ArticlePDF Available

Predicting coccolithophore rain ratio responses to calcite saturation state

Authors:

Abstract

The response of the coccolithophore calcite and particulate organic carbon (POC) ratio, also known as the rain ratio, to calcite saturation state (Omega(calcite)) is increasingly being used as an input parameter for modelling ocean feedbacks to changes in atmospheric pCO(2). Previously, this relationship has only been determined from a small number of studies from a single genus of coccolithophore. However, there is an increasing abundance of literature calcite: POC - Omega(calcite) data for several coccolithophore genera. Here, Omega(calcite) and calcite: POC data were collated from literature studies of coccolithophore responses to changes in ocean carbonate chemistry. These Omega(calcite) data were recalculated using a standardized pH scale and constants. Calcite: POC responses to Omega(calcite) were then determined using quantile regression for 2 major orders of coccolithophore: Isochrysidales (Emiliania, Gephyrocapsa) and Coccolithales (Coccolithus, Calcidiscus). These 2 coccolithophore groups display qualitatively and quantitatively different responses to Omega(calcite). A general combined expression was calculated to describe the response of the calcite: POC ratio to Omega(calcite) for coccolithophores as a single functional group for use when the relative contributions of each order to the coccolithophore community are unknown.
MARINE ECOLOGY PROGRESS SERIES
Mar Ecol Prog Ser
Vol. 500: 57–65, 2014
doi: 10.3354/meps10657 Published March 17
INTRODUCTION
Production of calcite and organic carbon by marine
coccolithophores is sensitive to changes in the car-
bonate system caused by oceanic uptake of CO2from
the atmosphere (Rost et al. 2008). However, a reduc-
tion in the ratio of calcite to particulate organic car-
bon (POC) production (the rain ratio) will in turn
enhance the ability of the oceans to take up CO2due
to the different impacts that these 2 biogenic pro-
cesses have on alkalinity and dissolved inorganic
carbon (Zeebe & Wolf-Gladrow 2001). Conversely,
an increase in the coccolithophore calcite:POC ratio
will reduce the potential for ocean CO2uptake.
Recently, coccolithophore calcite:POC res pon ses to
changes in carbonate chemistry (e.g. due to ocean
acidification) have started to be used as input param-
eters for global carbon cycle models (Gehlen et al.
2007, Gangsto et al. 2011, Pinsonneault et al. 2012).
Input parameters relating calcite:POC ratios to sea-
water carbonate chemistry, specifically to calcite sat-
uration state (Ωcalcite), have until now only been based
on a small dataset of experimental observations for
the coccolithophore Emiliania (Gehlen et al. 2007,
Gangsto et al. 2011).
In the last few years, there has been some attempt
to analyse the response of Emiliania to changes in
seawater carbonate system parameters (Findlay et al.
2011). There is an ever-growing number of studies in
which calcite:POC ratios for both Emiliania and other
coccolithophores have been measured as a function
of carbonate chemistry. However, there are several
problems involved in extracting a calcite:POC versus
Ωcalcite relationship from these studies before it can be
applied to a coccolithophore component in a biogeo-
chemical model.
© Inter-Research 2014 · www.int-res.com*Corresponding author: s.r.fielding@outlook.com
Predicting coccolithophore rain ratio
responses to calcite saturation state
Samuel R. Fielding*
School of Environmental Sciences, University of Liverpool, Liverpool, UK
ABSTRACT: The response of the coccolithophore calcite and particulate organic carbon (POC)
ratio, also known as the rain ratio, to calcite saturation state (Ωcalcite) is increasingly being used as
an input parameter for modelling ocean feedbacks to changes in atmospheric pCO2. Previously,
this relationship has only been determined from a small number of studies from a single genus of
coccolithophore. However, there is an increasing abundance of literature calcite:POC − Ωcalcite
data for several coccolithophore genera. Here, Ωcalcite and calcite:POC data were collated from
literature studies of coccolithophore responses to changes in ocean carbonate chemistry. These
Ωcalcite data were recalculated using a standardized pH scale and constants. Calcite:POC re -
sponses to Ωcalcite were then determined using quantile regression for 2 major orders of coccolitho-
phore: Isochrysidales (Emiliania, Gephyrocapsa) and Coccolithales (Coccolithus, Calcidiscus).
These 2 coccolithophore groups display qualitatively and quantitatively different responses to
Ωcalcite. A general combined expression was calculated to describe the response of the calcite:POC
ratio to Ωcalcite for coccolithophores as a single functional group for use when the relative contribu-
tions of each order to the coccolithophore community are unknown.
KEY WORDS: Calcification · Phytoplankton · Carbon cycling · Modelling · Quantile regression ·
Emiliania
Resale or republication not permitted without written consent of the publisher
This authors' personal copy may not be publicly or systematically copied or distributed, or posted on the Open Web,
except with written permission of the copyright holder(s). It may be distributed to interested individuals on request.
Mar Ecol Prog Ser 500: 57– 65, 2014
First, not all studies provide Ωcalcite data. Studies
which do, use different methods to calculate Ωcalcite
from measured carbonate system parameters such as
total alkalinity and dissolved inorganic carbon, mak-
ing the accuracy of comparisons between experi-
ments uncertain. Second, experiments are carried
out under a diverse range of environmental condi-
tions. Culture studies are made under varying light,
temperature and nutrient regimes and during differ-
ent growth phases which have all been shown to
affect calcite:POC ratios (Paasche 1998, 2001, Zon-
dervan et al. 2002). Additionally, measurements de -
rived from in situ ocean microbial communities
include non-calcifying organisms which skew cal-
cite:POC values downwards. Third, intraspecific va -
riation of the calcite:POC response of Emiliania to
ocean acidification has been observed (Langer et al.
2009). Further, other species of coccolithophore have
displayed calcite:POC responses to ocean acidifica-
tion that are significantly different from those of
E. huxleyi (Langer et al. 2006).
Therefore, calcite:POC values from the literature
are a result of a wide range of factors (e.g. light, intra-
specific variation) other than the change in the car-
bonate system, making comparison between individ-
ual studies problematic.
In an attempt to synthesize these coccolithophore
ocean acidification studies, I (1) collated data from
Emiliania and from all other species of coccolitho-
phore for which calcite:POC ratios have been meas-
ured as a function of carbonate chemistry, (2) recal-
culated literature Ωcalcite values from carbonate
system measurements using a standardized set of
constants, and (3) used quantile regression (Koenker
& Bassett 1978) to determine genus- and order-level
response parameters for calcite:POC versus Ωcalcite.
METHODS
Data collection
Data were obtained from studies where coccolitho-
phore calcite and POC were measured in experi-
ments mimicking future or past changes in atmos-
pheric pCO2and subsequent changes in ocean CO2,
58
Species Strain Lat. Lon. n Source
Isochrysidales
Emiliania huxleyi CCMP371 32 −62 12 Feng et al. (2008), Lefebvre et al. (2012)
E. huxleyi PeECE III 2005 60 5 42 Müller et al. (2010)
E. huxleyi PLYB92/11 60 5 63 Zondervan et al. (2002), Bach et al. (2011),
Borchard et al. (2011)
E. huxleyi PLYM219 −47 168 48 Iglesias-Rodriguez et al. (2008), Shi et al.
(2009)
E. huxleyi Raunefjorden 2009 60 5 12 Lohbeck et al. (2012), Müller et al. (2012)
monoclonal
E. huxleyi Raunefjorden 2009 60 5 7 Lohbeck et al. (2012)
multiclonal
E. huxleyi RCC1212 −34 17 4 Langer et al. (2009)
E. huxleyi RCC1216 −42 170 12 Langer et al. (2009), Fiorini et al. (2011b),
Richier et al. (2011), Rokitta & Rost (2012)
E. huxleyi RCC1238 34 140 4 Langer et al. (2009)
E. huxleyi RCC1256 63 −20 12 Langer et al. (2009), Hoppe et al. (2011)
E. huxleyi TW1 38 2 2 Sciandra et al. (2003)
E. huxleyi CS369 −43 148 9 Gao et al. (2009)
Gephyrocapsa oceanica PC7/1 38 −9 5 Zondervan et al. (2001)
Coccolithales
Coccolithus braarudii RCC1200 −25 12 73 Langer et al. (2006), Müller et al. (2010),
Krug et al. (2011)
Calcidiscus leptoporus AC370 −34 17 2 Fiorini et al. (2011b)
Calcidiscus quadriperforatus RCC1135 −36 16 12 Langer et al. (2006), Langer (2011)
Pleurochrysis carterae Japan 40 142 2 Casareto et al. (2009)
Syracosphaerales
Syracosphaera pulchra AC418 41 14 4 Fiorini et al. (2011a)
Table 1. Coccolithophore strains used in this study with their contribution to the dataset (n =number of Ωcalcite − calcite:POC
data points) and the latitude (Lat.) and longitude (Lon.) of the location of their isolation (shown in Fig. 1); (–) indicates °S and
°W for latitude and longitude respectively. Ωcalcite: calcite saturation state; POC: particulate organic carbon
Author copy
Fielding: Coccolithophore rain ratios
pH and Ωcalcite (Table 1). Only calcite and POC data
from monospecific cultures in exponential growth
phase were included, as measurements which were
taken in stationary phase cultures (e.g. De Bodt et al.
2010) or from multispecies assemblages such as
mesocosms (e.g. Delille et al. 2005) may either under-
or overestimate POC, respectively. Similarly, data
from studies where pH was kept constant while other
carbonate system parameters were varied (e.g. Rick-
aby et al. 2010) were not included. Two studies did
not include sufficient data to accurately calculate
Ωcalcite (Nimer & Merrett 1993, Buitenhuis et al. 1999).
When tabular data were not available, graphical data
were extracted using En gauge Digitizer 4.1 (http://
digitizer. source forge. net/).
Carbonate system standardization
The literature studies analysed here used varying
methods and constants for calculating Ωcalcite from
carbonate system parameters. Therefore, to allow
inter-comparability between studies, carbonate sys-
tem parameters including Ωcalcite were recalculated
from the original data using CO2Sys 2.1 (Pierrot et al.
2006) with dissociation constants for carbonic acid of
Roy et al. (1993), the total pH scale, dissociation con-
stants for KSO4from Dickson (1990), the total boron
formulation of Lee et al. (2010) and culture tempera-
ture, salinity and phosphate concentrations from
each study. Where possible, experimentally meas-
ured total alkalinity (AT) and dissolved inorganic
carbon (DIC) were used as input parameters for cal-
culating Ωcalcite (Riebesell et al. 2010). Where 1 or
more of these parameters were not available, they
were back-calculated from pH and either pCO2or
the fugacity of CO2(fCO2) using the
appropriate pH scales and constants
(see ‘Results’ and ‘Discussion’ for
caveats of this method) before being
used to calculate Ωcalcite. Where salin-
ity data were not available but the
provenance of the sea water used
in cultures was known, salinity was
approximated from local annual
mean sea-surface salinity following
Antonov et al. (2010). Where neither
salinity nor sea water provenance da -
ta were available, a salinity of 35 was
used. All studies provided details of
culture temperature and initial phos-
phate concentrations.
Modelling and statistical analysis
Quantile regression can be used to extract informa-
tion about the relationship between 2 parameters
(e.g. calcite:POC ratio and Ωcalcite) from the upper
edge of a scatterplot for a dataset when other limiting
factors (e.g. light) contribute to a large proportion of
the variation (Cade et al. 1999, Cade & Noon 2003).
Quantile regression models with non-zero inter-
cepts (Table 2) were used to approximate the upper
edge (ideally the 99th quantile) of the dataset using R
2.15.3 with the quantreg 4.98 package. Rogers (1992)
suggested that the number of data points (n) should
be greater than [5/(1 − τ)], where τis the quantile,
with 0.99 representing the 99th quantile. Therefore,
for each dataset modelled in this study, regressions
were calculated for the largest possible quantile
without violating the upper limit constrained by n.
RESULTS AND DISCUSSION
Data overview and partitioning
Data were derived from 38 individual datasets from
21 separate studies on coccolithophores from diverse
locations (Table 1, Fig. 1). Culture experiments were
conducted on 6 genera of coccolithophore. The ge -
nus Emiliania contributed the largest number of
data sets, with 28 out of the 38. A smaller number of
datasets was also available for Gephyrocapsa (n = 1),
Coccolithus (n = 3), Calcidiscus (n = 3), Syraco -
sphaera (n = 2) and Pleurochrysis (n = 1).
The calcite:POC ratio from the 2 studies that used
Emiliania but did not include sufficient data for cal-
59
Model Equation kDeviance
Isochrysidales
Logarithmic a+ bln(x) 2 2.62
Rectangular hyperbola c+ (ax)/(b + x) 3 2.63
Power c+ a(xb) 3 2.64
Exponential c+ a(1 − ebx ) 3 2.66
Quadratic c+ (ax) + (bx2) 3 2.67
Linear a+ (bx) 2 2.78
Coccolithales
Log normal d+ (a/x)e–0.5{[ln(x/c)]/b}24 3.97
Gaussian ae–0.5[(x–c)/b]23 4.48
Lorentzian a/{1 + [(x − c)/b]2} 3 4.49
Linear a+ (bx) 2 4.69
Table 2. Pre-selection quantile regression models. krepresents the number of
independently varied parameters (a,b,c,d) in the model. Models are plotted in
Fig. 2
Author copy
Mar Ecol Prog Ser 500: 57– 65, 2014
culating Ωcalcite (Nimer & Merrett 1993, Buitenhuis et
al. 1999) fell well below the upper edge of the other
Emiliania data (Fig. 2).
Data for Pleurochrysis and Syracosphaera were
limited, and at present there is no evidence to sug-
gest that the calcite:POC ratio of either genus res -
ponds to carbonate system parameters at all. These
data were omitted from subsequent analysis. The
remaining data were partitioned at the order level,
which broadly corresponds with ecotypic differentia-
tion within the coccolithophores: Isochrysidales
(Emiliania, Gephyrocapsa) are bloom-forming (Rhodes
et al. 1995, Paasche 2001), relatively small and fast
growing, whereas the Coccolithales (Coccolithus,
Calcidiscus) are generally larger, slower growing
and do not readily form blooms.
The Isochrysidales dataset (232 Ωcalcite − cal cite:
POC data points) almost exclusively comprised Emil-
iania (~98%), while Gephyrocapsa was represented
by only 5 data from 1 study (~2%). The Coccolithales
dataset (87 Ωcalcite − calcite: POC data points) was
largely comprised of Coccolithus (~84%) with a
lesser contribution from Calcidiscus (~16%). The use
of this order-level binning approach therefore pro-
vides a useful starting point from which
to model coccolithophore calcite: POC
response to ocean acidification in the ab -
sence of further data better describing
responses at the genus level for e.g. Cal-
cidiscus or Gephyrocapsa.
Model fitting and selection
The upper edge of the Isochrysidales
data was best described by a logarithmic
curve closely followed by a rectangular
hyperbola with a <1% higher deviance
(where deviance is inversely propor-
tional to the quality of fit; Table 2). Addi-
60
dd
d
d
d
d
d
d
d
d
dd
d
d
d
d
d
d
d
Emiliania
d
Gephyrocapsa
d
Coccolithus
d
Calcidiscus
d
Pleurochrysis
d
Syracosphaera
Fig. 1. Geographical locations of the coccolithophore strains used in the
literature compilation in this study
Ωcalcite
2.5
2.0
1.5
1.0
0.5
0
2.5
2.0
1.5
1.0
0.5
0
0 2 4 6 8 10 12 14 16 0 2 4 6 8 10 12 14 16
Log normal
Gaussian
Lorentzian
Logarithmic
Quadratic
Power
Rectangular hyperbola
Exponential
Coccolithus
Calcidiscus
Pleurochrysis
Syracosphaera
Emiliania
Gephyrocapsa
Calcite:POC
a
b
c
d
Fig. 2. Ratio of coccolithophore calcite:particulate organic carbon (POC) as a function of calcite saturation state (Ωcalcite) for (a)
Isochrysidales and (b) Coccolithales and Syracosphaerales together with quantile regression models describing the upper
edge of datasets for (c) Isochrysidales (97th quantile) and (d) Coccolithales (94th quantile). See Table 2 for model deviances
Author copy
Fielding: Coccolithophore rain ratios
tionally, power, exponential and quadratic models
had deviances which were only <2% higher than
that of the logarithmic model. Despite the quadratic
model predicting decreased calcite: POC ratios at
high Ωcalcite (Fig. 2), there is currently no evidence to
suggest that either Emiliania or Gephyrocapsa dis-
play optimal relationships as a function of ocean
acidification. Further, there are only 2 data with
higher Ωcalcite than at the visual calcite:POC maxi-
mum at Ωcalcite of ~8.5, making prediction of calcite:
POC ratios uncertain in this range. Therefore, no
attempt was made to fit other peak models (e.g.
Gaussian, log normal) to the Isochrysidales dataset.
To select the most parsimonious model from the
group of models with similar deviances fit to the
Isochrysidales data, they were compared using
Akaike’s information criterion (AIC; Akaike 1974):
AIC = −2Lm+ 2k, where kis the number of inde-
pendently varied parameters in the model, and Lmis
the maximized log likelihood (−2Lmis equivalent to
the deviance of the model fit). Smaller AIC values
therefore indicate models with better relative fits to
the data. The logarithmic model not only had the
lowest deviance but also had only 2 parameters
against the 3 parameters of the other models. As a
result, AIC selected the logarithmic model as the
most parsimonious descriptor of the upper edge of
the Isochrysidales data with an AIC score of 6.62
compared to 8.63 for the rectangular hyperbolic
model. A logarithmic response has previously been
found to be a good fit for a smaller dataset of the
Emiliania calcite:POC ratio as a function of Ωcalcite
(Findlay et al. 2011).
Conversely, strains of both Coccolithus and Cal-
cidiscus have been shown to have calcite:POC
optima as a function of ocean acidification (Langer et
al. 2006). Therefore, peak models were fit to the
Coccolithales data, with a log normal curve having
the best fit for the upper edge of the dataset with a
deviance >5% lower than the next best model
(Table 2). Gaussian and Lorentzian models did not fit
the data as closely as a result of their forced symme-
try and subsequent overestimation of calcite:POC
ratios at lower values of Ωcalcite.
As the simplest mathematical model, and having
previously been used to approximate calcite:POC
ratios as a function of Ωcalcite (Gangsto et al. 2011, Pin-
sonneault et al. 2012), a linear equation was also fit to
the upper edge of each dataset. However, these mod-
els had higher deviances than non-linear models
(Table 2), reflecting the non-linearity of the relation-
ship between Ωcalcite and calcite:POC ratios apparent
in the data presented here.
Genus, geographical and environmental bias
Culture strains of Emiliania were derived from
a diverse range of ocean basins (Fig. 1). However,
there were only 2 strains of Calcidiscus which were
both isolated from a geographically restricted area in
the south Atlantic, while other genera were each rep-
resented by only a single strain. This sampling bias
towards Emiliania is common in studies of coccolitho-
phores due to the ease of isolating and maintaining
this genus in laboratory culture. The results of this
study should therefore be interpreted with this
experimental and geographical bias in mind.
Additionally, there is a lack of data derived from
strains isolated far from continental land masses
(Fig. 1). Considerable intraspecific variation in phys-
iology and morphology between near-shore and
oceanic genotypes has been shown for Emiliania
(Conte et al. 1995, Paasche 2001, Fielding et al.
2009), although this distinction has not yet been
observed for other species of coccolithophore. Fur-
ther, this distinction has not yet been observed for the
calcite:POC response to changes in the seawater car-
bonate system. However, this bias away from oceanic
strains should be taken into account when interpret-
ing the results of this study.
Previously, light intensity and day length (Paasche
1999, Zondervan et al. 2002), nutrient limitation
(Paasche 1998) and salinity (Paasche et al. 1996) have
been shown to impact coccolithophore calcite:POC
ratios, and it may be hypothesized that other environ-
mental factors such as temperature also exhibit an
influence. Both the Isochrysidales and the Coccol-
ithales datasets were derived from cultures grown
across a wide range of environmental conditions.
Culture conditions ranged from 14 to 24°C, salinities
of 29.7 to 38, day lengths from 12 to 24 h and photon
flux densities between 15 and 580 µmol photons m−2
s−1. Variation in calcite:POC ratios due to these fac-
tors can be accounted for by scatter in the data below
the upper edge of the dataset (Cade et al. 1999).
While ~96% of the data in this study were from cul-
tures grown under nutrient-replete conditions, ~4 %
were from Emiliania growing in N- and P-limited
che mo stats. However, there is currently no consen-
sus in the literature as to whether nutrient limitation
increa ses or decreases coccolithophore calcite:POC
ratios.
Data from Borchard et al. (2011) show that the
calcite:POC ratio of a Norwegian coastal strain of
Emiliania decreased negligibly from ~0.16 in already
P-limited cultures to ~0.14 in increasingly P-limited
cultures over a range of Ωcalcite. Müller et al. (2012)
61
Author copy
Mar Ecol Prog Ser 500: 57– 65, 2014
showed that in a different Norwegian coastal strain
of Emiliania, the calcite:POC ratio was reduced from
>1 in nutrient-replete cultures to between 0.75 and
0.50 under N limitation, also over a range of Ωcalcite.
These N- and P-limited calcite:POC data (included in
this study) lie well below the upper edge of the data-
set, and are therefore unlikely to have influenced
quantile regression estimates.
Similarly, Fe limitation has been shown to decrease
calcite:POC ratios from 1.29 (Fe-replete) to ~0.04 in
cultures of an Emiliania strain from the north-east
Pacific, with a reduction in calcite:POC ratios also
observed when the N source was changed from
ammonium to nitrate (Muggli & Harrison 1996).
Contrastingly, data from a North Atlantic strain of
Emiliania, not included in this study, show calcite:
POC ratios rising from ~1 in nutrient-replete batch
cultures to as high as 1.5 in both N- and P-limited
chemostats (Paasche 1998). Similarly, Riegman et al.
(2000) showed an increase in calcite:POC ratios in a
Norwegian coastal strain of Emiliania in both N- and
P-limited chemostats from 0.23 to 0.50 and from 0.4 to
0.5, respectively. Kaffes et al. (2010) also showed a
slight increase in calcite:POC ratios from 0.62 to 0.70
with a reduction in nitrate concentration. However,
carbonate system parameters for some of these stud-
ies (Muggli & Harrison 1996, Paasche 1998, Riegman
et al. 2000, Kaffes et al. 2010) are not detailed enough
to relate these data to Ωcalcite.
Nutrient limitation does not play a large part in the
calcite:POC variation observed in this study due to the
small percentage of nutrient limitation studies in-
cluded. However, nutrient limitation is likely to have
some influence on the coccolithophore calcite: POC
ratio and its response to changes in the carbonate sys-
tem in the ocean, although at present we lack culture
data with which to quantify this phenomenon. Future
experiments using both chemostat and batch cultures
over a wide range of Ωcalcite will be necessary to elicit
how macro- and micronutrient limitation controls coc-
colithophore calcite:POC ratios in relation to this, and
other, carbonate system variables.
Order-level partitioning
Although strains of both Coccolithus and Calcidis-
cus display calcite:POC optima as a function of ocean
acidification, the applicability of combining both of
these genera is open to debate. Each genus con-
tributes more to different parts of the response curve:
Coccolithus data delineate the calcite:POC response
at lower (<2) Ωcalcite, whereas Calcidiscus data repre-
sent the upper edge of the data at higher (>5) Ωcalcite.
However, there is no evidence to suggest that Coc-
colithus follows the response curve of Calcidiscus
and vice versa. Therefore, additional calcite:POC
data from Coccolithus at higher Ωcalcite and from Cal-
cidiscus at lower Ωcalcite are required to support the
use of the combined relationship presented in this
study. Nonetheless, the combination of these 2 data -
sets may still provide a useful simplification of the
calcite:POC response of the Coccolithales for incor-
poration into carbon cycle models.
Conversely, the inclusion of the small number of
Gephyrocapsa data in the Isochrysidales dataset has
only minimal influence on quantile regression model
selection or parameter estimates. Therefore, the re -
gression results for the Isochrysidales dataset essen-
tially describe Emiliania. However, the Gephyro-
capsa calcite:POC ratio has been shown to react to
ocean acidification in a similar way to Emiliania
(Zondervan et al. 2001), while there is currently no
evidence to suggest that Gephyrocapsa might exhibit
a calcite:POC optimum as a function of ocean acidifi-
cation. Therefore, for the purposes of using these
results as input parameters for biogeochemical mod-
elling, it is suggested that they are used to encom-
pass both Emiliania and Gephyrocapsa. Nonetheless,
further data for Gephyrocapsa are required to better
support this conclusion.
Problems with carbonate system
standardization
Due to the use of varying carbonic acid dissociation
constants and pH scales in the literature studies col-
lated here, a standardization approach was used to
attempt to minimize errors introduced while calculat-
ing Ωcalcite.
It has been shown that calculated values of pCO2,
Ωcalcite and other carbonate system parameters from
measured ATand DIC can differ from parameters
calculated from pH and either ATor DIC by up to
30% (Hoppe et al. 2012). Therefore, in this study, as
the most common carbonate system parameters
available were ATand DIC, where other carbonate
system data were given they were converted back to
these 2 initial parameters. This calculation step is
likely to have introduced a certain amount of error
inherent in the constants and in the propagation of
the error from the initial measurements. However,
the error introduced by conversion from one set of
constants to another is small compared with errors
from comparing values derived using different sets of
62
Author copy
Fielding: Coccolithophore rain ratios
constants and pH scales (Zeebe & Wolf-Gladrow
2001). The benefits of the method used here of recal-
culating the carbonate system using a standardized
set of constants and from a standardized starting
point (i.e. ATand DIC) are therefore likely to out-
weigh any error derived from back-calculation.
Additional error in calculating Ωcalcite may derive
from estimates of culture salinity where measured
salinity values were not given. For the purposes of cal-
culating calcite saturation state (using carbonate and
calcium ion concentrations), seawater calcium con-
centration is assumed to be proportional to salinity.
Therefore, estimated salinity values could potentially
result in incorrect Ωcalcite. However, salinity only had
to be approximated for < 3% of the dataset used in this
study, and where it was estimated, it would have
caused an error in Ωcalcite of ~0.05 per unit salinity er-
ror. Further, all data where salinity was estimated are
for Emiliania and fall well below the upper edge of the
data. Therefore, minimal impact on the overall trends
in calcite:POC ratios observed here is expected.
Calcite sub-saturation
Theoretically, abiogenic calcium carbonate should
dissolve completely below an Ωcalcite of 1, as seawater
would be sub-saturated with respect to calcite. In-
deed, models incorporating coccolithophore calcite:
POC ratios force values to 0 below an Ωcalcite of 1
(Gehlen et al. 2007, Gangsto et al. 2011, Pinsonneault
et al. 2012). However, the present study highlights
that coccolithophore genera belonging to both the
Iso chrysidales and the Coccolithales can produce cal-
cite in seawater sub-saturated with respect to calcite
(Fig. 2). Below an Ωcalcite of 1, the calcite:POC ratio for
both clades is reduced to less than 50% of the maxi-
mum calcite: POC ratio. Persistent
calcite production under these condi-
tions is likely due to coccolithophore
calcite having an organic coating
which partially protects it from disso-
lution (Henriksen et al. 2004, Godoi
et al. 2009, Hassenkam et al. 2011).
Parameter estimates for
modelling
The present study uses quantile re -
gression to determine the res ponse
of the calcite:POC ratio to Ωcalcite by
modelling the upper edge of the scat-
terplot which represents the response of the calcite:
POC ratio to Ωcalcite when environmental variables
other than the investigated response parameter (i.e.
Ωcalcite) are not limiting (Cade et al. 1999). Once the
shape (e.g. linear, logarithmic, exponential) of the re-
sponse has been determined, an average regression
model can be calculated through the data using the
50th quantile. These 50th quantile estimates may be
useful for approximating an averaged res ponse of the
calcite:POC ratio to Ωcalcite under a range of unknown
environmental conditions which can then be used for
biogeochemical model parameterization. A logarithmic
50th quantile re gres sion was made through the Iso -
chrysidales data and a log normal 50th quantile re-
gression was made through the Coccolithales data
(Fig. 3). For modelling coccolithophore calcite:POC
response to Ωcalcite where the relative contributions of
63
Ωcalcite
3
2
1
0
0 2 4 6 8 10 12 14 16
All data
Isochrysidales, 97th
Isochrysidales, 50th
Coccolithales, 94th
Coccolithales, 50th
50th mean
Isochrysidales & Coccolithales
Calcite:POC
Fig. 3. Quantile regression models describing the upper edges
and median (50th quantile) estimates for the ratios of calcite:
particulate organic carbon (POC) of both Cocco lithales and
Isochrysidales as a function of calcite saturation state, Ωcalcite.
The model describing the mean of the two 50th quantile esti-
mates is also shown. See Table 3 for model parameters
Model Quantile Equation
Isochrysidales
Logarithmic − 97th 0.5650 + 0.2951ln(Ωcalcite)
upper edge
Logarithmic − 50th 0.4210 + 0.2094 ln(Ωcalcite)
median
Coccolithales
Log normal − 94th –1.1700 + (66.712/Ωcalcite)e–0.5{[ln(Ωcalcite/93.990)]/1.7413}2
upper edge
Log normal − 50th –0.7304 + (281.93/Ωcalcite)e–0.5{[ln(Ωcalcite/1919.1)]/2.3209}2
median
All data
Log normal − – –0.5904 + (1159.1/Ωcalcite)e 0.5{[ln(Ωcalcite/30949)]/2.7777}2
mean of medians
Table 3. Post-selection quantile regression models. Models are plotted in Fig. 3.
Ωcalcite is the calcite saturation state
Author copy
Mar Ecol Prog Ser 500: 57– 65, 2014
64
different coccolithophore genera, orders or ecotypes
is unknown, the mean of the two 50th quantile regres-
sions may be taken (Fig. 3). This mean curve can be
described by a log normal model (Table 3).
Aside from the non-zero calcite:POC ratio below
Ωcalcite of 1, the mean 50th quantile response pre-
sented here (Fig. 3) is both qualitatively and quanti-
tatively similar to those used previously (Gehlen et
al. 2007, Gangsto et al. 2011) and therefore partially
supports the results of these modelling studies. How-
ever, the inclusion of non-zero calcite:POC ratios for
Ωcalcite <1 in biogeochemical models may more
closely represent coccolithophore calcite production
in the ocean.
CONCLUSION
The relationships presented here represent the first
attempt to objectively quantify the relationship be -
tween coccolithophore calcite:POC ratios and Ωcalcite
for both Isochrysidales (Emiliania, Gephyrocapsa)
and Coccolithales (Coccolithus, Calcidiscus). The
50th quantile regression estimates for both Isochrysi-
dales and Coccolithales are recommended for use in
biogeochemical models incorporating either of these
2 distinct groups. Alternatively, the mean 50th quan-
tile response is recommended for use where cocco-
lithophores are modelled as a single functional
group. However, the inclusion of further data from
under-sampled genera, from nutrient-limited experi-
ments and from open-ocean locations will allow for a
more accurate reappraisal of these relationships.
Acknowledgements. Thanks to G. Langer and K. Lohbeck
for providing data.
LITERATURE CITED
Akaike H (1974) A new look at statistical-model identifica-
tion. IEEE Trans Automat Contr 19: 716−723
Antonov JI, Seidov D, Boyer TP, Locarnini RA and others
(2010) World ocean atlas 2009, Vol 2: Salinity. In: Levitus S
(ed) NOAA Atlas NESDIS 69. US Governmet Printing
Office, Washington, DC
Bach LT, Riebesell U, Schulz KG (2011) Distinguishing
between the effects of ocean acidification and ocean
carbonation in the coccolithophore Emiliania huxleyi.
Limnol Oceanogr 56: 2040−2050
Borchard C, Borges AV, Handel N, Engel A (2011) Biogeo-
chemical response of Emiliania huxleyi (PML B92/11) to
elevated CO2and temperature under phosphorous limita-
tion: a chemostat study. J Exp Mar Biol Ecol 410: 61−71
Buitenhuis ET, de Baar HJW, Veldhuis MJW (1999) Photo -
synthesis and calcification by Emiliania huxleyi (Prymne-
siophyceae) as a function of inorganic carbon species.
J Phycol 35: 949−959
Cade BS, Noon BR (2003) A gentle introduction to quantile
regression for ecologists. Front Ecol Environ 1: 412−420
Cade BS, Terrell JW, Schroeder RL (1999) Estimating effects
of limiting factors with regression quantiles. Ecology 80:
311−323
Casareto BE, Niraula MP, Fujimura H, Suzuki Y (2009) Effects
of carbon dioxide on the coccolithophorid Pleurochrysis
carterae in incubation experiments. Aquat Biol 7: 59−70
Conte MH, Thompson A, Eglinton G, Green JC (1995) Lipid
biomarker diversity in the coccolithophorid Emiliania hux-
leyi (Prymnesiophyceae) and the related species Gephyro-
capsa oceanica. J Phycol 31: 272−282
De Bodt C, Van Oostende N, Harlay J, Sabbe K, Chou L (2010)
Individual and interacting effects of pCO2and tempera-
ture on Emiliania huxleyi calcification: study of the calcite
production, the coccolith morphology and the coccosphere
size. Biogeosciences 7: 1401−1412
Delille B, Harlay J, Zondervan I, Jacquet S and others (2005)
Response of primary production and calcification to
changes of pCO2during experimental blooms of the
coccolithophorid Emiliania huxleyi. Global Biogeochem
Cycles 19: GB2023, doi:10.1029/2004GB002318
Dickson AG (1990) Standard potential of the reaction: AgCl(s)
+ 12H2(g) = Ag(s) + HCl(aq), and the standard acidity con-
stant of the ion HSO4in synthetic sea water from 273.15 to
318.15 K. J Chem Thermodyn 22: 113−127
Feng Y, Warner ME, Zhang Y, Sun J, Fu FX, Rose JM,
Hutchins DA (2008) Interactive effects of increased pCO2,
temperature and irradiance on the marine coccolithophore
Emiliania huxleyi (Prymnesiophyceae). Eur J Phycol 43:
87−98
Fielding SR, Herrle JO, Bollmann J, Worden RH, Montagnes
DJS (2009) Assessing the applicability of Emiliania huxleyi
coccolith morphology as a sea-surface salinity proxy. Lim-
nol Oceanogr 54: 1475−1480
Findlay HS, Calosi P, Crawfurd KJ (2011) Determinants of the
PIC: POC response in the coccolithophore Emiliania hux-
leyi under future ocean acidification scenarios. Limnol
Oceanogr 56: 1168−1178
Fiorini S, Middelburg JJ, Gattuso JP (2011a) Effects of
elevated CO2partial pressure and temperature on the coc-
colithophore Syracosphaera pulchra. Aquat Microb Ecol
64: 221−232
Fiorini S, Middelburg JJ, Gattuso JP (2011b) Testing the
effects of elevated pCO2on coccolithophores (Prymnesio-
phyceae): comparison between haploid and diploid life
stages. J Phycol 47: 1281−1291
Gangsto R, Joos F, Gehlen M (2011) Sensitivity of pelagic cal-
cification to ocean acidification. Biogeosciences 8: 433−458
Gao KS, Ruan ZX, Villafane VE, Gattuso JP, Helbling EW
(2009) Ocean acidification exacerbates the effect of UV
radiation on the calcifying phytoplankter Emiliania hux-
leyi. Limnol Oceanogr 54: 1855−1862
Gehlen M, Gangstø R, Schneider B, Bopp L, Aumont O, Ethe
C (2007) The fate of pelagic CaCO3production in a high
CO2ocean: a model study. Biogeosciences 4: 505−519
Godoi RHM, Aerts K, Harlay J, Kaegi R, Ro CU, Chou L, Van
Grieken R (2009) Organic surface coating on coccolitho-
phores — Emiliania huxleyi: its determination and implica-
tion in the marine carbon cycle. Microchem J 91: 266−271
Hassenkam T, Johnsson A, Bechgaard K, Stipp SLS (2011)
Tracking single coccolith dissolution with picogram reso-
lution and implications for CO2sequestration and ocean
acidification. Proc Natl Acad Sci USA 108: 8571−8576
Henriksen K, Young JR, Bown PR, Stipp SLS (2004) Coccolith
biomineralisation studied with atomic force micro scopy.
Author copy
Fielding: Coccolithophore rain ratios
Palaeontology 47: 725−743
Hoppe CJM, Langer G, Rost B (2011) Emiliania huxleyi shows
identical responses to elevated pCO2in TA and DIC
manipulations. J Exp Mar Biol Ecol 406: 54−62
Hoppe CJM, Langer G, Rokitta SD, Wolf-Gladrow DA, Rost B
(2012) Implications of observed inconsistencies in carbon-
ate chemistry measurements for ocean acidification stud-
ies. Biogeosciences 9: 2401−2405
Iglesias-Rodriguez MD, Halloran PR, Rickaby REM, Hall IR
and others (2008) Phytoplankton calcification in a high-
CO2world. Science 320: 336−340
Kaffes A, Thoms S, Trimborn S, Rost B and others (2010) Car-
bon and nitrogen fluxes in the marine coccolithophore
Emiliania huxleyi grown under different nitrate concen-
trations. J Exp Mar Biol Ecol 393: 1−8
Koenker R, Bassett G (1978) Regression quantiles. Economet-
rica 46: 33−50
Krug SA, Schulz KG, Riebesell U (2011) Effects of changes in
carbonate chemistry speciation on Coccolithus braarudii:
a discussion of coccolithophorid sensitivities. Biogeo-
sciences 8: 771−777
Langer G (2011) CO2mediation of adverse effects of sea water
acidification in Calcidiscus leptoporus. Geo chem Geophys
Geosyst 12: Q05001, doi:10.1029/2010 GC 003393
Langer G, Geisen M, Baumann KH, Kläs J, Riebesell U, Thoms
S, Young JR (2006) Species-specific responses of calcifying
algae to changing seawater carbonate chemistry. Geochem
Geophys Geosyst 7: Q09006, doi:10.1029/ 2005 GC001227
Langer G, Nehrke G, Probert I, Ly J, Ziveri P (2009) Strain-
specific responses of Emiliania huxleyi to changing sea-
water carbonate chemistry. Biogeosciences 6: 2637−2646
Lee K, Kim TW, Byrne RH, Millero FJ, Feely RA, Liu YM
(2010) The universal ratio of boron to chlorinity for the
North Pacific and North Atlantic oceans. Geochim Cos-
mochim Acta 74: 1801−1811
Lefebvre SC, Benner I, Stillman JH, Parker AE and others
(2012) Nitrogen source and pCO2synergistically affect
carbon allocation, growth and morphology of the cocco-
lithophore Emiliania huxleyi: potential implications of
ocean acidification for the carbon cycle. Glob Change Biol
18: 493−503
Lohbeck KT, Riebesell U, Reusch TBH (2012) Adaptive evolu-
tion of a key phytoplankton species to ocean acidification.
Nat Geosci 5: 346−351
Muggli DL, Harrison PJ (1996) Effects of nitrogen source on
the physiology and metal nutrition of Emiliania huxleyi
grown under different iron and light conditions. Mar Ecol
Prog Ser 130: 255−267
Müller MN, Schulz KG, Riebesell U (2010) Effects of long-
term high CO2exposure on two species of coccolitho-
phores. Biogeosciences 7: 1109−1116
Müller MN, Beaufort L, Bernard O, Pedrotti ML, Talec A,
Sciandra A (2012) Influence of CO2and nitrogen limitation
on the coccolith volume of Emiliania huxleyi (Hapto-
phyta). Biogeosciences 9: 4155−4167
Nimer NA, Merrett MJ (1993) Calcification rate in Emiliania
huxleyi Lohmann in response to light, nitrate and avail-
ability of inorganic carbon. New Phytol 123: 673−677
Paasche E (1998) Roles of nitrogen and phosphorus in cocco-
lith formation in Emiliania huxleyi (Prymnesiophyceae).
Eur J Phycol 33: 33−42
Paasche E (1999) Reduced coccolith calcite production under
light-limited growth: a comparative study of three clones
of Emiliania huxleyi (Prymnesiophyceae). Phycologia 38:
508−516
Paasche E (2001) A review of the coccolithophorid Emili -
ania huxleyi (Prymnesiophyceae), with particular refer-
ence to growth, coccolith formation, and calcification-pho-
tosynthesis interactions. Phycologia 40: 503−529
Paasche E, Brubak S, Skattebøl S, Young JR, Green JC (1996)
Growth and calcification in the coccolithophorid Emiliania
huxleyi (Haptophyceae) at low salinities. Phycologia 35:
394−403
Pierrot D, Lewis E, Wallace DWR (2006) MS Excel program
developed for CO2system calculations. ORNL/CDIAC-
105a. Carbon Dioxide Information Analysis Center, Oak
Ridge National Laboratory, US Department of Energy,
Oak Ridge, TN
Pinsonneault AJ, Matthews HD, Galbraith ED, Schmittner A
(2012) Calcium carbonate production response to future
ocean warming and acidification. Biogeosciences 9:
2351−2364
Rhodes LL, Peake BM, Mackenzie AL, Marwick S (1995) Coc-
colithophores Gephyrocapsa oceanica and Emiliania hux-
leyi (Prymnesiophyceae = Haptophyceae) in New Zealand
coastal waters characteristics of blooms and growth in
laboratory culture. NZ J Mar Freshw Res 29: 345−357
Richier S, Fiorini S, Kerros ME, von Dassow P, Gattuso JP
(2011) Response of the calcifying coccolithophore Emilia-
nia huxleyi to low pH/high pCO2: from physiology to
molecular level. Mar Biol 158: 551−560
Rickaby REM, Henderiks J, Young JN (2010) Perturbing
phytoplankton: response and isotopic fractionation with
changing carbonate chemistry in two coccolithophore spe-
cies. Clim Past 6: 771−785
Riebesell U, Fabry VJ, Hansson L, Gattuso JP (eds) (2010)
Guide to best practices for ocean acidification research
and data reporting. Publications Office of the European
Union, Luxembourg
Riegman R, Stolte W, Noordeloos AAM, Slezak D (2000)
Nutrient uptake, and alkaline phosphate (EC3: 1: 3: 1) activ-
ity of Emiliania huxleyi (Prymnesiophyceae) during
growth under N and P limitation in continuous cultures. J
Phycol 36: 87−96
Rogers WH (1992) Quantile regression standard errors. Stata
Tech Bull 9: 16−19
Rokitta SD, Rost B (2012) Effects of CO2and their modulation
by light in the life-cycle stages of the coccolithophore
Emiliania huxleyi. Limnol Oceanogr 57: 607−618
Rost B, Zondervan I, Wolf-Gladrow D (2008) Sensitivity of
phytoplankton to future changes in ocean carbonate
chemistry: current knowledge, contradictions and re -
search directions. Mar Ecol Prog Ser 373: 227−237
Roy RN, Roy LN, Vogel KM, Porter-Moore C and others (1993)
The dissociation constants of carbonic acid in seawater at
salinities 5 to 45 and temperatures 0 to 45°C. Mar Chem
44: 249−267
Sciandra A, Harlay J, Lefèvre D, Lemée R, Rimmelin P, Denis
M, Gattuso JP (2003) Response of coccolithophorid Emilia-
nia huxleyi to elevated partial pressure of CO2under
nitrogen limitation. Mar Ecol Prog Ser 261: 111−122
Shi D, Xu Y, Morel FMM (2009) Effects of the pH/pCO2con-
trol method on medium chemistry and phytoplankton
growth. Biogeosciences 6: 1199−1207
Zeebe RE, Wolf-Gladrow DA (2001) CO2in seawater: equilib-
rium, kinetics, isotopes. Elsevier, Amsterdam
Zondervan I, Zeebe RE, Rost B, Riebesell U (2001) Decrea sing
marine biogenic calcification: a negative feedback on ris-
ing atmospheric pCO2. Global Biogeochem Cycles 15:
507−516
Zondervan I, Rost B, Riebesell U (2002) Effect of CO2concen-
tration on the PIC/POC ratio in the coccolithophore Emil-
iania huxleyi grown under light-limiting conditions and
different daylengths. J Exp Mar Biol Ecol 272: 55−70
65
Editorial responsibility: Ronald Kiene,
Mobile, Alabama, USA
Submitted: May 22, 2013; Accepted: November 19, 2013
Proofs received from author(s): Februar y 17, 2014
Author copy
... Uncertainties in our model results here reflect uncertainties in modeled parameterization of CO 2 -calcification feedback, which actually reflects uncertainties in our understanding of the calcification response to changing ocean chemistry. The reported response of the CaCO 3 production rate to ocean acidification varies dramatically between experiments using different species of calcifying groups or manipulation methods 20,26 . Therefore, modeling simulations based on different results of experimental studies would result in different estimates of the effect of CO 2 -calcification feedback. ...
Article
Full-text available
Ocean uptake of anthropogenic CO2 reduces pH and saturation state of calcium carbonate materials of seawater, which could reduce the calcification rate of some marine organisms, triggering a negative feedback on the growth of atmospheric CO2. We quantify the effect of this CO2-calcification feedback by conducting a series of Earth system model simulations that incorporate different parameterization schemes describing the dependence of calcification rate on saturation state of CaCO3. In a scenario with SRES A2 CO2 emission until 2100 and zero emission afterwards, by year 3500, in the simulation without CO2-calcification feedback, model projects an accumulated ocean CO2 uptake of 1462 PgC, atmospheric CO2 of 612 ppm, and surface pH of 7.9. Inclusion of CO2-calcification feedback increases ocean CO2 uptake by 9 to 285 PgC, reduces atmospheric CO2 by 4 to 70 ppm, and mitigates the reduction in surface pH by 0.003 to 0.06, depending on the form of parameterization scheme used. It is also found that the effect of CO2-calcification feedback on ocean carbon uptake is comparable and could be much larger than the effect from CO2-induced warming. Our results highlight the potentially important role CO2-calcification feedback plays in ocean carbon cycle and projections of future atmospheric CO2 concentrations.
Article
Full-text available
The response of Emiliania huxleyi (Lohmann) W. W. Hay et H. Mohler, Calcidiscus leptoporus (G. Murray et V. H. Blackman) J. Schiller, and Syracosphaera pulchra Lohmann to elevated partial pressure of carbon dioxide (pCO 2) was investigated in batch cultures. We reported on the response of both haploid and diploid life stages of these three species. Growth rate, cell size, particulate inorganic carbon (PIC), and particulate organic carbon (POC) of both life stages were measured at two different pCO 2 (400 and 760 parts per million [ppm]), and their organic and inorganic carbon production were calculated. The two life stages within the same species generally exhibited a similar response to elevated pCO 2 , the response of the haploid stage being often more pronounced than that of the diploid stage. The growth rate was consistently higher at elevated pCO 2 , but the response of other processes varied among species. Calcification rate of C. leptoporus and of S. pulchra did not change at elevated pCO 2 , whereas it increased in E. huxleyi. POC production and cell size of both life stages of S. pulchra and of the haploid stage of E. huxleyi markedly decreased at elevated pCO 2. It remained unaltered in the diploid stage of E. huxleyi and C. leptoporus and increased in the haploid stage of the latter. The PIC:-POC ratio increased in E. huxleyi and was constant in C. leptoporus and S. pulchra. Elevated pCO 2 has a significant effect on these three coccolithophore species , the haploid stage being more sensitive. This effect must be taken into account when predicting the fate of coccolithophores in the future ocean.
Article
Full-text available
The growing field of ocean acidification research is concerned with the investigation of organism responses to increasing p CO<sub>2</sub> values. One important approach in this context is culture work using seawater with adjusted CO<sub>2</sub> levels. As aqueous p CO<sub>2</sub> is difficult to measure directly in small-scale experiments, it is generally calculated from two other measured parameters of the carbonate system (often A <sub> T </sub>, C <sub> T </sub> or pH). Unfortunately, the overall uncertainties of measured and subsequently calculated values are often unknown. Especially under high p CO<sub>2</sub>, this can become a severe problem with respect to the interpretation of physiological and ecological data. In the few datasets from ocean acidification research where all three of these parameters were measured, p CO<sub>2</sub> values calculated from A <sub> T </sub> and C <sub> T </sub> are typically about 30% lower (i.e. ~300 μatm at a target p CO<sub>2</sub> of 1000 μatm) than those calculated from A <sub> T </sub> and pH or C <sub> T </sub> and pH. This study presents and discusses these discrepancies as well as likely consequences for the ocean acidification community. Until this problem is solved, one has to consider that calculated parameters of the carbonate system (e.g. p CO<sub>2</sub>, calcite saturation state) may not be comparable between studies, and that this may have important implications for the interpretation of CO<sub>2</sub> perturbation experiments.
Article
Full-text available
Emiliania huxleyi, a small oceanic coccolithophore, was isolated from the NE subarctic Pacific and maintained under oceanic conditions. Coccolith- forming cultures were grown with either NO3- or NH4+ as the primary nitrogen source. Fe-stress was induced experimentally, and physiological parameters including metal quotas (Fe, Mn, Zn, Cu) were measured for both NO3-- and NH4+-grown cells to determine whether it was advantageous for the cells to grow on NH4+ rather than NO3- under Fe-stressed conditions. The parameters used to observe the cell's physiological status were specific growth rate (μ), cell volume (CV), carbon (C), chlorophyll a (chl a), and nitrogen (N) per cell volume. Under Fe-replete conditions (100 nM Fe), no physiological parameters were significantly different (p < 0.05) between NO3-- and NH4+-grown cells. However, under Fe-stressed conditions, CV (NH4+ > NO3-), chl a CV-1 (NH4+ < NO3-), Mn CV-1 (NH4+ < NO3- ), and Mn:C (NH4+ < NO3-) were all significantly different (p < 0.05) for NO3-- and NH4+-grown cells. Under Fe-stressed conditions, NO3--grown cells of E. huxleyi maintained the same or greater levels of chl a, N, Fe, Mn, and Cu as NH4+-grown cells, largely due to the drastic decrease in CV of NO3--grown cells under Fe-stress. Although NO3--grown cells substantially decreased their CVs under Fe-stressed conditions (whereas NH4+-grown cells did not), both NO3-- and NH4+-grown cells reduced their CVs equally under limiting-irradiance, Fe-replete conditions. A reduction in CV by the NO3--grown cells is particularly advantageous for cells living in a low Fe environment, since it reduces the cellular requirements for photosynthate, N, and Fe. The fact that these already small cells become smaller, along with their unique ability to maintain chlorophyll synthesis at Fe levels limiting to cell division, may help explain why E. huxleyi is a member of the numerically dominant size class in the NE subarctic Pacific.
Article
Full-text available
Coccolithophores, a key phytoplankton group, are one of the most studied organisms regarding their physiological response to ocean acidification/carbonation. The biogenic production of calcareous coccoliths has made coccolithophores a promising group for paleoceanographic research aiming to reconstruct past environmental conditions. Recently, geochemical and morphological analyses of fossil coccoliths have gained increased interest in regard to changes in seawater carbonate chemistry. The cosmopolitan coccolithophore Emiliania huxleyi (Lohm.) Hay and Mohler was cultured over a range of pCO2 levels in controlled laboratory experiments under nutrient replete and nitrogen limited conditions. Measurements of photosynthesis and calcification revealed, as previously published, an increase in particulate organic carbon production and a moderate decrease in calcification from ambient to elevated pCO2. The enhancement in particulate organic carbon production was accompanied by an increase in cell diameter. Changes in coccolith volume were best correlated with the coccosphere/cell diameter and no significant correlation was found between the coccolith volume and the particulate inorganic carbon production. The conducted experiments revealed that the coccolith volume of E. huxleyi is variable with aquatic CO2 concentration but its sensitivity is rather small in comparison with its sensitivity to nitrogen limitation. Comparing coccolith morphological and geometrical parameters like volume, mass and size to physiological parameters under controlled laboratory conditions is an important step to understand variations in fossil coccolith geometry.
Article
Full-text available
Anthropogenic carbon dioxide (CO2) emissions are acidifying the ocean, affecting calcification rates in pelagic organisms and thereby modifying the oceanic alkalinity cycle. However, the responses of pelagic calcifying organisms to acidification vary widely between species, contributing uncertainty to predictions of atmospheric CO2 and the resulting climate change. Meanwhile, ocean warming caused by rising CO2 is expected to drive increased growth rates of all pelagic organisms, including calcifiers. It thus remains unclear whether anthropogenic CO2 will ultimately increase or decrease the globally-integrated pelagic calcification rate. Here, we assess the importance of this uncertainty by introducing a variable dependence of calcium carbonate (CaCO3) production on calcite saturation state (ΩCaCO3) in the University of Victoria Earth System Climate Model, an intermediate complexity coupled carbon-climate model. In a series of model simulations, we examine the impact of this parameterization on global ocean carbon cycling under two CO2 emissions scenarios, both integrated to the year 3500. The simulations show a significant sensitivity of the vertical and surface horizontal alkalinity gradients to the parameterization, as well as the removal of alkalinity from the ocean through CaCO3 burial. These sensitivities result in an additional oceanic uptake of carbon when calcification depends on ΩCaCO3 (of up to 13 % of total carbon emissions), compared to the case where calcification is insensitive to acidification. In turn, this response causes a reduction of global surface air temperature of up to 0.4 °C in year 3500, a 13 % reduction in the amplitude of warming. Narrowing these uncertainties will require better understanding of both temperature and acidification effects on pelagic calcifiers. Preliminary examination suggests that alkalinity observations can be used to constrain the range of uncertainties and may exclude large sensitivities of CaCO3 production on ΩCaCO3.
Article
Anthropogenic carbon dioxide (CO<sub>2</sub>) emissions are acidifying the ocean, affecting calcification rates in pelagic organisms, and thereby modifying the oceanic carbon and alkalinity cycles. However, the responses of pelagic calcifying organisms to acidification vary widely between species, contributing uncertainty to predictions of atmospheric CO<sub>2</sub> and the resulting climate change. At the same time, ocean warming caused by rising CO<sub>2</sub> is expected to drive increased growth rates of all pelagic organisms, including calcifiers. It thus remains unclear whether anthropogenic CO<sub>2</sub> emissions will ultimately increase or decrease pelagic calcification rates. Here, we assess the importance of this uncertainty by introducing a dependence of calcium carbonate (CaCO<sub>3</sub>) production on calcite saturation state (Ω<sub>CaCO<sub>3</sub></sub>) in an intermediate complexity coupled carbon-climate model. In a series of model simulations, we examine the impact of several variants of this dependence on global ocean carbon cycling between 1800 and 3500 under two different CO<sub>2</sub> emissions scenarios. Introducing a calcification-saturation state dependence has a significant effect on the vertical and surface horizontal alkalinity gradients, as well as on the removal of alkalinity from the ocean through CaCO<sub>3</sub> burial. These changes result in an additional oceanic uptake of carbon when calcification depends on Ω<sub>CaCO<sub>3</sub></sub> (of up to 270 Pg C), compared to the case where calcification does not depend on acidification. In turn, this response causes a reduction of global surface air temperature of up to 0.4 °C in year 3500. Different versions of the model produced varying results, and narrowing this range of uncertainty will require better understanding of both temperature and acidification effects on pelagic calcifiers. Nevertheless, our results suggest that alkalinity observations can be used to constrain model results, and may not be consistent with the model versions that simulated stronger responses of CaCO<sub>3</sub> production to changing saturation state.
Article
In a recent Concepts paper in Ecology, Thomson et al. emphasized that assumptions of conventional correlation and regression analyses fundamentally conflict with the ecological concept of limiting factors, and they called for new statistical procedures to address this problem. The analytical issue is that unmeasured factors may be the active limiting constraint and may induce a pattern of unequal variation in the biological response variable through an interaction with the measured factors. Consequently, changes near the maxima, rather than at the center of response distributions, are better estimates of the effects expected when the observed factor is the active limiting constraint. Regression quantiles provide estimates for linear models fit to any part of a response distribution, including near the upper bounds, and require minimal assumptions about the form of the error distribution. Regression quantiles extend the concept of one-sample quantiles to the linear model by solving an optimization problem of minimizing an asymmetric function of absolute errors. Rank-score tests for regression quantiles provide tests of hypotheses and confidence intervals for parameters in linear models with heteroscedastic errors, conditions likely to occur in models of limiting ecological relations. We used selected regression quantiles (e.g., 5th, 10th, ..., 95th) and confidence intervals to test hypotheses that parameters equal zero for estimated changes in average annual acorn biomass due to forest canopy cover of oak (Quercus spp.) and oak species diversity. Regression quantiles also were used to estimate changes in glacier lily (Erythronium grandiflorum) seedling numbers as a function of lily flower numbers, rockiness, and pocket gopher (Thomomys talpoides fossor) activity, data that motivated the query by Thomson et al. for new statistical procedures. Both example applications showed that effects of limiting factors estimated by changes in some upper regression quantile (e.g., 90-95th) were greater than if effects were estimated by changes in the means from standard linear model procedures. Estimating a range of regression quantiles (e.g., 5-95th) provides a comprehensive description of biological response patterns for exploratory and inferential analyses in observational studies of limiting factors, especially when sampling large spatial and temporal scales.
Article
In laboratory experiments with the coccolithophore species Emiliania huxleyi and Gephyrocapsa oceanica, the ratio of particulate inorganic carbon (PIC) to particulate organic carbon (POC) production decreased with increasing CO2 concentration ([CO2]). This was due to both reduced PIC and enhanced POC production at elevated [CO2]. Carbon dioxide concentrations covered a range from a preindustrial level to a value predicted for 2100 according to a “business as usual” anthropogenic CO2 emission scenario. The laboratory results were used to employ a model in which the immediate effect of a decrease in global marine calcification relative to POC production on the potential capacity for oceanic CO2 uptake was simulated. Assuming that overall marine biogenic calcification shows a similar response as obtained for E. huxleyi or G. oceanica in the present study, the model reveals a negative feedback on increasing atmospheric CO2 concentrations owing to a decrease in the PIC/POC ratio.