Atmospheric science: Marked decline in atmospheric carbon dioxide concentrations during the Paleogene

Abstract

The relation between the partial pressure of atmospheric carbon dioxide (pCO2) and Paleogene climate is poorly resolved. We used stable carbon isotopic values of di-unsaturated alkenones extracted from deep sea cores to reconstruct pCO2 from the middle Eocene to the late Oligocene (approximately 45 to 25 million years ago). Our results demonstrate that pCO2 ranged between 1000 to 1500 parts per million by volume in the middle to late Eocene, then decreased in several steps during the Oligocene, and reached modern levels by the latest Oligocene. The fall in pCO2 likely allowed for a critical expansion of ice sheets on Antarctica and promoted conditions that forced the onset of terrestrial C4 photosynthesis.

Full text

numbers CZ551658 to CZ552046. We thank members
of the Rubin and Pa
¨
a
¨
bo laboratories for insightful
discussions and support. This work was performed
under the auspices of the U.S. Department of Energy’s
Office of Science Biological and Environmental Re-
search Program and by the University of California;
Lawrence Berkeley National Laboratory; Lawrence
Livermore National Laboratory; and Los Alamos Na-
tional Laboratory under contract numbers DE-AC03-
76SF00098, W-7405-Eng-48, and W-7405-ENG-36,
respectively, with support from NIH grants U1
HL66681B and T32 HL07279 and at the Max Planck
Institute for Evolutionary Anthropology.
Supporting Online Material
www.sciencemag.org/cgi/content/full/1113485/DC1
Materials and Methods
Tables S1 to S3
References
12 April 2005; accepted 26 May 2005
Published online 2 June 2005;
10.1126/science.1113485
Include this information when citing this paper.
Marked Decline in Atmospheric
Carbon Dioxide Concentrations
During the Paleogene
Mark Pagani,
1
James C. Zachos,
2
Katherine H. Freeman,
3
Brett Tipple,
1
Stephen Bohaty
2
The relation between the partial pressure of atmospheric carbon dioxide ( pCO
2
)
and Paleogene climate is poorly resolved. We used stable carbon isotopic values
of di-unsaturated alkenones extracted from deep sea cores to reconstruct pCO
2
fromthemiddleEocenetothelateOligocene(È45 to 25 million years ago). Our
results demonstrate that pCO
2
ranged between 1000 to 1500 parts per million
by volume in the middle to late Eocene, then decreased in several steps during
the Oligocene, and reached modern levels by the latest Oligocene. The fall in
pCO
2
likely allowed for a critical expansion of ice sheets on Antarctica and
promoted conditions that forced the onset of terrestrial C
4
photosynthesis.
The early Eocene EÈ52 to 55 million years
ago (Ma)^ climatewasthewarmestofthepast
65 million years. Mean annual continental
temperatures were considerably elevated rel-
ative to those of today, and high latitudes
were ice-free, with polar winter temperatures
È10-C warmer than at present (1–3). After
this climatic optimum, surface- and bottom-
water temperatures steadily cooled over È20
million of years (4, 5), interrupted by at least
one major ephemeral warming in the late mid-
dle Eocene (6). High-latitude cooling even-
tually sustained small Antarctic ice sheets by
the late Eocene (7), culminating in a striking
climate shift across the Eocene/Oligocene
boundary (E/O) at 33.7 Ma. The E/O climate
transition, Earth_s first clear step into Bicehouse[
conditions during the Cenozoic, is associated
with a rapid expansion of large continental
ice sheets on Antarctica (8, 9) in less than
È350,000 years (10, 11).
Changes in the partial pressure of atmo-
spheric carbon dioxide (pCO
2
) are largely
credited for the evolution of global climates
during the Cenozoic (12–14). However, the
relation between pCO
2
and the extraordinary
climate history of the Paleogene is poorly
constrained. Initial attempts to estimate early
Paleogene pCO
2
have provided conflicting
results, with both high (15) and low (i.e., sim-
ilar to modern) (16) estimates of pCO
2
.This
deficiency in our understanding of the history
of pCO
2
is critical, because the role of CO
2
in forcing long-term climate change during
some intervals of Earth_s history is equivocal.
For example, Miocene pCO
2
records (È25 to
5 Ma) argue for a decoupling between global
climate and CO
2
(15–17). These records sug-
gest that Miocene pCO
2
was rather low and
invariant across periods of both inferred
global warming and high-latitude cooling
(17). Clearly, a more complete understanding
of the relation between pCO
2
and climate
change requires the extension of paleo-pCO
2
records back into periods when Earth was
substantially warmer and ice-free.
Paleoatmospheric CO
2
concentrations can
be estimated from the stable carbon isotopic
compositions of sedimentary organic molecules
known as alkenones. Alkenones are long-
chained (C
37
-C
39
) unsaturated ethyl and meth-
yl ketones produced by a few species of
Haptophyte algae in the modern ocean (18).
Alkenone-based pCO
2
estimates derive from
records of the carbon isotopic fractionation that
occurred during marine photosynthetic carbon
fixation (e
p
). Chemostat experiments conducted
under nitrate-limited conditions indicate that
alkenone-based e
p
values (e
p37:2
)varyasa
function of the concentration of aqueous CO
2
(ECO
2aq
^) and specific growth rate (19–21).
These experiments also provide evidence that
cell geometry accounts for differences in e
p
among marine microalgae cultured under sim-
ilar conditions (21). In contrast, results from
dilute batch cultures conducted under nutrient-
replete conditions yield substantially lower
e
p37:2
values, a different relation for e
p
versus
m/CO
2aq
(where m 0 algal growth rate), and a
minimal response to ECO
2aq
^ (22). Thus, com-
parison of the available culture data suggests
that different growth and environmental condi-
tions potentially trigger different carbon up-
take pathways and carbon isotopic responses
(23). A recent evaluation of the efficacy of the
alkenone-CO
2
approach, using sedimentary
alkenones in the natural environment, sup-
ported the capacity of the technique to re-
solve relatively small differences in water
column ECO
2aq
^ across a variety of marine
environments when phosphate concentra-
tions and temperatures are constrained (24).
In our study, we extended records of the
carbon isotopic composition of sedimentary
alkenones (d
13
C
37:2
) from the middle Eocene
to the late Oligocene and established a record
of pCO
2
for the past È45 million years.
Samples from Deep Sea Drilling Project sites
516, 511, 513, and 612 and Ocean Drilling
Program site 803 (Fig. 1) were used to re-
construct d
13
C
37:2
and e
p37:2
records ranging
from the middle Eocene to the late Oligocene
(È25 to 45 Ma). These sites presumably rep-
resent a range of oceanic environments with a
variety of surface-water nutrient and algal-
growth conditions and thus reflect a set of
environmental and physiological factors af-
fecting both d
13
C
37:2
and e
p37:2
values.
These data are presented as a composite
record, in large part because the measurable
concentration of di-unsaturated alkenones var-
ied both spatially and temporally. Moreover,
continuous alkenone records spanning the
entire Eocene and Oligocene from individual
sites were not recovered. As a consequence,
most of the Oligocene record is represented at
site 516, whereas the majority of the Eocene is
represented at site 612 (Fig. 2A). Age models
for each site were developed by linearly
interpolating between biostratigraphic datums
(25–31), calibrated to the Geomagnetic Polar-
ity Time Scale (32).
Eocene d
13
C
37:2
values range from È–30
to –35 per mil (°), with the most negative
values (sites 511 and 513) occurring near the
E/O boundary. d
13
C
37:2
values increase sub-
stantially through the Oligocene with maxi-
mumvaluesofÈ–27° by È25.5 Ma. This
trend is briefly reversed near the end of the
Oligocene as d
13
C
37:2
values become more
negative, reaching È–32° by 25 Ma (Fig.
2A). The overall pattern of
13
C enrichment
continues into the Miocene, establishing a
clear secular trend from the middle Eocene to
the middle Miocene (Fig. 2B). These isotopic
1
Department of Geology and Geophysics, Yale Univer-
sity, 210 Whitney Avenue, New Haven, CT 06511, USA.
2
Earth Sciences Department, University of California,
1156 High Street, Santa Cruz, CA 95064, USA.
3
De-
partment of Geosciences, Pennsylvania State Univer-
sity, University Park, PA 16802, USA.
R EPORTS
22 JULY 2005 VOL 309 SCIENCE www.sciencemag.org
600

trends do not mirror changes in the d
13
Cof
dissolved inorganic carbon (d
13
C
DIC
) because
d
13
C records of bulk carbonate (33) and ben-
thic foraminifera (10) indicate small changes in
d
13
C
DIC
for the Eocene to Oligocene relative
to the change in d
13
C
37:2
. Nonetheless, inter-
pretations of long-term trends in d
13
C
37:2
are
enhanced when d
13
C
37:2
values are converted
to e
p37:2
(34), thus eliminating the influence
of d
13
C
DIC
.
The temporal pattern of e
p37:2
is similar to
that of d
13
C
37:2
(Fig. 2, C and D), consistent
with other studies (17). Higher values of e
p37:2
(È19.5 to 21.5°) characterize the Eocene
and then decrease through the Oligocene.
The e
p37:2
values recorded for the Eocene
and earliest Oligocene are higher than any
recorded for the modern ocean (Fig. 2D).
Given our present understanding of the con-
trols on e
p37:2
, the decrease in e
p37:2
from the
Eocene through the Oligocene could be driven
by a consistent change in the cell dimensions
of alkenone-producing algae over time, a sec-
ular increase in growth rates of alkenone-
producing algae, or a long-term decrease in
ECO
2aq
^ and/or increased utilization of bicar-
bonate (EHCO
3
–
^) as a result of low ECO
2aq
^
(35). At present, evolutionary changes in algal
cell geometries are poorly constrained. If the
long-term decrease in e
p37:2
were driven solely
by changes in algal cell dimensions, it would
require a pattern of increasing ratios of cell
volume to surface area with time. If e
p
scales
linearly with the ratio of cell volume to sur-
face area (21), the observed change in e
p37:2
values would require an È60% increase in the
cell diameters of alkenone-producing algae
from the Eocene to the Miocene (i.e., sites 516
and 612). Further, given that Miocene and
Modern e
p37:2
values are similar, Eocene coc-
colithophores would have to have been È60%
smaller than modern alkenone producers, such
as Emiliania huxleyi, with cell diameters of
È5 mm(21). However, the available data
suggest that placoliths from probable alke-
none producers, specifically species within
the genus Reticulofenestra, were substantial-
ly larger than modern species and then de-
creased through the Oligocene and early
Miocene (36, 37). If we reasonably assume
that placolith geometry scales to cell geome-
try (38), then cell diameters decreased during
the late Paleogene. A trend of decreasing cell
diameters should lead to an increase in e
p37:2
values (21), which is contrary to our measure-
ments. Thus, although a long-term change in
cell geometry might have influenced the rel-
ative magnitude of Paleogene e
p37:2
values, it
was not responsible for the pattern observed
in our record.
Alternatively, variations in e
p37:2
could be
ascribed to variations in the specific growth
rates of alkenone-producing algae (m
alk
), with
higher m
alk
values associated with lower e
p37:2
values. Under this scenario, Eocene and early
Fig. 1. Site location
map. Sites 612, 516,
803, 511, and 513
were used to recon-
struct Eocene and Ol-
igocene e
p37:2
values.
Sites 588, 608, 730,
and 516 were used to
reconstruct Miocene
e
p37:2
values.
90° 120° 150°E 180° 150°W 120° 90° 60° 30°W 0° 30°E 60°
60°S
40°S
20°S
0°
20°N
40°N
60°N
608
612
730
516
513
511
803
588
-25
-26
-27
-28
-29
-30
-31
-32
-33
-34
-35
-36
-37
δ
13
C
37:2
(‰, PDB)
Miocene
Oligocene
Eocene
-15
-17
-19
-21
-23
-25
-27
-29
-31
-33
-35
-37
δ
13
C
37:2
(‰, PDB)
site 612
site 511
site 513
site 803
site 516
site 516 (ref 53)
site 608 (ref 17)
site 588 (ref 17,42)
site 730 (ref 17)
Miocene
Oligocene
Eocene
14
15
16
17
18
19
20
21
22
23
24
15 20 25 30 35 40 45 0
Miocene
Oligocene
Eocene
Age (Ma)
ε
p
(‰)
ε
p
(‰)
3
5
7
9
11
13
15
17
19
21
23
10 15 20 25 30 35 40 45 50
A
g
e (Ma)
Miocene
Oligocene
Eocene
5
A
B
C
D
Fig. 2. (A) Stable carbon isotopic composition of di-unsaturated alkenones. Each data point rep-
resents one measurement or an average of multiple measurements, with error bars bracketing the
range of values for each sample. PDB, Pee Dee belemnite standard. (B) Compilation of the carbon
isotopic composition of di-unsaturated alkenones from this study and Pagani et al.(17, 42, 53). (C)
Paleogene e
p37:2
values. e
p37:2
is calculated from the d
13
C of di-unsaturated alkenones as follows:
e
p37:2
0 [(dd þ 1000/dp þ 1000) – 1]  10
3
,wheredd is the carbon isotopic composition of CO
2aq
calculated from mixed-layer carbonates and dp is the carbon isotopic composition of haptophyte
organic matter enriched by 4.2° relative to alkenone d
13
C(54). Carbon isotopic compositions of
mixed-layer carbonates were used to calculate dd by assuming equilibrium conditions and applying
temperature-dependent isotope equations (55, 56). Mixed-layer temperatures were calculated from
the d
18
O of planktonic foraminifera (57) as follows: site 612, Acarinina spp.; site 513, Subbotina spp.
and Chiloguembelina cubensis; and site 511, Subbotina spp. Temperatures for sites 516 and 803 were
estimated from the d
18
OcompositionsoftheG60-mm carbonate fraction, assuming that the d
18
O
composition of seawater changed from –0.75° during the Eocene to –0.5° during the Oligocene.
Error bars reflect the range of e
p37:2
values calculated by applying the maxima and minima of both
the measured d
13
C of di-unsaturated alkenones and calculated temperatures. (D) Compilation of
e
p37:2
values from this study and Pagani et al.(17, 42, 53). Dashed horizontal lines bracket the range of
e
p37:2
values from surface waters of modern oceans. In general, higher and lower e
p37:2
values come
from oligotrophic and eutrophic environments, respectively. The shaded box represents the range of
e
p37:2
values from oligotrophic sites where [PO
4
3–
] ranges between 0.0 to 0.2 mmol/liter.
R EPORTS
www.sciencemag.org SCIENCE VOL 309 22 JULY 2005
601

Oligocene e
p37:2
values must reflect substantial-
ly lower m
alk
than modern m
alk
found in oligo-
trophic waters where EPO
4
3–
^ is È0 mmol/liter
(Fig. 2D). That is, algal growth rates during
the Paleogene from both eutrophic and oligo-
trophic environments would have to be lower
than the lowest growth rates found in the
modern oligotrophic ocean. Further, if growth
rates were indeed the first-order control on
e
p37:2
values, the lowest Miocene e
p37:2
values
would require substantially higher algal growth
rates in oligotrophic settings, comparable to
those of the highly productive Peru upwelling
margin (Fig. 2D). Therefore, we conclude that
rather extraordinary changes in m
alk
are
required to explain the temporal pattern of
e
p37:2
values and thus are not the primary
cause for the observed long-term trends. In-
stead, we contend that the Cenozoic evolution
of e
p37:2
was forced primarily by changes in
ECO
2aq
^ and pCO
2
. Accordingly, these records
would qualitatively reflect high surface-water
ECO
2aq
^ during the middle to late Eocene, a
pattern of decreasing ECO
2aq
^ through the
Oligocene, and near-modern levels during the
Neogene. If the change in e
p37:2
values during
the Paleogene was brought about by an in-
creased utilization of HCO
3
–
over CO
2aq
,then
it implies that ECO
2aq
^ became increasingly
limiting to algal growth in both oligotrophic
and eutrophic environments. Although this
would compromise quantitative estimates of
atmospheric pCO
2
, it would still support a
scenario of decreasing pCO
2
with time. Until
evidence emerges to the contrary, we must as-
sume that the physiological processes respon-
sible for e
p37:2
in the past were similar to those
operating in modern surface waters (19, 24)
and use these data to estimate both ECO
2aq
^
and pCO
2
over the past 45 million years.
The conversion of e
p37:2
values to pCO
2
requires an estimate of surface-water EPO
4
3–
^
(39) and temperature for each site. For this
study, we assumed that modern surface-water
distributions of EPO
4
3–
^ at each site between 0
and 100 m encompassed the probable range at
any given time, and we applied temperatures
derived from the oxygen isotope composition
of coeval carbonates in order to convert es-
timates of ECO
2aq
^ to pCO
2
. This approach
assumes relative air-sea equilibrium, which
may not be valid for every site. However, al-
though disequilibrium could lead to overesti-
mates of pCO
2
, our treatment of the data
ultimately yields a range of CO
2
concentra-
tions that reflects the uncertainty associated
with this effect. On a broad scale, our results
indicate that CO
2
concentrations during the
middle to late Eocene ranged between 1000
and 1500 parts per million by volume (ppmv)
(40) and then rapidly decreased during the
Oligocene, reaching modern levels by the lat-
est Oligocene (Fig. 3A). In detail, a trend
toward lower CO
2
concentrations is evident
from the middle to late Eocene, reaching lev-
els by the E/O boundary that could have trig-
gered the rapid expansion of ice on east
Antarctica (2). An episode of higher pCO
2
in
the latest Oligocene occurs concomitantly with
a È2-million-year low in the mean d
18
Ocom-
position of benthic foraminifera (Fig. 3B), in-
dicating that global climate and the carbon
cycle were linked from the Eocene to the late
Oligocene. This association weakens in the
Neogene, when long-term patterns of climate
and pCO
2
appear to be decoupled (17).
In addition to climate, the change in CO
2
implied by our record would have substantial-
ly affected the growth characteristics of ter-
restrial flora. In particular, the expansion of C
4
grasses has received considerable attention as
an indicator of environmental change (41, 42).
The C
4
pathway concentrates CO
2
at the site
of carboxylation and enhances rates of photo-
synthesis by eliminating the effects of photo-
respiration under low CO
2
concentrations (43).
Moreover, higher rates of carbon assimilation
can be maintained under water-stressed condi-
tions. This results in a water-use efficiency
(water loss per unit of carbon assimilated) in
C
4
plants that is twice that of C
3
plants at
È25-C(44). Given our understanding of the
environmental parameters affecting C
3
and C
4
plants, a prevalent supposition has emerged
that C
4
photosynthesis originated as a response
to stresses associated with photorespiration
(41, 45). The CO
2
threshold below which C
4
photosynthesis is favored over C
3
flora is
estimated at È500 ppmv (41), a level that is
breached during the Oligocene. Molecular
phylogenies (46, 47) and isotopic data (48)
place the origin of C
4
grasses before the Mio-
cene between 25 to 32 Ma (46, 47, 49), the
interval when CO
2
concentrations approached
modern levels. This confluence strongly sug-
gests that C
4
photosynthesis evolved as a re-
sponse to increased photorespiration rates
forced by a substantial drop in pCO
2
during
the Oligocene. Near-global expansion of C
4
ecosystems ensued later in the Miocene (41),
possibly driven by drier climates and/or changes
in patterns of precipitation (42).
References and Notes
1. J. C. Zachos, L. D. Stott, K. C. Lohmann, Paleocean-
ography 9, 353 (1994).
2. K. G. Miller, R. G. Fairbanks, G. S. Mountain, Paleocean-
ography 2, 1 (1987).
3. L. D. Stott, J. P. Kennett, N. J. Shackleton, Proc.
Ocean Drilling Program Sci. Results 113, 849 (1990).
4. N. J. Shackleton, J. P. Kennett, Init. Rep. Deep Sea
Drill. Proj. 29, 743 (1975).
5. J. C. Zachos, M. Pagani, L. Sloan, E. Thomas, K. Billups,
Science 292, 686 (2001).
6. S. Bohaty, J. C. Zachos, Geology 31, 1017 (2003).
7. J. V. Browning, K. G. Miller, D. K. Pak, Geology 24,
639 (1996).
8. C. Robert, J. P. Kennett, Geology 25, 587 (1997).
9. J. C. Zachos, B. N. Opdyke, C. N. Quinn, C. E. Jones,
A. N. Halliday, Chem. Geol. 161, 165 (1999).
10. J. C. Zachos, T. M. Quinn, K. A. Salamy, Paleoceanog-
raphy 11, 251 (1996).
11. H. K. Coxall, P. A. Wilson, H. Pa
¨
like, C. H. Le, Nature
433, 53 (2005).
12. M. E. Raymo, Geology 19, 344 (1991).
13. R. A. Berner, Z. Kothavala, Am. J. Sci. 301, 182 (2001).
14. R. M. DeConto, D. Pollard, Nature 421, 245 (2003).
15. P. N. Pearson, M. R. Palmer, Nature 406, 695 (2000).
16. D. L. Royer et al., Nature 292, 2310 (2001).
17. M. Pagani, M. A. Arthur, K. H. Freeman, Paleoceanog-
raphy 14, 273 (1999).
18. M. H. Conte, J. K. Volkman, G. Eglinton, in The Hapto-
phyte Algae,J.C.Green,B.S.C.Leadbeater,Eds.
(Clarendon Press, Oxford, 1994), pp. 351–377.
19. R. R. Bidigare et al., Global Biogeochem. Cycles 11,
279 (1997).
20. R. R. Bidigare et al., Global Biogeochem. Cycles 13,
251 (1999).
21. B. N. Popp et al., Geochim. Cosmochim. Acta 62,69
(1998).
22. U. Riebesell, A. T. Revill, D. G. Hodsworth, J. K. Volkman,
Geochim. Cosmochim. Acta 64, 4179 (2000).
23. E. A. Laws, B. N. Popp, R. R. Bidigare, Geochem.
Geophys. Geosyst. 2, 2000GC000057 (2001).
0
500
1000
1500
2000
2500
10 15 20 25 30 35 40 45 50
5
0
Miocene
Oligocene
Eocene
A
g
e (Ma)
pCO
2
(ppmv)
-1
0
1
2
3
4
5
δ
18
O (‰, PDB)
A
B
Fig. 3. (A) pCO
2
estimates calculated from
e
p37:2
. e
p
0 e
f
– b/[CO
2aq
](39), where b 0
{(118.52[PO
4
3–
]) þ 84.07}/(25 – e
p37:2
), calcu-
lated from the geometric mean regression of all
available data (19, 20, 23, 58, 59). [CO
2aq
]values
were calculated using mean e
p37:2
values and a
range of [PO
4
3–
]valuesforeachsite.[PO
4
3–
]
ranges applied for individual sites were as
follows: site 612, 0.5 to 0.3 mmol/liter; site 516,
0.4to0.2mmol/liter; sites 511 and 513, 1.10 to
0.8 mmol/liter; site 803, 0.3 to 0.1 mmol/liter; and
site 588, 0.3 to 0.2 mmol/liter. Values of CO
2aq
were converted to pCO
2
by applying Henry’s
Law (60), calculated assuming a salinity of 35
and surface-water temperatures derived from
d
18
O of marine carbonates. Maximum pCO
2
estimates were calculated using maximum tem-
peratures (61) for each sample and maximum
[PO
4
3–
] for each site. Intermediate and minimum
(dashed line) pCO
2
estimates were calculated
using intermediate and minimum temperatures
for each sample and minimum [PO
4
3–
]foreach
site. An analytical treatment of error propagation
suggests that relative uncertainties in recon-
structed CO
2
values are È20% for the Miocene
data and approach 30 to 40% for Paleogene
samples with higher (20 to 24°) e
p37:2
values
(62). (B) Global compilation of benthic oxygen
isotope records (5).
R EPORTS
22 JULY 2005 VOL 309 SCIENCE www.sciencemag.org
602

24. M. Pagani, K. H. Freeman, N. Ohkouchi, K. Caldeira,
Paleoceanography 17, 1069 (2002).
25. P. Valentine, Init. Rep. Deep Sea Drill. Proj. 95, 359
(1987).
26. K. G. Miller, W. A. Berggren, J. Zhang, J. A. A. Palmer,
Palaios 6, 17 (1991).
27. W. Wei, S. W. Wise, Mar. Micropaleontol. 14, 119
(1989).
28. C. Pujol, Init. Rep. Deep Sea Drill. Proj. 72, 623 (1983).
29. R. M. Leckie, C. Farnham, M. G. Schmidt, Proc. Ocean
Drill. Program 130, 113 (1993).
30. S. W. Wise Jr., Init. Rep. Deep Sea Drill. Proj. 71, 481
(1983).
31. I. A. Basov, P. F. Ciesielski, V. A. Krasheninnikov, F. M.
Weaver, S. W. Wise Jr., Init. Rep. Deep Sea Drill. Proj.
71, 445 (1983).
32. W. A. Berggren, D. V. Kent, C. C. I. Swisher, M. P.
Aubry, in Geochronology, Time Scales and Global
Stratigraphic Correlation, W. A. Berggren, D. V. Kent,
M. P. Aubry, J. Hardenbol, Eds. (Special Publication
No. 54, Society for Sedimentary Geology, Tulsa, OK,
1995), pp. 129–212.
33. N. J. Shackleton, M. A. Hall, A. Boersma, Init. Rep.
Deep Sea Drill. Proj. 74, 599 (1984).
34. Calculation of e
p37:2
requires knowledge of the d
13
Cof
ambient CO
2aq
(d
13
C
CO
2
aq
) during alkenone produc-
tion and of temperature, which can be approximated
from the d
13
C of shallow-dwelling foraminifera,
assuming isotopic and chemical equilibria among all
the aqueous inorganic carbon species and atmospheric
CO
2
, as well as foraminiferal calcite (17). In this
study, records of planktonic foraminifera coeval with
alkenone measurements were available from sites 511,
513, and 803. Site 612 had well-preserved planktonic
and benthic foraminifera, but some samples lacked
coeval samples of planktonic foraminifera. In these
cases, the isotopic compositions of planktonic foram-
inifera were modeled by calculating the average
difference between benthic and planktonic foraminif-
era and adding this value to the isotopic compositions
of benthic foraminifera. Site 516 had poor carbonate
preservation and lacked an adequate foraminiferal
record. For this site, surface d
13
C
CO
2
aq
and values were
modeled from the d
13
C compositions of the G60-mm
fine fraction (FF), assuming an isotopic offset between
the FF and shallow-dwelling foraminifera of þ0.5°,as
indicated by Miocene (50) and Eocene (this study)
records from this site. Similarly, surface-water temper-
atures, required in the calculation of both d
13
C
CO
2
aq
and pCO
2
, were estimated from the d
18
O compositions
of shallow-dwelling planktonic foraminifera or modeled
from the d
18
O compositions of the G60 mm FF, assum-
ing an isotopic offset between the FF and shallow-
dwelling foraminifera of –1.5° (50).
35. B. Rost, U. Riebesell, S. Burkhardt, Limnol. Oceanogr.
48, 55 (2003).
36. J. Backman, J. O. R. Hermelin, Palaeogeogr. Palaeo-
climatol. Palaeoecol. 57, 103 (1986).
37. J. Young, J. Micropalaeontol. 9, 71 (1990).
38. J. Young, personal communication.
39. e
p37:2
is related to [CO
2aq
] by the expression e
p
0 e
f
–
b/[CO
2aq
], where e
f
represents the carbon isotope frac-
tionation due to carboxylation. b represents the sum
of physiological factors, such as growth rate and cell
geometry, affecting the total carbon isotope dis-
crimination. In the modern ocean, b is highly cor-
related to surface-water [PO
4
3–
](19). However, it is
unlikely that [PO
4
3–
] alone is responsible for the
variability in growth rate inferred from variation in b.
Instead, [PO
4
3–
] may represent a proxy for other
growth-limiting nutrients, such as specific trace ele-
ments that exhibit phosphate-like distributions.
40. For comparison with our record, middle to late Eocene-
age estimates of ocean pH using the boron isotopic
compositions of foraminifera (15) yield early Eocene
CO
2
concentrations that are potentially 10 times high-
er than preindustrial levels (È3500 ppmv), reaching
levels as low as È350 ppmv during the middle to late
Eocene. Our Eocene estimates do not support a sce-
nario of low pCO
2
during this time.
41. T. E. Cerling et al., Nature 389, 153 (1997).
42. M. Pagani, K. H. Freeman, M. A. Arthur, Science 285,
876 (1999).
43. R. W. Pearcy, J. Ehleringer, Plant Popul. Biol. 7, 1 (1984).
44. M. D. Hatch, Biochim. Biophys. Acta 895, 81 (1987).
45. J. R. Ehleringer, R. F. Sage, L. B. Flanagan, R. W. Pearcy,
Trends Ecol. Evol. 6,95(1991).
46. B. S. Gaut, J. F. Doebley, Proc. Natl. Acad. Sci. U.S.A.
94, 6809 (1997).
47. E. A. Kellogg, in C
4
Plant Biology, R. F. Sage, R. K.
Monson, Eds. (Academic Press, New York, 1999), pp.
313–371.
48. D. L. Fox, P. L. Koch, Geology 31, 809 (2003).
49. R. E. Sage, New Phytol. 161, 341 (2004).
50. A. Ennyu, M. A. Arthur, M. Pagani, Mar. Micro-
paleontol. 46, 317 (2002).
51. D. P. Schrag, Chem. Geol. 161, 215 (1999).
52. P. N. Pearson et al., Nature 413, 481 (2001).
53. M. Pagani, M. A. Arthur, K. H. Freeman, Paleoceanog-
raphy 15, 486 (2000).
54. B. N. Popp, F. Kenig, S. G. Wakeham, E. A. Laws, R. R.
Bidigare, Paleoceanography 13, 35 (1998).
55. W. G. Mook, J. C. Bommerson, W. H. Staberman,
Earth Planet. Sci. Lett. 22, 169 (1974).
56. C. S. Romanek, E. L. Grossman, J. W. Morse, Geochim.
Cosmochim. Acta 56, 419 (1992).
57. J. Erez, B. Luz, Geochim. Cosmochim. Acta 47, 1025
(1983).
58. B. N. Popp et al.,inReconstructing Ocean History: A
Window into the Future, F. Abrantes, A. Mix, Eds.
(Plenum, New York, 1999), pp. 381–398.
59. M. E. Eek, M. J. Whiticar, J. K. B. Bishops, C. S. Wong,
Deep-Sea Res. II 46, 2863 (1999).
60. R. F. Weiss, Mar. Chem. 2, 203 (1974).
61. All carbonates are assumed to be diagenetically altered
to some degree, which acts to increase their d
18
O
composition (51, 52), yielding minimum temperatures.
In order to compensate for this uncertainty, three
temperature estimates were used in the calculation of
e
p37:2
and pCO
2
, reflecting minimum temperatures
calculated directly from the d
18
O value of carbonates
(Temp
min
), intermediate temperatures (Temp
min
þ 3-C),
and maximum temperatures (Temp
min
þ 6-C).
62. K. H. Freeman, M. Pagani, in A History of Atmospheric
CO
2
and its Effects on Plants, Animals, and Ecosystems,
J. R. Ehleringer, T. E. Cerling, M. D. Dearing, Eds.
(Springer, New York, 2005), pp. 35–61.
63. The authors thank two anonymous reviewers who
helped improve the quality of the manuscript. We also
thank B. Berner and K. Turekian for coffee and ani-
mated conversations that helped develop and inspire
ideas. This work was funded by a grant from NSF.
21 January 2005; accepted 7 June 2005
Published online 16 June 2005;
10.1126/science.1110063
Include this information when citing this paper.
Global Mammal Conservation:
What Must We Manage?
Gerardo Ceballos,
1
*
Paul R. Ehrlich,
2
Jorge Sobero
´
n,
3
.
Irma Salazar,
1
John P. Fay
2
We present a global conservation analysis for an entire ‘‘flagship’’ taxon, land
mammals. A combination of rarity, anthropogenic impacts, and political
endemism has put about a quarter of terrestrial mammal species, and a larger
fraction of their populations, at risk of extinction. A new global database and
complementarity analysis for selecting priority areas for conservation shows
that È11% of Earth’s land surface should be managed for conservation to
preserve at least 10% of terrestrial mammal geographic ranges. Different ap-
proaches, from protection (or establishment) of reserves to countryside bio-
geographic enhancement of human-dominated landscapes, will be required to
approach this minimal goal.
Research on population and species extinctions
shows an accelerating decay of contemporary
biodiversity. This pressing environmental prob-
lem is likely to become even worse in coming
decades (1–3). Although impacts of human
activities are global in scope, they are not uni-
formly distributed. The biota of certain coun-
tries and regions can be identified as being
most at risk, having both exceptionally high
richness and endemism and exceptionally rapid
rates of anthropogenic change. Because re-
sources for conservation are limited, ecolo-
gists must provide managers and politicians
with solid bases for establishing conserva-
tion priorities (4) to minimize population and
species extinctions (5), to reduce conservation
conflicts (6, 7), and to preserve ecosystem
services (8).
Even for charismatic taxa, we lack a
global view of patterns of species distribu-
tions useful for establishing conservation
priorities. Such a view would allow evalua-
tion of the effort required, for example, to
preserve all species in a given taxon. It would
also be relevant to setting global conservation
goals such as protecting a certain percentage
of Earth_s land surface (9). More restricted
approaches such as identifying hot spots
and endemic bird areas have called attention
to relatively small areas where large num-
bers of species might be protected (10–13).
For instance, recently the number of verte-
brate species that lack populations within
major protected areas was estimated (12).
But now more comprehensive analyses are
possible.
1
Instituto de Ecologı
´
a, UNAM, Apdo. Postal 70-275,
Me
´
xico D.F. 04510, Me
´
xico.
2
Center for Conservation
Biology, Department of Biological Sciences, Stanford
University, Stanford, CA 94305–5020, USA.
3
Comisio
´
n
Nacional de Biodiversidad, Periferico-Insurgentes 4903,
Mexico.
*To whom correspondence should be addressed.
E-mail: gceballo@miranda.ecologia.unam.mx
.Present address: Natural History Museum, Dyke
Hall, University of Kansas, Lawrence, KS 66045, USA.
R EPORTS
www.sciencemag.org SCIENCE VOL 309 22 JULY 2005
603