposed to a higher or lower temperature; 2) phenotypic changes in photosynthetic metabolism that occur in response to growth at different temperatures; and 3) genetic differences in photosynthetic metabolism between algal species or ecotypes from different thermal environments. -from Author
Article
Genotypic variation in the temperature optimum for resource-saturated growth of microalgae has been used to provide envelopes of μm (maximum specific growth rate) as a function of temperature. The Q10 value for μm for batch-cultured algae with optimal growth temperatures in the range 5-40°C is 1.88; rather higher values (Q10 = 2.08-2.19) are found, albeit with lower μ values at a given temperature, for continuous cultures. For resource-limited growth, the phenotypic effect of suboptimal temperatures on growth, when light is the limiting resource, is often less marked than when growth is light saturated. When a chemical nutrient is limiting, the temperature effect on growth of a given genotype is often, but not invariably, decreased. -from Authors
Article
A 10-yr record of the thermal characteristics of four lakes at the North-Temperate Lakes Long-Term Ecological Research site was analyzed and used to validate simulations of lake physics with the dynamic reservoir simulation model. Simulations of cool, warm, and intermediate years were rerun with meteorology from four general circulation models with a doubling of CO2. In all simulations with doubled CO2 there is an earlier onset of stratification, increased summer epilimnetic temperature (1-7°C), and an increased intensity and longer duration of stratification. Maximum surface temperatures at times may exceed upper lethal limits of warm and cool water fish in some scenarios. Suitable thermal habitat for cold water, cool water, and warm water fish generally increases in all scenarios after climate change. Changes in the vertical migration of Daphnia, however, are expected to vary depending on the interaction of thermal stratification and fish habitat use. In northern Wisconsin lakes with cold water planktivores, habitat overlap between fish and zooplankton is expected to decrease, while in southern Wisconsin lakes habitat overlap is expected to increase. Although most physical responses of lakes to climate change are consistent among all climate scenarios, biological responses will likely be more variable owing to the complex nature of factors determining ecological interactions in lakes.