ogh
Ins
ta,
cia,
Review(or perhaps even greater) threat to the biology of our
planet.
A large proportion of the CO
2
emitted by the activities
of humans is taken up by the oceans. This uptake
accounts for 50% of the CO
2
released from the combus-
tion of fossil fuels and cement production, or 30% if land-
use practices such as deforestation are also included [1].
As CO
2
dissolves in the oceans, it takes part in a series
of reactions leading to a drop in pH (Box 1). This change
in seawater chemistry affects marine organisms and
ecosystems in several ways (Boxes 2 and 3), with the
clearest impacts being felt by organisms that produce
shells and skeletons composed of calcium carbonate.
Due to the CO
2
-induced decrease in concentration of
carbonate ions in seawater, the saturation state of the
seawater pH, and the decreasing trend in ocean pH will be
more pronounced in some areas than in others. Higher
latitudes, for instance, will experience the greatest change,
whereas the tropics will apparently experience less change
[10]. Consequently, conditions are likely to become very
hostile for calcifying species in the northern high latitude
regions over the next decade [10] and in the Southern
Ocean over the next few decades [11,12]. This does not,
however, militate against large-scale changes which are
occurring (and will continue to occur) in tropical regions
[3–6].
Understanding the implications of these changes in
seawater chemistry for marine organisms and ecosystems
is in its infancy. However, interest in this issue has
deepened in the past five years, with a growing number
of researchers investigating the widespread implications
and potentially catastrophic consequences of oceanCorresponding author: Pelejero, C. (pelejero@icm.cat).Paleo-perspectives
acidification
Carles Pelejero
1
, Eva Calvo
2
and Ove Hoe
1
Institucio´ Catalana de Recerca i Estudis Avanc¸ats (ICREA) and
37-49, E-08003, Barcelona, Catalonia, Spain
2
Institut de Cie`ncies del Mar, CSIC, Pg. Marı´tim de la Barcelone
3
Global Change Institute, The University of Queensland, St. Lu
The anthropogenic rise in atmospheric CO
2
is driving
fundamental and unprecedented changes in the chem-
istry of the oceans. This has led to changes in the
physiology of a wide variety of marine organisms and,
consequently, the ecology of the ocean. This review
explores recent advances in our understanding of ocean
acidification with a particular emphasis on past changes
to ocean chemistry and what they can tell us about
present and future changes. We argue that ocean con-
ditions are already more extreme than those experi-
enced by marine organisms and ecosystems for
millions of years, emphasising the urgent need to adopt
policies that drastically reduce CO
2
emissions.
Ocean acidification: the ‘evil twin’ of global warming
One of the major environmental challenges facing society
is the impact of increased levels of atmospheric CO
2
and
other greenhouse gases on the physical and biological
systems on earth. These increased levels are mainly due
to the combustion of fossil fuels and changes in land use
and deforestation. So far, most research has focused on
the impacts arising from global warming which has dri-
ven (and is continuing to drive) large fundamental
changes in biological systems. However, the steady acid-
ification of the oceans (nicknamed the ‘evil twin’ of global
warming) is another insidious consequence of rising
levels of atmospheric CO
2
. Although it will be hard to
quantify the effects separately, and indeed their syner-
gistic behaviour, evidence gathered over the last years
suggests that ocean acidification could represent an equal332 0169-5347/$ – see front matter  2010 Elsevin ocean
-Guldberg
3
titut de Cie`ncies del Mar, CSIC, Pg. Marı´tim de la Barceloneta,
37-49, E-08003, Barcelona, Catalonia, Spain
Queensland, QLD 4072, Australia
precipitated calcium carbonate decreases progressively
as pH decreases, making calcification more difficult (Box
1). In addition, ocean acidification can also disrupt pH-
sensitive physiological processes such as gas exchange
and reproduction ([2] and references therein). Reduced
pH will also alter the chemistry of nutrients and the
chemical form of metals in seawater, which might
enhance their role as micronutrients (e.g. iron) or their
toxic potential (e.g. copper and zinc). On a broader scale,
changes to the physiology of keystone organisms such as
pteropods, coccolithophorids, foraminifera and corals
could lead to vast changes to fundamentally important
ecosystems and/or food webs, with consequences for the
carbon cycle and the exchange of gases between the ocean
and the atmosphere [2–8].
The surface waters of the oceans have already acidified
by an average of 0.1 pH units from pre-industrial levels [2],
with future changes depending ultimately on the rate at
which CO
2
is emitted by the activities of humans. Surface-
water pH reductions could range from 0.2 units if 1,200 GT
C(1GT=10
15
g) are released over 1000 years to nearly 0.8
units if 5,000 GT C are released over 200 years [9]. By the
end of the twenty-first century, projections based on differ-
ent scenarios indicate that ocean pH will have decreased by
0.3 to 0.4 pH units ([10], Box 1). As this review will
show, current conditions are probably more extreme than
those experienced by the oceans over a timescale of
millions of years, and future conditions will certainly be
more extreme. However, an important degree of spatial
(Box 4) and temporal (Figure 1) heterogeneity exists iner Ltd. All rights reserved. doi:10.1016/j.tree.2010.02.002 Available online 30 March 2010

Box 1. Basics of chemistry changes in seawater from the marine absorption of CO
2
.
As CO
2
dissolves in seawater, it participates in a series of chemical
equilibrium reactions ((1) in Figure Ia) which result in increased
concentrations of bicarbonate and hydrogen ions (protons) and
therefore a decrease in seawater pH (Figure Ib). This also leads to a
decrease in the concentration of carbonate ions (Figure Ib), with
implications for marine organisms that need them as building blocks
in the construction of their calcium carbonate skeletons or shells. For
most of these organisms, experimental studies have demonstrated
that this change in seawater chemistry makes calcification more
difficult (Boxes 2 and 3).
Marine organisms can precipitate calcium carbonate in the form of
calcite (e.g. coccolithophorids, foraminifera), aragonite (e.g. corals,
pteropods), or high-magnesium calcite (e.g. crustose coralline algae).
These crystal structures have different degrees of stability. Calcite is
thermodynamically the most stable, followed by aragonite and finally
by high-magnesium calcite, which is the least stable. The tendency for
a structure to dissolve is strongly influenced by the saturation state
(V) of each particular mineral phase, which is related to the
concentration of calcium and carbonate ions in the seawater
(equation (2) in Figure Ia). Over geologically short timescales (<1
million years), the concentration of calcium ions in seawater does not
vary considerably, so a decrease in carbonate ion concentration due
to ocean acidification will reduce V. Precipitation of calcium
carbonate is thermodynamically favourable where V > 1(super-
saturation) and unfavourable where V < 1 (under-saturation). How-
ever, the V threshold for biogenic precipitation depends on each
marine species. Coral reef communities and structures, for example,
develop properly in seawaters where V values of aragonite are >3.3
[5,83] because the calcification process has to exceed rates of bio-
erosion.
As oceans turn more acidic, the saturation state of aragonite
progressively decreases, reducing the effective area where coral reefs
can develop (Figure Ic). Evidence for these declining trends in V has
recently been reported from the Caribbean [85]. The chemical reaction
(3) in Figure Ia is one of the ways to represent the precipitation of
calcium carbonate, and illustrates the counter-intuitive fact that, when
marine organisms calcify, there is a release of CO
2
in the surrounding
water. However, the amount of CO
2
released per mol of calcium
carbonate precipitated depends on a range of factors including
temperature, salinity and pCO
2
[86]; in today’s oceans, the CO
2
released per mol of calcium carbonate precipitated is of 0.6 mol, but
this figure will increase as the atmospheric concentration of CO
2
rises.
Figure I. Chemistry of ocean acidification and global changes in aragonite
saturation. (a) Schematic view of the anthropogenic perturbation of the carbon
cycle in which part of the CO
2
emitted from the combustion of fossil fuels and
deforestation is absorbed by the oceans. (1) Chemical equilibrium reactions in
which CO
2
intervenes as it dissolves in seawater, where K
1
and K
2
are the
dissociation constants for carbonic acid and bicarbonate ion, respectively. (2)
Definition of the saturation state (V), where K
sp
is the solubility constant of each
mineral phase. (3) Calcium carbonate precipitation reaction that illustrates the
release of CO
2
as precipitation occurs. (b) Near past (1800) and near future (2100)
evolution of surface pH (orange, on total scale) and atmospheric CO
2
(magenta)
from [10] based on prescribed fossil- fuel and land- use CO
2
emissions from
historical data for the period 1820 to 2000 and considering the A2 IPCC SRES
emissions scenario afterwards. Evolution of bicarbonate (green) and carbonate
ion (blue) concentrations, computed from dissolved inorganic carbon and pH
data from [10]. These calculations were done using the CO2SYS.XLS program
(Pelletier et al., available at www.ecy.wa.gov/programs/eap/models.html). The
grey cloud of points shows all 1  1 degree mixed surface layer (upper 50 m) pH
values in the oceans, computed from total carbon and alkalinity data (see Box 4
for more details). (c) Reconstructions for the past and present, and predictions
for the future changes in the aragonite saturation state (V, equation (2) in Figure
Ia) at different levels of atmospheric CO
2
concentrations (white numbers in
ppmv; reprinted with permission from [5]). Note the dramatic reduction in blue
areas, which are representative of the approximate minimum V
aragonite
values
needed for carbonate coral reef communities to develop properly [83].
Review
Trends in Ecology and Evolution Vol.25 No.6
333