ARTICLE IN PRESS

MARMIC-01239; No of Pages 8
Marine Micropaleontology xxx (2008) xxx–xxx

Contents lists available at ScienceDirect

Marine Micropaleontology
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / m a r m i c r o

Modelling the temperature dependent growth rates of planktic foraminifera

Fabien Lombard a,⁎, Laurent Labeyrie a, Elisabeth Michel a, Howard J. Spero b, David W. Lea c
a
Laboratoire des Sciences du Climat et de l'Environnement/Institut Pierre-Simon Laplace, laboratoire Commissariat à l'Energie Atomique/Centre National de la Recherche
Scientifique/Université de Versailles Saint-Quentin, avenue de la Terrasse, F-91198 Gif-sur-Yvette CEDEX, France
b
Department of Geology, University of California Davis, Davis CA, USA
c
Dept. of Earth Science, University of California, Santa Barbara, CA 93106-9630, USA

a r t i c l e i n f o a b s t r a c t

Article history: The temperature influence on foraminifera growth rate was analysed using a mechanistic
Received 24 April 2008 formulation that take into account enzyme inactivation at extreme temperatures. Growth rates
Received in revised form 17 September 2008 are calculated using available published and unpublished laboratory culture experiments for
Accepted 19 September 2008
eight species, including Neogloboquadrina pachyderma (sinistral and dextral forms),
Available online xxxx
Neogloboquadrina dutertrei, Globigerina bulloides, Globigerinoides ruber, Globigerinoides
sacculifer, Globigerinella siphonifera and Orbulina universa. Modeled growth formulas readily
Keywords:
reproduce the observed growth patterns for all species. Similar growth patterns are observed
Foraminifera
for the species that have the same symbiotic algae G. ruber, G. sacculifer, and O. universa.
Growth rates
Temperature However, different growth patterns are observed for herbivorous species (Neogloboquadrina
genus) compared to carnivorous species with or without symbionts. Our growth estimates
correspond well to in situ observations from both plankton tows and sediment traps. These
estimates will help to improve the quantification of the effects of environmental parameters on
foraminifera species distribution and abundance.
© 2008 Elsevier B.V. All rights reserved.

1. Introduction analysis of the foraminiferal specific distribution (Imbrie and

Kipp, 1971)). However, applications of such methods are
Zooplanktonic foraminifera are widely distributed from limited by the lack of constraints on their growth environ-
polar to equatorial seas (Bé and Tolderlund, 1971). Foramini- ment, especially for temperature (season and depth). For-
fera are protozoans that construct calcareous shells during aminifera species distribution and fluxes depend on
their lifecycle. Once planktonic foraminifera have completed phytoplankton productivity (Ortiz et al., 1995; Watkins et al.,
gametogenesis or died, empty shells sink rapidly through the 1996; Eguchi et al., 1999; Schiebel et al., 2001), season of
water column to the seafloor. Despite their low abundance in occurrence (Žarić et al., 2006) and the depth of growth
the plankton, foraminifera are responsible for 32–80% of the (Cleroux et al., 2007). Yet, most paleoceanographic recon-
global CaCO3 flux to the sediments (Schiebel, 2002). Fossilised structions are based on indirect links or statistical relation-
tests collected from deep sea sediment cores are commonly ships between climatic indicators (from atlases or local
used to reconstruct past ocean climate, notably as sea water measurements) and foraminifera assemblages observed in
temperature proxies (for example by their Mg:Ca ratio sediment core tops, plankton tows and sediment traps (Bé and
(Mashiotta et al., 1999; Elderfield and Ganssen, 2000) or Tolderlund, 1971). Core tops integrate decades to millennia of
ocean hydrological variability. Sediment traps provide good
temporal constraints, but neither core tops nor traps provide
independent constraints on the foraminifera depth of growth.
⁎ Corresponding author. Tel.: +33 169 823 534; fax: +33 169 823 568. Plankton may be collected by nets with parallel records of
E-mail addresses: Fabien.Lombard@lsce.ipsl.fr (F. Lombard),
Laurent.Labeyrie@lsce.ipsl.fr (L. Labeyrie), Elisabeth.michel@lsce.cnrs-gif.fr
hydrological parameters, but the method is time consuming,
(E. Michel), spero@geology.ucdavis.edu (H.J. Spero), lea@geol.ucsb.edu and only a small number of studies group a large number of
(D.W. Lea). plankton samples and the associated hydrological constraints.

0377-8398/$ – see front matter © 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.marmicro.2008.09.004

Please cite this article as: Lombard, F., et al., Modelling the temperature dependent growth rates of planktic foraminifera,
Marine Micropaleontology (2008), doi:10.1016/j.marmicro.2008.09.004
 ARTICLE IN PRESS
2 F. Lombard et al. / Marine Micropaleontology xxx (2008) xxx–xxx

Even fewer studies give simultaneous information on food point (b50 μE m− 2 s− 1) at which symbiont photosynthesis
availability (e.g. Kuroyanagi and Kawahata, 2004). exceeds respiration needs (Rink et al., 1998).
More precise knowledge on the conditions of foraminifera 3) Extreme salinity concentrations affect foraminiferal devel-
growth could be extracted from laboratory studies under opment, but growth is stable or only slightly impacted in
controlled conditions (e.g. Caron et al., 1987; Hemleben et al., the salinity range 30–40 psu (Bijma et al., 1990; Hemleben
1987; Bijma et al., 1990; Bijma et al., 1992; Spero and Lea, et al., 1987). Therefore we considered only observations
1996). But this type of information is often collected as a by- corresponding to the 30–40 psu salinity range.
product of geochemical studies. Sound interpretations may be
To be consistent in the analysis of published data, only
obtained only within the framework of recently acquired
mean initial size, final size of gametogenetic individuals and
knowledge on plankton biology (Kooijman, 2000). In that
mean survival time (i.e., time elapsed between the first
context, we propose an analytical formulation of the
measurement and gametogenesis) are considered. The num-
sensitivity of planktic foraminifera growth to temperature
ber of observations (foraminifera) corresponding to these
based on enzymatic thermodynamic properties for eight of
mean values are tabulated simultaneously. That number will
the most studied species with available laboratory data. This
be used as a weighting factor in the analysis. Individual growth
is one of the first steps necessary to reach an understanding of
rates, when recorded, are kept only as an indication of
the biological factors controlling the location, season and
individual variability. However, observations obtained at the
depth habitat of the foraminifera species proxies for paleo-
same temperature but with different protocols were consid-
ceanographic research.
ered as distinct data points, because different light intensity,
salinity, feeding frequency or food composition could influ-
2. Material and methods
ence foraminiferal growth. Published studies do not generally
consider experiments where all foraminifera died before
Growth observations on foraminifera were collected from
reaching gametogenesis. Such experiments often correspond
different sources (Table 1) including previously published
to extreme growth conditions which need to be taken in
papers (Caron et al., 1987a,b; Hemleben et al., 1987; Faber
account. We use such data but only give them a small weight in
et al., 1989; Bijma et al., 1990, 1992; Spero and Lea, 1993, 1996;
the analysis (i.e., number of observations = 1, regardless of the
Bijma et al., 1998), unpublished data associated with
number of individuals considered, which was usually 10–20).
published papers (Lea et al., 1996; Mashiotta et al., 1997;
The organic carbon weight of specimens is calculated from
Bemis et al., 1998; Mashiotta et al., 1999; Russel et al., 2004;
the initial and final size using a conversion factor (0.089 pgC
von Langen et al., 2005; Kimoto and Tsuchiya, 2006) or
μm− 3; Michaels et al., 1995) and assuming a spherical volume
previously unpublished data from two of the authors (HJS and
of foraminifera. The growth rate (µ, d− 1) is calculated
DWL). These data comprises eight different species including
assuming an exponential growth of foraminifera, following
Neogloboquadrina pachyderma (sinistral and dextral forms),
the formulation:
Neogloboquadrina dutertrei, Globigerina bulloides, Globigeri-
noides ruber, Globigerinoides sacculifer, Globigerinella siphoni- lnðWf =Wi Þ
fera and Orbulina universa. μ= ð1Þ
Δt
Growth rate is generally quantified by the number of
chambers precipitated within the observation period (as where Wi and Wf are initial and final foraminifera organic
chamber d− 1 or µm d− 1), for each studied foraminifera. Such carbon weights, respectively, in units of µgC ind− 1, and Δt (d)
data may not be used directly as these units strongly depend the time elapsed between the two measurements.
on the mean size of the considered specimens but also on the The temperature influence on growth (µ) is generally
fact that some foraminifera species have small and numerous considered with the convenient Q10 value that quantifies the
chambers whereas other have few large chambers. Therefore growth rate increase for a 10 °C increase:
it is necessary to reconsider the growth of the different
T=10
foraminifera species with a biologically relevant growth rate. cðT Þ = c0 Q10 ð2Þ
In our study, growth rate is quantified by the organic carbon
weight increase for each studied foraminifera. where µ0 (d− 1) is the growth rate at 0 °C. However, Q10 is only
Data selection is based on several criteria: adapted for small temperature ranges, and it varies as a
function to temperature. The Arrhenius relationship on the
1) Morphology: we consider individuals that constructed at other hand is more stable over a larger temperature range
least two chambers, underwent gametogenesis, and did (Koojiman, 2000). This relationship uses temperature on the
not show chamber resorption. These criteria are the same Kelvin scale and has the following form:
as those used in published studies (Caron et al., 1987a,b;
 
Hemleben et al., 1987; Faber et al., 1989; Bijma et al., 1990, T T
cðT Þ = cðT1 Þexp A − A ð3Þ
1992; Spero and Lea, 1993, 1996; Bijma et al., 1998), thus T1 T
keeping the data as homogeneous as possible.
2) Foraminiferal growth is strongly limited when animals are where µ(T1) (d− 1) is the growth rate for a chosen reference
not fed (Bé et al., 1981) or not illuminated (symbiont- temperature T1 and TA (°K) is the Arrhenius temperature. This
bearing species) (Caron et al., 1981). Specimens had to be relationship is equivalent to the Q10 formulation with:
maintained on a regular feeding schedule (fed daily, or  
every two or three days) and maintained under a dark/ 10dTA
Q10 = exp ð4Þ
light cycle with sufficient light to exceed the compensation TdT1

Please cite this article as: Lombard, F., et al., Modelling the temperature dependent growth rates of planktic foraminifera,
Marine Micropaleontology (2008), doi:10.1016/j.marmicro.2008.09.004
Marine Micropaleontology (2008), doi:10.1016/j.marmicro.2008.09.004
Please cite this article as: Lombard, F., et al., Modelling the temperature dependent growth rates of planktic foraminifera,

Table 1
Source and temperatures corresponding to the different growth observations of foraminifera

Species Source Temperature tested °C (specimens number)

F. Lombard et al. / Marine Micropaleontology xxx (2008) xxx–xxx

ARTICLE IN PRESS
G. sacculifer Bijma et al. (1990) 11.6 (3) 13.9 (14) 15.6 (17) 31 (21) 32.9 (1)
Hemleben et al. (1987) 19.5 (97; 59; 54; 36) 23.5 (80; 46; 56; 36)
26.5 (82; 57; 60; 47) 29.5 (94; 42; 60; 36)
Bijma et al. (1992) 23.5 (80; 46; 19; 24; 9; 49; 25; 27; 15) 26.5 (82; 57; 13; 8)
Caron et al. (1987a,b) 19.5 (141) 22 (114) 25 (111) 28 (112)
Spero and Lea (1993) 29 (16; 11; 10; 6)
Spero and Lea (pers. comm.) 21 (21) 23.3 (10) 25.1 (10) 26.9 (7) 29.2 (46) 29.3 (16)
G. siphonifera Bijma et al. (1990) 10 (1) 11.7 (20) 13.4 (33) 15.8 (22) 23.5 (92) 29.4 (17) 31 (1) 32.4 (1)
Bijma et al. (1992) 23.5 (42; 25; 31) 26.5 (31; 24)
Bijma et al. (1998) 23.5 (45; 15; 65; 9; 34; 27; 38; 30)
Faber et al. (1989) 26 (31; 30; 38; 22; 56; 55; 46)
Spero and Lea (pers. comm.) 23.3 (5) 25.1 (1) 29.3 (9)
O. universa Bijma et al. (1990) 11.6 (6) 16 (2) 17 (8) 18 (10) 25.8 (16) 30.7 (20) 32.5 (1)
Bijma et al. (1992) 23.5 (22; 28; 18) 26.5 (19; 15)
Caron et al. (1987a,b) 19.5 (60) 22 (63) 25 (62) 28 (32; 122)
Spero and Lea (pers. comm.) 15 (23) 16 (2) 17 (18) 21 (4) 22 (51) 23.3 (2) 25 (3) 27 (49) 29 (22) 29.2 (16)
G. ruber Bijma et al. (1990) 11.6 (4) 13.5 (2) 15.5 (6) 27.8 (9) 30.6 (5) 32.9 (1)
Bijma et al. (1992) 23.5 (19) 26.5 (36) 27.9 (13)
Spero and Lea (pers. comm.) 21 (28) 23.3 (23) 25.1 (29) 26.9 (9) 29.2 (62) 29.3 (31)
N. dutertrei Bijma et al. (1990) 13 (1) 14.4 (14) 16 (13) 17.9 (4) 28.6 (1) 31.5 (2) 33 (1)
Spero and Lea (pers. comm.) 6 (1) 9 (4) 12.9 (13) 16.2 (14) 19.2 (35)
G. bulloides Spero and Lea (1996) 16 (49; 30) 22 (30; 57)
Spero and Lea (pers. comm.) 6 (1) 9.9 (5) 12.9 (42) 14 (18) 16 (35) 17 (22) 22 (43) 25 (19) 28.3 (1)
N. pachyderma (dex.) Spero and Lea (pers. comm.) 4 (1) 6 (11) 9 (80) 12.9 (56) 16.2 (67) 19.2 (114) 23 (1)
N. pachyderma (sin.) Unpublished data 4.7 (1) 8 (1) 12 (1)

3
 ARTICLE IN PRESS
4 F. Lombard et al. / Marine Micropaleontology xxx (2008) xxx–xxx

Please cite this article as: Lombard, F., et al., Modelling the temperature dependent growth rates of planktic foraminifera,
Marine Micropaleontology (2008), doi:10.1016/j.marmicro.2008.09.004
 ARTICLE IN PRESS
F. Lombard et al. / Marine Micropaleontology xxx (2008) xxx–xxx 5

Table 2
Parameters of the mechanistic model used to reproduce the growth rate of the different foraminifera species

Parameters (±SD)

Species μ (T1) TA TL TH TAL TAH r2

O. universa 0.23 (±0.09) 626 (±3851) 289.8 (±3.4) 304.0 (± 0.5) 30,953 (± 13,119) 249,412 (± 389,004) 0.75
G. sacculifer 0.30 (±0.01) 2155 (±686) 289.3 (±0.6) 304.4 (± 0.3) 94,385 (± 23,740) 171,209 (± 48,514) 0.95
G. siphonifera 0.17 (±0.02) 7105 (±1861) 284.9 (±0.3) 301.8 (± 0.5) 349,991 (± 283,836) 130,852 (± 66,186) 0.77
G. ruber 0.24 (±0.11) 1086 (±5772) 292.3 (±2.3) 303.4 (± 0.5) 53,765 (± 21,035) 165,490 (± 102,487) 0.90
N. dutertrei 0.10 (±0.02) 6876 (±4805) 280.1 (±1.6) 298.8 (± 3.0) 210,746 (± 602,206) 57,855 (± 39,102) 0.63
G. bulloides 0.25 (±0.02) 6482 (±1256) 281.1 (± 0.3) 298.5 (± 0.3) 295,374 (± 266,947) 159,618 (± 44,756) 0.95
N. pachyderma (dex.) 0.13 (±0.01) 6584 (±1080) 277.8 (±0.3) 293.7 (± 0.5) 300,239 (± 249,645) 110,583 (± 35,218) 0.96
N. pachyderma (sin.) 0.37 (±0.05) 6584 a – 279.8 (± 0.7) – 59,491 (± 15,098) 0.90
a
The same TA from dextral N. pachyderma is assumed for the sinistral form.

The Arrhenius relationship, however, does not consider the model) and the 95% envelope defined by the t-test of
important physiological phenomena such as enzyme inacti- significance.
vation at low or high temperature that leads to a sharp
decrease in the observed growth rates at both ends of the 3. Results
optimal temperature range (Kooijman, 2000). Sharpe and
DeMichele (1977) proposed a mechanistic formulation for Growth rates are reported as a function of temperature
these rate reductions derived from the Arrhenius rate (Fig. 1). They follow approximately the same pattern for the
kinetics. This formulation is based on several simplifying different foraminifera species: a well-defined optimal tem-
assumptions: the growth rate of an organism at a given perature range with progressive growth rate increase, and no
temperature is assumed to be governed by a single rate- growth below or above that range. As a result, each species
controlling enzyme which is reversibly inactivated at low and shows a maximum growth rate at a temperature close to the
high temperatures; the total concentration of enzymes, in upper limit of the growth range. Within each species, the
both active and inactive form, is assumed to remain constant mean growth values derived from the different studies are
and independent of temperature; and the growth rate is a relatively similar. However, variability among specimens is
function of the ratio of active enzymes to inactive ones. Using important and differences between individuals can reach one
this formulation, growth rate µ is described as a function of order of magnitude for the same temperature.
temperature T (in Kelvin) (Kooijman, 2000): Growth models have been calibrated over their whole
temperature ranges for all species except for N. pachyderma
 
cðT1 Þexp TTA1 − TTA sinistral. For this species, due to the scarcity of observations,
μ ðT Þ =     ð5Þ only the maximal temperature decrease was fitted, assuming
1 + exp TTAL − TTALL + exp TTAH
H
− TTAH TA similar to the dextral form. The result of the growth model
calibration for each species is showed on Fig. 1 and the
where µ(T1) is the growth rate for an arbitrary chosen corresponding parameters in Table 2.
temperature T1 (20 °C or 293 K in this study) without The model correctly reproduces the general growth
considering enzymes inactivation and assuming only a normal pattern of all species, with r2 above 0.9 for five species (N.
increase of rate with temperature (Eq. (3)); TA is the Arrhenius pachyderma (sin. and dex. forms), G. bulloides, G. rubber and
temperature (Kooijman, 2000); TL and TH relate to the lower G. sacculifer), and between 0.63 and 0.75 for the remaining
and upper temperature boundaries of the growth tolerance three (G. siphonifera, O. universa and N. dutertrei); the lower
range and TAL and TAH are the Arrhenius temperatures for the portion of variance explained for these three could originate
decrease in growth rate respectively below and above these from discrepancies between the experimental protocols used
boundaries. All T (in K) are taken to be positive and in most in the different studies. The large deviations of the 95%
organisms the growth pattern correspond to TAH N TAL N TA. The confidence envelop observed for some of the species at the
upper part of the relationship relates the classical increase of lower and upper limits of the optimum temperature range
growth rate as a function of temperature whereas the lower derive from an insufficient number of data points covering the
part relates the enzymatic fraction that is in its active state. growth rate decrease for these extreme conditions. Orbulina
This is the analytical formula used in the present study. universa, G. sacculifer and G. ruber follow a similar growth
The parameters are adjusted by fitting Eq. (5) to the square pattern with an optimum growth between 20 and 29 °C and
root of the laboratory observed growth rates (Heitzer et al., only a slight increase of the growth rate between these
1991; Alber and Schaffner, 1992), taking into account the temperatures (TA comprised for the different species between
number of observations (Table 1) using a least square 626 and 2155 K), a slow growth decrease at the lower
minimisation (Nelder–Mead simplex method). Uncertainties temperature limit (TAL between 30,900 and 94,300 K) and
are estimated by the r2 (fraction of data variance explained by minimum and maximum of the temperature growth limits

Fig. 1. Growth rates (d− 1) of O. universa (A), G. sacculifer (B), G. siphonifera (C), G. ruber (D), N. dutertrei (E), G. bulloides (F), N. pachyderma dextral (G) and sinistral
forms (H) in relation to experimental temperature (°C). Observations from laboratory cultures are represented by large dots for the mean population at one
temperature or by small dots in the case of individual growth rates. Model results (line) and 95% confidence intervals of the model (dashed lines) are represented
for each species. Observations based on plankton tow samples (Bé and Tolderlund, 1971) are shown as stepped filled bars, and the flux observed in sediment traps
(Žarić et al., 2005) are shown in terms of overall distribution as a line and as optimal SST as an open bar.

Please cite this article as: Lombard, F., et al., Modelling the temperature dependent growth rates of planktic foraminifera,
Marine Micropaleontology (2008), doi:10.1016/j.marmicro.2008.09.004
 ARTICLE IN PRESS
6 F. Lombard et al. / Marine Micropaleontology xxx (2008) xxx–xxx

thermophilic strains (Pomeroy and Wiebe, 2001; Ratkowsky

et al., 2005; Miroshnichenko and Bonch-Osmolovskaya,
2006), macrophyte algae (Eggert and Wiencke, 2000),
phytoplankton algae (Suzuki and Takahashi 1995; El-Sabaawi
and Harrison, 2006), fresh water crustaceans (Yurista, 1999),
and corals (Howe and Marshall, 2002). To our knowledge,
however, this pattern has never been observed for oceanic
zooplankton. It is important to note that the pattern we
reconstruct for foraminifera growth rate is non symmetrical
and significantly different from the Gaussian response of
growth rate in function to temperature hypothesized by Fraile
et al. (2008) based on sediment trap flux observations Žarić et
al. (2005) In addition, and despite the fact that populations
observed in sediment traps certainly depends on other
parameters than just individual growth rate, the asymme-
trical growth pattern we observe seems to correspond to the
asymmetrical pattern in abundance observed in sediment
Fig. 2. Comparison of the modelled growth rate (d− 1) of the different traps in response to sea surface temperature (Žarić et al.,
foraminifera species in relation to experimental temperature (°C). The color 2005).
coding is: O. universa (green), G. sacculifer (blue), G. siphonifera (cyan), G.
The growth estimates used in our study derive from
ruber (red), N. dutertrei (black), G. bulloides (pink), N. pachyderma dextral
(dashed blue) and sinistral forms (dashed red). (For interpretation of the
different sources and protocols with different feeding inter-
references to colour in this figure legend, the reader is referred to the web vals, light illuminations or salinity. This may explain their
version of this article.) large dispersion that leads to lower r2 of the model
adjustment for some species (Fig. 1A–D). A large variability
is also observed between individuals cultivated within
respectively at 11 and 32 °C. Compared to these three species, apparently similar conditions (Fig. 1). At least part of that
the other species have a larger growth rate increase in variability may correspond to different foraminifera
function of temperature (TA between 6400 and 7100 K) and a responses to culture stress. Effectively, if some individuals
sharper growth rate decrease at their lower temperature limit recover rapidly from sampling and then grow relatively fast,
(TAL N 200,000 K). The optimal temperature ranges and others need more times to recover and produce only few shell
maximal growth limits are around 20–29 °C and 11–32 °C chambers in culture before reaching the gametogenesis stage.
respectively for G. siphonifera, 8–25 °C and 6–32 °C for N. For these individuals, growth rates may be slightly under-
dutertrei, 9–25 °C and 7–28 °C for G. bulloides and 6–20 °C and estimated and not represent the in situ growth behaviour of
4–23 °C for dextral N. pachyderma. Neogloboquadrina pachy- these species. Despite this variability, these growth observa-
derma sinistral presents an optimal growth below 5 °C and tions follow with a good confidence the general shape of the
negligible growth at T N 12 °C. A comparison of the different growth behaviour defined for each species.
species (Fig. 2) gives the highest growth rate for G. sacculifer The temperature dependence derived from laboratory
at high temperatures (17–30 °C), G. bulloides at intermediate growth rates is qualitatively in good agreement with the in
temperature (8–17 °C) and N. pachyderma s. under 7 °C. Or- situ occurrence and abundance of species observed with
bulina universa, G. siphonifera and G. ruber have intermediate planktons tows (Bé and Tolderlund, 1971) or in sediment traps
growth rates and dextral N. pachyderma and N. dutertrei have (Žarić et al., 2005) if we take into account available mean
the lowest growth rates. annual or seasonal sea surface temperatures (Fig. 1). Even if
the model fit is low for some species, (r2 between 0.63 and
4. Discussion and conclusions 0.95), the agreement between their growth pattern and in situ
observations is good for the warm species O. universa,
To our knowledge, this study is the first attempt that G. sacculifer, G. siphonifera, G. ruber and N. dutertrei. Their
quantifies the changes of foraminiferal growth rate with in situ temperature limits correspond to the laboratory
temperature, based on the increase of organic weight during determined growth, and their temperature range of maximum
laboratory cultures. Contrary to rates expressed in term of size abundance correspond to the optimal growth temperature
increase per day or chamber formation per day, this technique range in the model (Fig. 1A–E). This means that the parameter
is independent of the size of the organism and allows a TL and TH correspond to the in situ observations. Discrepancies
comparison of growth rates between species that have that do not appear to be attributable to the model (r2: 0.90–
different shell or chamber sizes. Our study shows that a 0.95) are seen for colder species, which we attribute to an
simple mechanistic formulation based on enzyme activity and insufficient knowledge of the precise conditions of growth in
inactivation at extremes temperatures can efficiently repre- the oceans (in particular seasonality in temperature and depth
sent the foraminifera growth pattern. This model has been habitat). For G. bulloides, our model estimation corresponds to
fitted with success for the eight species studied and allows for the observations in sediment traps but individuals were
characterization of their complete growth curves, including observed in colder conditions in plankton nets (Fig. 1F). This
the upper and lower growth limits, except for N. pachyderma could originate from a disparity between the mean annual sea
sinistral. Similar growth patterns have been observed for surface temperature as used in Bé and Tolderlund (1971) and
different species of bacteria, including marine endobiotic or the spring to summer occurrence of G. bulloides under

Please cite this article as: Lombard, F., et al., Modelling the temperature dependent growth rates of planktic foraminifera,
Marine Micropaleontology (2008), doi:10.1016/j.marmicro.2008.09.004
 ARTICLE IN PRESS
F. Lombard et al. / Marine Micropaleontology xxx (2008) xxx–xxx 7

temperate conditions (Kuroyanagi et al., 2002; King and light and food availability in order to reproduce the natural
Howard, 2003; Bárcena et al., 2004; Marchant et al., 2004). seasonal and spatial variability of foraminiferal species
But this can also originate from discrepancies between the abundance. Ideally, it would also be valuable to explore
diversity of genetic types observed in situ (Darling et al., 1999; growth patterns in laboratory culture using different for-
Kucera and Darling, 2002), whereas all our observations on G. aminiferal genotypes of common species to test the signifi-
bulloides comes from the same location (California coast) and cance of genetic diversity on the growth pattern of cryptic
cannot represent the overall diversity. The oceanic distribution species.
of N. pachyderma sinistral and dextral as defined from sediment
traps (Žarić et al., 2005) is apparently larger than the growth Acknowledgments
limits modelled from foraminifera cultures (Fig. 1G–H). This
could originate from their adaptation to a sub-surface habitat if We greatly thank J.-C. Duplessy, E. Cortijo, G. Gorsky and
sea surface temperature is too high and sufficient food is the members of the Forclim Team for constructive discussions
available (Kuroyanagi and Kawahata, 2004), although part of and their improvement of the manuscript, the French program
that discrepancy may also be derived from a higher culture ANR05-BLAN0275-01 Forclim, the CEA and the CNRS for their
stress near the upper temperature limits for these species. For base support to the LSCE. We are grateful to the staff of the
N. pachyderma, Bé and Tolderlund (1971) give only information Wrigley Marine Science Center for providing a world-class
on the total abundance of this species with the indication that field station for the experiments that generated much of these
the dextral to sinistral shift occurs at an approximate data, and the many students, assistants and colleagues who
temperature of 7.2 °C. This corresponds well to the shift in participated in the foraminifera culturing program during the
growth rate modelled between sinistral and dextral forms past decade. This research was supported by U.S. National
(7.8 °C Fig. 2). Science Foundation (NSF) Grants 9416595, 0550703 (HJS) and
When comparing the species growth rates (Table 2, Fig. 2), 9415991, 9729327 (DWL). During the writing of this manu-
different growth patterns are observed. For G. sacculifer, O. script, HJS was supported by the NSF while he worked at the
universa and G. ruber, the increase in growth rate with Foundation. Any opinions, findings, and conclusions or
temperature appears relatively small. This is also the case for recommendations expressed in this paper are those of the
the growth rate increase below the optimum temperature authors and do not necessarily reflect the views of the NSF.
range when compared to the other species. Knowing that
these three species have the same symbiotic algae species References
(Gast and Caron, 2001; Shaked and de Vargas, 2006), the
temperature sensitivity of that algae could potentially drive Alber, S., Schaffner, D., 1992. Evaluation of data transformations used with the
square root and schoolfield models for predicting bacterial-growth rate.
this particular growth pattern. A clear distinction can also be Appl. Environ. Microbiol. 58, 3337–3342.
made between carnivorous species that have a high growth Bárcena, M.A., et al., 2004. Planktonic response to main oceanographic
rate and herbivorous species (N. dutertrei, N. pachyderma, changes in the Alboran Sea (Western Mediterranean) as documented in
sediment traps and surface sediments. Mar. Micropaleontol. 53,
Hemleben et al., 1989) that have a low growth rate. This 423–445.
difference could originate both from an effectively different Bé, A.W.H., Tolderlund, D.S., 1971. 6. Distribution and ecology of living
growth rate linked to the energetic content of their prey or planktonic foraminifera in surface waters of the Atlantic and Indian
Oceans. In: Funnell, B.M., Riedel, W.R. (Eds.), Micropaleontology of
from an incorrect feeding of the herbivorous species in the
Oceans. Cambridge University Press, London, pp. 105–149.
laboratory cultures due to difficulties to provide them with Bé, A.W.H., Caron, D.A., Anderson, O.R., 1981. Effect of feeding frequency on
the appropriate food. life processes of the planktonic foraminifer Globigerinoides sacculifer in
laboratory cultures. J. Mar. Biol. Assoc. U. K. 61, 257–277.
In addition, the species-specific temperature dependence
Bemis, B.E., Spero, H.J., Bijma, J., Lea, D.W., 1998. Reevaluation of the oxygen
of growth rates (Fig. 2) may explain at least in part the isotopic composition of planktonic foraminifera: experimental results
foraminifera species distribution in the oceanic environment. and revised paleotemperature equation. Paleoceanography 13, 150–160.
There is evidence that the predominance of one population Bijma, J., Faber, W.W.J., Hemleben, C., 1990. Temperature and salinity limits
for growth and survival of some planktonic foraminifers in laboratory
over others could depend on specific growth rates (Eppley cultures. J. Foraminiferal Res. 20, 95–116.
1972; Goldman and Carpenter, 1974). Species that have a Bijma, J., Hemleben, C., Oberhänsli, H., Spindler, M., 1992. The effect of
higher growth rate would be successful competitors and increased water fertility on tropical spinose planktonic foraminifers in
laboratory cultures. J. Foraminiferal Res. 22, 242–256.
dominate the other species however other processes such as Bijma, J., Hemleben, C., Huber, B.T., Erlenkeuser, H., Kroon, D.,1998. Experimental
reproduction, mortality and predation rates had also to be determination of the ontogenic stable isotope variability in two morpho-
considered. This may also be the case for foraminifera. Our types of Globigerinella siphonifera. Mar. Micropaleontol. 35, 141–160.
Caron, D.A., Bé, A.W.H., Anderson, O.R., 1981. Effect of variations in light
model results show that, depending on water temperature intensity on life processes of planktonic foraminifer Globigerinoides
range, N. pachyderma, G. bulloides and G. sacculifer should sacculifer in laboratory cultures. J. Mar. Biol. Assoc. U. K. 62, 435–451.
dominate the foraminifera population. N. pachyderma is Caron, D.A., Faber, W.W.J., Bé, A.W.H., 1987a. Effect of temperature and
salinity on the growth and survival of the planktonic foraminifer Globi-
effectively found to be the dominant species in polar areas,
gerinoides sacculifer. J. Mar. Biol. Assoc. U. K. 67, 323–341.
G. bulloides in temperate regions and, depending on the Caron, D.A., Faber, W.W.J., Bé, A.W.H., 1987b. Growth of the spinose planktonic
oligotrophy of the zone, G. ruber and G. sacculifer are foraminifer Orbulina universa in laboratory culture and the effect of
temperature on life processes. J. Mar. Biol. Assoc. U. K. 67, 343–358.
dominant in tropical areas (Bé and Tolderlund, 1971). This
Cleroux, C., Cortijo, E., Duplessy, J.-C., Zahn, R., 2007. Deep dwelling
model provides a new perspective on foraminiferal species foraminifera as thermocline temperature recorder. Geochem. Geophys.
distribution and abundance in the open ocean that could Geosystems 8, Q04N11. doi:10.1029/2006GC001474.
strengthen paleo-ecological interpretation of fossil data. Darling, K.F., Wade, C.M., Kroon, D., Leigh Brown, A.J., Bijma, J., 1999. The
diversity and distribution of modern planktic foraminiferal small subunit
However, for a fully integrated application, this model needs ribosomal RNA genotypes and their potential as tracers of present and
to be combined with other environmental parameters such as past ocean circulation. Paleoceanography 14, 3–12.

Please cite this article as: Lombard, F., et al., Modelling the temperature dependent growth rates of planktic foraminifera,
Marine Micropaleontology (2008), doi:10.1016/j.marmicro.2008.09.004
 ARTICLE IN PRESS
8 F. Lombard et al. / Marine Micropaleontology xxx (2008) xxx–xxx

Eggert, A., Wiencke, C., 2000. Adaptation and acclimatation of growth and Mashiotta, T.A., Lea, D.W., Spero, H.J., 1997. Experimental determination of
photosynthesis of five Antarctic red algae to low temperatures. Polar. cadmium uptake in shells of the planktonic foraminifera Orbulina
Biol. 23, 609–618. universa and Globigerina bulloides: Implication for surface water
Eguchi, N.O., Kawahata, H., Taira, A., 1999. Seasonal response of planktonic paleoreconstructions. Geochim. Cosmochim. Acta 61, 4053–4065.
foraminifera to surface ocean condition: sediment trap results from the Mashiotta, T.A., Lea, D.W., Spero, H.J., 1999. Glacial–interglacial changes in
central North Pacific Ocean. J. Oceanogr. 55, 681–691. Subantarctic sea surface temperature and δ18O-water using foraminiferal
El-Sabaawi, R., Harrison, P.J., 2006. Interactive effects of irradiance and Mg. Earth Planet Sci. Lett. 170, 417–432.
temperature on the photosynthetic physiology of the pennate diatom Michaels, A.F., Caron, D.A., Swanberg, N.R., Howse, F.A., Michaels, C.M., 1995.
Pseudo-nitzschia granii (Bacilariophyceae) from the Northeast Subarctic Planktonic sarcodines (Acantharia, Radiolaria, Foraminifera) in surface
Pacific. J. Phycol. 42, 778–785. waters near Bermuda: abundance, biomass and vertical flux. J. Plankton
Elderfield, H., Ganssen, G., 2000. Past temperature and δ18O of surface ocean Res. 17, 131–163.
waters inferred from foraminiferal Mg/Ca ratios. Nature 405, 442–445. Miroshnichenko, M.L., Bonch-Osmolovskaya, E.A., 2006. Recent develop-
Eppley, R.W., 1972. Temperature and phytoplankton growth in the sea. Fish. ments in the thermophilic microbiology of deep-sea hydrothermal vents.
Bull. 70, 1063–1085. Extremophiles 10, 85–96.
Faber, W.W.J., Anderson, O.R., Caron, D.A.,1989. Algal–foraminiferal symbiosis in Ortiz, J.D., Mix, A.C., Collier, R.W., 1995. Environmental control of living
the planktonic foraminifer Globigerinella aequilateralis: II. Effects of two symbiotic and asymbiotic foraminifera of the California Current.
symbiont species on foraminiferal growth and longevity. J. Foraminiferal Paleoceanography 10, 987–1009.
Res. 19, 185–193. Pomeroy, L.R., Wiebe, W.J., 2001. Temperature and substrates as interactive
Fraile, I., Schulz, M., Mulitza, S., Kucera, M., 2008. Predicting the global limiting factors for marine heterotrophic bacteria. Aquat. Microb. Ecol. 23,
distribution of planktonic foraminifera using a dynamic ecosystem 187–204.
model. Biogeosciences 5, 891–911. Ratkowsky, D.A., Olley, J., Ross, T., 2005. Unifying temperature effects on the
Gast, R.J., Caron, D.A., 2001. Photosymbiotic associations in planktonic growth rate of bacteria and the stability of globular proteins. J. Theor.
foraminifera and radiolaria. Hydrobiologia 461, 1–7. Biol. 233, 351–362.
Goldman, J., Carpenter, E.J., 1974. A kinetic approach to the effect of Rink, S., Kühl, M., Bijma, J., Spero, H.J., 1998. Microsensor studies of
temperature on algal growth. Limnol. Oceanogr. 19, 756–766. photosynthesis and respiration in the symbiotic foraminifer Orbulina
Heitzer, A., Kohler, H., Reichert, P., Hamer, G., 1991. Utility of phenomen- universa. Mar. Biol. 131, 583–595.
ological models for describing temperature-dependence of bacterial- Russel, A.D., Hönisch, B., Spero, H.J., Lea, D.W., 2004. Effects of seawater
growth. Appl. Environ. Microbiol. 57, 2656–2665. carbonate ion concentration and temperature on shell U, Mg, and Sr in
Hemleben, C., Spindler, M., Breitinger, I., Ott, R., 1987. Morphological and cultured planktonic foraminifera. Geochim. Cosmochim. Acta 68,
physiological response of Globigerinoides sacculifer (Brady) under varying 4347–4361.
laboratory conditions. Mar. Micropaleontol. 12, 305–324. Schiebel, R., 2002. Planktic foraminiferal sedimentation and the marine calcite
Hemleben, C., Spindler, M., Anderson, O.R., 1989. Modern Planktonic budget. Glob. Biogeochem. Cycles 16, 1065. doi:10.1029/2001GB001459.
Foraminifera. Springer-Verlag, New York. 363 pp. Schiebel, R., Waniek, J., Bork, M., Hemleben, C., 2001. Planktic foraminiferal
Howe, S., Marshall, A.T., 2002. Temperature effects on calcification rate and production simulated by chlorophyll redistribution and entrainment of
skeletal deposition in the temperate coral, Plesiastrea versipora nutrients. Deep-Sea Res. Part I 48, 721–740.
(Lamarck). J. Exp. Mar. Biol. Ecol. 275, 63–81. Shaked, Y., de Vargas, C., 2006. Pelagic photosymbiosis: rDNA assessment of
Imbrie, J., Kipp, N.G., 1971. A new micropaleontological method for diversity and evolution of dinoflagelate symbionts and planktonic
paleoclimatology: Application to a Late Pleistocene Caribbean core, The foraminiferal host. Mar. Ecol. Prog. Ser. 325, 59–71.
Late Cenozoic Glacial Ages. Yale University Press, New Haven, pp. 71–181. Sharpe, P.J.H., DeMichele, D.W., 1977. Reaction kinetics of poikilotherm
Kimoto, K., Tsuchiya, M., 2006. The “unusual” reproduction of planktic development. J. Theor. Biol. 64, 649–670.
foraminifera: an assexual reproductive phase of Neogloboquadrina Spero, H.J., Lea, D.W., 1993. Intraspecific stable isotope variability in the
pachyderma. Anu. Inst. Geociênc. 29, 461. planktonic foraminifera Globigerinoides sacculifer: results from labora-
King, A.L., Howard, W.R., 2003. Planktonic foraminiferal flux seasonality in tory experiments. Mar. Micropaleontol. 22, 221–234.
Subantarctic sediment traps: a test for paleoclimate reconstructions. Spero, H.J., Lea, D.W., 1996. Experimental determination of intraspecific
Paleoceanography 18, 1019. doi:10.1029/2002PA000839. stable isotope variability in Globigerina bulloides: Implications for
Kooijman, S.A.L.M., 2000. Dynamic Energy and Mass Budgets in Biological paleoceanographic reconstructions. Mar. Micropaleontol. 28, 231–246.
Systems. Cambridge university press, Cambridge. 426 pp. Suzuki, Y., Takahashi, M., 1995. Growth responses of several diatom species
Kucera, M., Darling, K.F., 2002. Cryptic species of planktonic foraminifera: isolated from various environments to temperature. J. Phycol. 31,
their effect on palaeoceanographic reconstructions. Philos. Trans. R. Soc. 880–888.
Lond. A 360, 695–719. von Langen, P.J., Pak, D.K., Spero, H.J., Lea, D.W., 2005. Effect of temperature
Kuroyanagi, A., Kawahata, H., 2004. Vertical distribution of living planktonic on Mg/Ca in neogloboquadrinid shells determined by live culturing.
foraminifera in the seas around Japan. Mar. Micropaleontol. 53, 173–196. Geochem. Geophys. Geosystems 6, Q10P03. doi:10.1029/2005GC000989.
Kuroyanagi, A., Kawahata, H., Nishi, H., Honda, M.C., 2002. Seasonal changes Watkins, J.M., Mix, A.C., Wilson, J., 1996. Living planktic foraminifera: tracers
in planktonic foraminifera in the northwestern North Pacific Ocean: of circulation and productivity regimes in the central equatorial Pacific.
sediment trap experiments from subarctic and subtropical gyres. Deep- Deep-Sea Res. Part II 43, 1257–1282.
Sea Res. Part II 49, 5627–5645. Yurista, P.M., 1999. Temperature-dependant energy budget of an Arctic
Lea, D.W., Martin, P.A., Chan, D.A., Spero, H.J., 1996. Experimental determina- Cladoceran, Daphnia middendorffiana. Freshw. Biol. 42, 21–34.
tion of stable isotope variability in Globigerina bulloides: implication for Žarić, S., Donner, B., Fischer, G., Mulitza, S., Wefer, G., 2005. Sensitivity of
paleoceanographic reconstructions. Mar. Micropaleontol. 28, 231–246. planktic foraminifera to sea surface temperature and export production
Marchant, M., Hebbeln, D., Giglioc, S., Coloma, C., González, H.E., 2004. as derived from sediment trap data. Mar. Micropaleontol. 55, 75–105.
Seasonal and interannual variability in the fluxof planktic foraminifera in Žarić, S., Schultz, M., Mulitza, S., 2006. Global prediction of planktic
the Humboldt Current System off central Chile (301S). Deep-Sea Res. foraminiferal fluxes from hydrographic and productivity data. Biogeos-
Part II 51, 2441–2455. ciences 3, 187–207.

Please cite this article as: Lombard, F., et al., Modelling the temperature dependent growth rates of planktic foraminifera,
Marine Micropaleontology (2008), doi:10.1016/j.marmicro.2008.09.004