Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 23
2
Effects of
Ocean Acidification
on the
Chemistry of Seawater
As atmospheric carbon dioxide (CO2) increases and dissolves into
the ocean, it modifies the chemistry of seawater. This chapter reviews the
current knowledge regarding the chemical changes brought about by
the increasing CO2--labeled collectively as ocean acidification--in the
past, the present, and the future. It first discusses the principal processes
that control the acidbase chemistry of seawater and the cycling of carbon
in the ocean. The chapter then examines how these processes are modified
by increasing CO2 concentrations. Most of these processes are well under
stood and the uncertainties have to do chiefly with the extent and the
timing of the chemical changes, not their nature. Next, previous instances
of acidification in the distant past are reviewed and their relevance to the
current situation are discussed. Finally, the chapter briefly touches on
efforts to mitigate or geoengineer solutions to climate change, and how
these efforts are related to ocean acidification. Additional detailed discus
sions of chemical changes related to acidification can be found in Zeebe
and WolfGladrow (2001) and Millero (2006).
2.1 SEAWATER CHEMISTRy
The principal weak acids and bases that can exchange hydrogen ion
in seawater and are thus responsible for controlling its pH are inorganic
carbon species and, to a lesser extent, borate. Inorganic carbon dissolved
in the ocean occurs in three principal forms: dissolved carbon dioxide
OCR for page 24
OCEANACIDIFICATION
(CO2.aq),1 bicarbonate ion (HCO3), and carbonate ion (CO32) (see Box 2.1
for definitions.). CO2 dissolved in seawater acts as an acid and provides
hydrogen ions (H+) to any added base to form bicarbonate:
CO2 (aq) + H2O H+ + HCO3 (1)
CO32 acts as a base and takes up H+ from any added acid to also form
bicarbonate:
H+ + CO32 HCO3 (2)
Borate [B(OH)4] also acts as a base to take up H+ from any acid to form
boric acid [B(OH)3]:
H+ + B(OH)4 B(OH)3 + H2O (3)
As seen in reactions 1 and 2, bicarbonate can act as an acid or a base (i.e.,
donate or accept hydrogen ions) depending on conditions.
Under presentday conditions, these reactions buffer the pH of sur
face seawater at a slightly basic value of about 8.1 (above the neutral value
around 7.0). At this pH, the total dissolved inorganic carbon (DIC ~ 2 mM)
consists of approximately 1% CO2, 90% HCO3, and 9% CO32 (Figure 2.1).
The total boric acid concentration (B(OH)4+ B(OH)3)) is about 1/5 that of
DIC. As discussed in section 2.2, increases in CO2 will increase the H+ con
centration, thus decreasing pH; the opposite occurs when CO2 decreases.
We note that isotope fractionation between B(OH)3 and B(OH)4 is used
for estimating past pH values (Box 2.2).
Life in the oceans modifies the amount and forms (or species) of
inorganic carbon and hence the acidbase chemistry of seawater. In the
sunlit surface layer, phytoplankton convert, or "fix," CO2 into organic
matter during the day--a process also known as photosynthesis or primary
production. This process simultaneously decreases DIC and increases the
pH. The reverse occurs at night, when a portion of this organic matter is
decomposed by a variety of organisms that regenerate CO2, resulting in
a daily cycle of pH in surface waters. A fraction of the particulate organic
matter sinks below the surface where it is also decomposed, causing verti
cal variations in the concentrations of inorganic carbon species and pH.
The net result is a characteristic maximum in CO2 concentration and
minima in pH and CO32 concentration around 500 to 1,000 meters depth
1The proper notation for carbon dioxide gas is CO2.g; carbon dioxide dissolved in water is
CO2.aq. However, for simplicity, these notations are not carried through the report; the text
provides adequate context to determine which form of CO2 is being discussed.
OCR for page 25
CHEMISTRYOFSEAWATER
BOX 2.1
Parameters of the Ocean Acid-base System
DIC = Dissolved Inorganic Carbon concentration
DIC = [CO2] + [HCO3] + [CO32]
Where the brackets indicate concentrations in mol/Kg.
pCO2 = partial pressure of CO2 (in ppm or µatm)
pCO2 = [CO2]/KH
Where KH is the solubility constant for CO2 in seawater (which varies with tempera-
ture, pressure and salinity)
Total Boric Acid = [B(OH)3 ]+ [B(OH)4]
TA = Total Alkalinity
TA = [HCO3] + 2[CO32] + [B(OH)4] + other minor bases
pH log10 [H+]
More formally, oceanographers use two different pH scales, the total and the
seawater pH scales:
pHT = log{[H+] + [HSO4]}
pHSWS = log{[H+] + [HSO4] + [HF]}
These two scales differ by about 0.01 units for a salinity S = 35 and temperature
T= 25°C.
in many areas of the open ocean as illustrated in Figure 2.2a. Because the
intensities of biological processes vary with season and the solubility of
CO2 varies with temperature, the pH and the concentrations of inorganic
carbon species exhibit cyclical seasonal variations. For reasons discussed
below, the vertical distribution of pH in the ocean varies with geographi
cal location, particularly as a function of latitude; this is illustrated in the
NorthSouth transect for the Pacific Ocean in Figure 2.2b.
Another important process affecting the acidbase chemistry of sea
water is the production of calcium carbonate (CaCO3). Marine life pro
duces the vast majority of CaCO3 in the ocean; mostly in the form of the
minerals calcite and aragonite (see Box 2.3). Even though these minerals
are supersaturated in surface seawater, they do not normally precipitate
spontaneously, but are formed by various organisms to serve as skel
etons or hard protective structures. The degree of supersaturation of these
minerals, quantified by the parameter (see Box 2.3), varies with tem
perature, depth and seawater inorganic carbon chemistry; is generally
highest in shallow, warm waters and lowest in cold waters and at depth
(Feely et al., 2004). When calcium carbonate sinks in the water column, it
OCR for page 26
OCEANACIDIFICATION
FIgURE 2.1 Typical concentrations of the major weak acids and weak bases in
seawater as a function of pH. This diagram is calculated for constant dissolved
inorganic carbon (DIC) and constant total boric acid using constants from Dickson
et al. (2007) and Lueker et al. (2000).
Figure 2-1
R01733
uneditable bitmapped image
BOX 2.2
Boron Isotopes as a Paleo-proxy for Seawater pH
Changes in ocean pH can be documented beyond the instrumental period
of direct measurements using a proxy based on the incorporation into CaCO3 of
the borate ion, B(OH)4 which has a lighter isotope composition than boric acid,
B(OH)3 (Spivack et al., 1993; Sanyal et al., 1995). For time scales shorter than
the residence time of boron in the ocean--5-10 million years--measured values in
sedimentary carbonates appear to accurately reflect the pH of the growth medium
for several calcifying taxa. Results from glacial-interglacial times generally reflect
the pH-buffering effect of the CaCO3 cycle (Hönisch, 2005), while records from
more recent time intervals reflect acidification of the ocean from rising CO2 con-
centrations over the past centuries (Liu, 2009).
OCR for page 27
CHEMISTRYOFSEAWATER
A
B
Figure 2-2a
R01733
uneditable bitmapped image
FIgURE 2.2 Inorganic carbon and pH vary as a function of depth and latitude.
(a) Vertical profiles typical of the midNorth Pacific showing variations of several
seawater chemical parameters with depth. Adapted from Morel and Hering (1993)
with calculations using constants from Figure 2-2b et al. (2007) and Lueker et al.
Dickson
R01733
(2000). (b) Typical distribution of pH with depth along a NorthSouth transect for
the Pacific Ocean. (Byrne uneditable
et al., 2010a). bitmapped image
OCR for page 28
OCEANACIDIFICATION
BOX 2.3
Calcium Carbonate Solubility
Many marine organisms deposit calcareous shells and skeletons made of
calcium carbonate (CaCO3), which is a soluble mineral (Sanyal et al., 1995). The
solubility of minerals such as CaCO3 varies depending upon the physical proper-
ties of the seawater (e.g., temperature, salinity, and pressure) and also the crystal
form of the mineral. The solubility is often expressed as the saturation state () of
a mineral: when >1, seawater is supersaturated with respect to CaCO3 and it will
remain solid; when <1, seawater is undersaturated and CaCO3 structures may
begin to dissolve, unless they are protected from dissolution (e.g., with an organic
coating). The saturation state is defined as follows:
2+ 2-
Ca sw × CO 3 sw
= 2+ 2-
Ca sat × CO 3 sat
The denominator refers to the stoichiometric solubility product (often desig-
nated as Ksp) of the Ca2+ and CO32 concentrations in a solution saturated with
respect to the given mineral, and the numerator is the product of the in situ con-
centrations. Under current pH conditions, CaCO3 is supersaturated in most surface
ocean waters. Calcium ion concentration varies little in the open ocean, but ocean
acidification decreases the concentration of CO32 and the degree of supersatura-
tion. In estuarine waters both Ca2+ and CO32 concentrations vary widely and can
frequently be below saturation.
Most calcium carbonate is precipitated by organisms in one of two forms:
calcite (which has a rhombohedral crystal structure) and aragonite (which is ortho-
rhombic). Vaterite, a third form, is rare but of interest because it is involved in
the early stages of calcite precipitation in some organisms and is highly soluble.
Normally, aragonite is about 1.5 times more soluble in seawater than calcite. How-
ever, the calcite crystal structure allows some ionic substitution of magnesium (Mg)
for calcium: calcite with > 4 mol% MgCO3 is called "high-Mg calcite" and is usually
more soluble than regular calcite.
FROM: Morse and Mackenzie, 1990 and Morse et al., 2006.
becomes less stable ( decreases) as a result of the decrease in CO32 con
centration and the increase in the solubility of the minerals caused by the
higher pressure and the lower temperature. The depth at which CaCO3
becomes undersaturated and begins to dissolve depends on its crystalline
form; this "saturation horizon" for calcite is deeper than that for aragonite
(see Box 2.3). Precipitation of CaCO3 at the surface lowers the ambient
pH, while its dissolution at depth increases it, partially compensating for
the inverse effects of the photosynthetic reduction of CO2 that raises pH
in surface waters and lowers pH in deeper waters as CO2 is regenerated
by metabolic oxidation.
OCR for page 29
CHEMISTRYOFSEAWATER
As illustrated in Figure 2.2b, the vertical distribution of pH is not
uniform throughout the oceans. The principal cause of these geographical
pH variations is the nonuniform distribution of the CO2 concentration
resulting from the lower solubility of CO2 gas at higher temperatures,
basinwide patterns of subsurface biological oxidation of organic matter
and dissolution of carbonate minerals, and upwelling of CO2rich deep
water or downwelling of CO2poor surface water (Sarmiento and Gruber,
2006). This is illustrated in Figure 2.3 (Part A) which shows CO 2 concen
tration as a function of depth in a NorthSouth transect across the North
Pacific Ocean. Upwelling around the equator increases CO2 concentration
near the surface at low latitudes compared to values in mid latitudes. An
increase in surface CO2 is also seen at high latitudes caused by the high
solubility of CO2 in cold water. High concentrations in deeper water result
from oxidation of organic matter. These geographical patterns in CO2 con
pCO2 pCO2 (µatm) P16N 2006
0
1350
1200 200
Pr essur e (db)
1050
400
900
750 600
600
800
450
A 300 1000
ARAG Ar agonite P16N 2006
4 0
3.5
200
3
Pr essur e (db)
2.5 400
2
1.5 600
1
800
0.5
B 0 1000
0° 10°N 20°N 30°N 40°N 50°N
L atitude
FIgURE 2.3 The distribution of (a) pCO2 and (b) aragonite saturation in the North
Pacific Ocean during a transect in March 2006. A pressure of 1 decibar (1 db on the
y axis) corresponds approximately to aFigure
depth of2-3 new (m). (Fabry et al., 2008b)
1 meter
OCR for page 30
0 OCEANACIDIFICATION
centration are reflected in consistent patterns of CO32 concentrations and
thus also in the degree of saturation () of CaCO3 minerals (see Figure 2.3,
Part B) and in the buffering capacity of the water (Egleston et al., 2010).
2.2 ANTHROPOgENIC CARBON DIOxIDE
EMISSIONS AND OCEAN ACIDIFICATION
The exchange of CO2 at the airwater interface is relatively fast, tak
ing place on a time scale of months to a year so that, on average, the
concentration of CO2 in surface seawater remains approximately at equi
librium with that of the atmosphere. As the concentration of atmospheric
CO2 gas increases year after year, some of it dissolves into the ocean
such that about a third of the total CO2 added to the atmosphere from
anthropogenic sources--including fossil fuel emissions, cement produc
tion and deforestation--over the past 150 years is now dissolved in the
oceans (Sabine et al., 2004; Khatiwala et al., 2009). The increase in dis
solved CO2 concentration decreases the pH and shifts the equilibrium of
inorganic carbon species in seawater, resulting in an increase in CO 2 and
HCO3 concentrations and a decrease in CO32 concentration (Figure 2.4).
For example, under present conditions in the mid North Pacific, for every
100 molecules of CO2 dissolved from the atmosphere, about 7 remain
as CO2, 15 react with B(OH)4, and 78 react with CO32, resulting in an
increase of HCO3 by 171 molecules. The buffering capacity of seawater--
the ability to resist changes in acidbase chemistry upon addition of an
acid such as CO2--depends on the concentration of bases, principally
CO32 and B(OH)4, to neutralize the acid (Figures 2.1 and 2.4). Upon acid
ification of the oceans, the buffering capacity of seawater will decrease
along with pH. Also, ocean water masses that are presently already high
in CO2 for any reason are less buffered against further increases in CO2
than those with lower CO2 (Egleston et al., 2010).
The decrease in carbonate ion concentration, CO32, that results from
ocean acidification will lead to reduced rates of calcification, along with
the a shoaling of the saturation horizons for calcium carbonate minerals
to shallower depths, and a change in the marine calcium carbonate cycle.
The resulting overall decrease in CaCO3 precipitation and burial will tend
to raise seawater pH, favoring the oceanic uptake of CO2, and providing
a small negative feedback on rising atmospheric CO2 and global warming
(Heinze, 2004). The extent of this feedback depends in part on the rela
tive contributions of calcite and aragonite, and hence of the organisms
that produce them, to the CaCO3 cycle. Model simulations (Gehlen et al.,
2007) show that an approximately 30% reduction in CaCO3 production
(which was hypothesized to occur when atmospheric CO2 reached 4x
preindustrial values) leads to an additional cumulative oceanic uptake
OCR for page 31
CHEMISTRYOFSEAWATER
(a.)
(b.)
FIgURE 2.4 Schematic (a) and calculations (b) showing the effect of increasing
CO2 concentration on acidbase species in seawater. Calculations are made for
constant alkalinity using constants from Dickson et al. (2007) and Lueker et al.
(2000). Note that the yaxis is on log scale.
Figure 2-4
R01733
uneditable bitmapped images (a and b)
OCR for page 32
OCEANACIDIFICATION
of ~6 petagrams (Pg) C, small relative to anthropogenic emissions and
other potential climatecarbon cycle feedbacks (Friedlingstein et al., 2006).
The reduction in carbonate production and its faster dissolution rate in
the water column could also decrease the ballasting of organic carbon by
CaCO3 that increases the sinking of organic carbon to the deep ocean (e.g.,
Armstrong et al., 2002; Klaas and Archer, 2002). This would cause more
organic carbon to decompose in shallow water and partially offset the
negative CO2 feedback resulting from lower calcification rates (Heinze,
2004). This effect could be enhanced by an increase in phytoplankton
production of extracellular organic carbon (see chapters 3 and 4) and
by the accelerated bacterial decomposition of organic matter at higher
temperature.
A decrease in seawater pH results in a readjustment of all minor acid
base species, in addition to inorganic carbon and borate. These include
a myriad of trace organic compounds, inorganic species such as the
hydroxyl ion, phosphate and ammonium, and trace metals bound to inor
ganic or organic compounds. The effect of pH on these chemical species
is of interest because several are important nutrients for phytoplankton
growth and the chemical forms affect availability for phytoplankton use.
For example, iron (Fe) is the most important trace nutrient for marine
phytoplankton and inorganic Fe compounds are more biologically avail
able than organicallybound Fe; acidification may cause Fe to become less
bioavailable because as the pH decreases, more Fe will become organically
bound (Shi et al., 2010). The effect of decreasing pH on Fe bioavailability
in surface water is further complicated by the lightinduced cycle between
oxidized and reduced Fe species, in which a key process--oxidation of
reduced Fe--slows down at lower pH. Such effects of acidification on the
chemistry and bioavailability of trace metals and other compounds in the
ocean have barely been studied at all and, unlike the changes in inorganic
carbon species, cannot be predicted with confidence.
In addition, recent studies have shown that ocean acidification can
affect the physical properties of seawater. At low frequencies, sound trans
mission in the ocean is attenuated by volume changes related to acidbase
equilibrium of some chemical species. Change in the proportions of such
systems, notably the boric acid and borate ion acidbase pair, may thus
result in a "noisier ocean" (Hester et al., 2008; Duda, 2009).
2.2.1 Projections for Surface Waters
Because the relationship between atmospheric CO2 and seawater car
bonate chemistry is well understood, it is a simple matter to calculate the
variations in average pH and inorganic carbon species concentrations in
OCR for page 33
CHEMISTRYOFSEAWATER
the surface waters of the open ocean based on the known variations
in atmospheric CO2 over the past 150 years (from actual measurements
or from ice core data). Independent estimation of past seawater pH have
been made using boron isotopes as well (see Box 2.2). Similarly, projec
tions for changes in seawater chemistry can be made for the future on
the basis of any future CO2 emission scenario such as those published by
the IPCC. Such calculations are shown in Figure 2.5 for the Pacific Ocean;
models show that, based on a "businessasusual" scenario of CO2 emis
sions, the surface ocean pH will decrease by about 0.3 units within the
next 100150 years (e.g., WolfGladrow et al., 1999; Caldeira and Wickett,
2003; Feely et al., 2004).
Figure 2.6 shows the results of actual measurements of surface sea
water chemistry at a station near Hawaii between 1998 and 2008. These
data confirm the validity of the calculations and demonstrate the pre
dicted trend of a decrease of about 0.0015 pH units per year. The data also
illustrate the seasonal cycle in pH and inorganic carbon species caused
Expected Changes in the CO2 System
8.4 2350 260
2000 240
2300
8.2 220
2250
200
pCO2 (µ atm)
1500 8.0
2200 180
DIC (µM)
pCO2
pH 160
CO3 (µM)
pH
DIC 7.8 2150
1000 140
CO32-
2-
2100 120
7.6
100
2050
500
7.4 80
2000
60
0 7.2 1950 40
1800 1900 2000 2100 2200
Year
FIgURE 2.5 Projected changes in the pH, and the concentrations of CO2 and
CO32 in surface seawater under a business as usual scenario for CO2 emissions
over the next two centuries. Calculations were made for a salinity of 35 and tem
perature of 25°C assuming constant alkalinity using the CO2sys program (Lewis
and Wallace, 1998). The projected future values of pCO2 in the atmosphere are
based on the estimates of Caldeira and Wickett (2003).
Figure 2-5
R01733
editable vectors
OCR for page 34
OCEANACIDIFICATION
by variations in biological activity discussed above. Because the buffer
ing capacity of seawater decreases with decreasing pH, it is expected that
these seasonal variations will amplify in the future.
2.2.2 Projections for Deeper Waters
While the CO2 concentration in the surface ocean tracks the increas
ing values in the atmosphere, the penetration of that CO2 into deep water
depends on the slow vertical mixing of the water column and the trans
port of water masses in the complex winddriven circulation and over
turning of the oceans (Sarmiento and Gruber, 2006). About half of the
anthropogenic CO2 is now found in the upper 400 meters, while the other
half has penetrated to deeper water, as illustrated in Figure 2.7 (Feely et
al., 2004). This slow penetration of CO2 into the deep ocean is reflected
in a slower decrease in pH at depth than at the surface. An illustration
of the time lag between surface and deep ocean acidification is shown in
Figure 2.8; according to these simple calculations, under a "businessas
usual" scenario of CO2 emissions, it will take about 500 years longer for
a 0.3 unit decrease to occur in deep waters compared to surface waters
(Caldeira and Wickett, 2003). However, in some regions where the vertical
movement of water is relatively fast, the time scale for deep penetration
of anthropogenic CO2 will be on the order of decades instead of centuries
(Sabine et al., 2004).
FIgURE 2.6 [next page] Timeseries of mean carbonic acid system measurements
within selected depth layers at Station ALOHA, 19882007. (First image) Partial
pressure of CO2 in seawater calculated from DIC and TA (blue symbols) and in
watersaturated air at in situ seawater temperature (red symbols). Linear regres
sions of the sea and air pCO2 values are represented by solid and dashed lines,
respectively. (Second, third, and fourth images) In situ pH, based on direct mea
surements (orange symbols) or as calculated from DIC and TA (green symbols), in
the surface layer and within layers centered at 250 and 1,000 m. Linear regressions
of the calculated and measured pH values are represented by solid and dashed
lines, respectively. (Dore et al., 2009)
OCR for page 35
CHEMISTRYOFSEAWATER
Figure 2-6
R01733
OCR for page 36
OCEANACIDIFICATION
FIgURE 2.7 Vertical distributions of anthropogenic CO2 concentrations (molmol kg1)
mol
and the saturation horizons for aragonite and calcite along northsouth transects in
the (A) Atlantic, (B) Pacific, and (C) Indian Oceans. A pressure of 1 decibar (1 db
on the yaxis) corresponds approximately to a depth of 1 meter (m). (Feely et al.,
2004) Figure 2-7
R01733
uneditable bitmapped image
OCR for page 37
CHEMISTRYOFSEAWATER
FIgURE 2.8 Atmospheric CO2 emissions, historical atmospheric CO2 levels and
predicted CO2 concentrations from this emissions scenario, together with changes
in ocean pH based on horizontally averaged chemistry. (Caldeira and Wickett,
2003)
Figure 2-8
R01733
As anthropogenic CO2 penetrates down in the water column, it
uneditable
decreases the CO 2 concentration bitmapped
and imageof CaCO super
hence the degree
3 3
saturation. The result is a slow upward migration, or shoaling, of the
saturation horizons for calcite and aragonite. This effect can already be
measured (Figure 2.7; Feely et al., 2004). As can be seen on Figure 2.7,
the extent of shoaling of the saturation horizons is uneven across ocean
basins, reflecting the differences in CO2 penetration caused by the com
plex movements of water masses.
2.2.3 Projections for Coastal Waters
The acidbase chemistry of coastal waters is much more complex
than that of open ocean surface and deep waters. It is affected by fresh
water and atmospheric inputs, the supply of both organic matter and
algal nutrients from land, and processes in the underlying sediments.
OCR for page 38
OCEANACIDIFICATION
Fresh water runoff tends to have higher dissolved CO2 concentrations
and lower pH than ocean water (Salisbury et al., 2008). In surface coastal
waters, high photosynthetic activity fueled by nutrient inputs can result
in low seasonal CO2 concentrations and high pH. In bottom waters, the
decomposition of organic matter, contributed either from land or from
local production, increases CO2 and decreases pH. A number of anthro
pogenic activities can exacerbate coastal acidification, principally those
that result in inputs of organic waste or algal nutrients, or that lead to the
formation of acid rain (Doney et al., 2007).
Many coastal areas also experience seasonal upwelling of CO2rich
deep water. In general, deep old waters in the ocean tend to have the
least invasion of fossil fuel CO2, but some upwelled waters are from shal
lower waters that are already subject to acidification by anthropogenic
CO2. This phenomenon has been shown to occur on the Pacific coast of
North America (Figure 2.9; Feely et al., 2008). On that coast, the seasonal
upwelling results in a natural seasonal cycle in pH and seawater carbon
ate chemistry; the extent and degree to which this has been amplified by
acidification, resulting in the breaching of corrosive, aragonite dissolving
water all the way to the surface, is an important research question. In
both river dominated and upwelling dominated coastal regions, future
trends in seawater carbon chemistry may also depend strongly on climate
change that influences wind patterns, upwelling and river flow. In shallow
waters, sediment dissolution can partly buffer acid inputs (Andersson et
al., 2003; Thomas et al., 2009).
2.2.4 Projections for High Latitudes
As seen in Figure 2.3, the cold waters of high latitude regions are
naturally low in carbonate ion concentration, owing to the increased solu
bility of CO2 at low temperature and ocean mixing patterns. As a result,
surface waters of these areas naturally have a lower degree of super
saturation of carbonate minerals and their acidbase chemistry is less
buffered than temperate and tropical surface waters. As the atmospheric
CO2 concentration increases, the pH and CO32 concentration in these
regions will decrease, and the saturation horizons of aragonite and cal
cite will move rapidly toward the surface (Olafsson et al., 2009). Seasonal
aragonite undersaturation in surface waters has already been observed in
the Canada Basin of the Arctic Ocean (Bates et al., 2009; YamamotoKawai
et al., 2009). Persistent undersaturation of surface waters with respect to
aragonite is projected to occur in high latitude regions by 2100, while in
lower latitude surface waters the degree or extent of supersaturation will
be reduced (Orr et al., 2005; Steinacher et al., 2009). This is illustrated in
Figure 2.10, which shows the projected changes in the aragonite satura
OCR for page 39
CHEMISTRYOFSEAWATER
FIgURE 2.9 Distribution of the depths of the undersaturated water (aragonite
saturation < 1.0; pH < 7.75) on the continental shelf of western North America
from Queen Charlotte Sound, Canada, to San Gregorio Baja California Sur, Mexico.
On transect line 5, the corrosive water reaches all the way to the surface in the
inshore waters near the coast. The black dots represent station locations. (Feely
et al., 2008)
OCR for page 40
0 OCEANACIDIFICATION
FIgURE 2.10 Surface water aragonite saturation state for the preindustrial ocean
(nominal year 1765), and years 1994, 2050, and 2100. Values for years 1765 and
1994 were computed from the global gridded data product GLODAP (Key et
Figure
al., 2004), whereas the saturation state 2-102050 and 2100 are the median of
for years
13 ocean general circulation models R01733
forced under the IPCC's IS92a "businessas
usual" CO2 emission scenario (Orr et bitmapped
uneditable al., 2005). (Fabry et al., 2008b)
image
OCR for page 41
CHEMISTRYOFSEAWATER
tion state of surface oceans under the "businessasusual" (i.e., IPCC's
IS92a) emissions scenario through the year 2100 (Orr et al., 2005; Fabry
et al., 2008b). Under current rates of CO2 emissions, models project that
surface waters of the Southern Ocean, the Arctic Ocean, and parts of the
subarctic Pacific will become undersaturated with respect to aragonite by
the end of this century, and, in some regions, as early as 2023 (Orr et al.,
2005; Steinacher et al., 2009).
2.3 CONTExT AND CONSTRAINTS FROM THE gEOLOgIC PAST
Information about past changes could be helpful for understanding
ongoing changes and their consequences. On time scales of thousands
of years and longer, the pH of the ocean is determined primarily by the
cycling of CaCO3 (and some silicate) minerals which are dissolved on
land, carried by rivers to the ocean where they are reprecipitated, and
eventually buried in sediments. Ocean acidification results from the fact
that this natural oceanic CaCO3 cycle cannot keep up with the rapid rise
in CO2. But eventually, over thousands of years, changes in CaCO3 cycling
will neutralize most of the excess acidity and restore the pH of the ocean
to nearpresentday value. Natural glacialinterglacial changes in atmo
spheric CO2 over the past 800,000 years, which are recorded in ice cores,
occurred over thousands of years, thus reducing the magnitude of change
in ocean pH for a given increase in atmospheric CO2 and allowing time
for the CaCO3 cycle to keep up (Ridgwell and Zeebe, 2005).
In the deeper geologic past, millions of years ago, atmospheric CO 2
concentrations were much higher than today, giving the Earth a warm
climate similar to the presentday tropics all the way to the high latitudes.
This is often referred to as "hot house" conditions as compared to present
day "ice house" conditions. Again, in these hothouse cycles, because the
CO2 concentration changed over millions of years, the CaCO3 cycle stabi
lized the pH of the ocean to these CO2 changes, as evidence by massive
CaCO3 deposits from those periods. While glacialinterglacial cycles and
hot houseice house cycles provide information regarding the response
of the ocean carbon cycle to changes in ocean pCO2 over thousands and
millions of years, they are not good analogs to current acidification of the
ocean by anthropogenic CO2.
2.4 MITIgATION AND gEOENgINEERINg
There is currently a great deal of international interest in mitigating
the impacts of climate change. However, this leads to the question of how
these mitigation strategies will affect ocean acidification and how ocean
acidification itself can be mitigated. Clearly, all mitigation strategies for
OCR for page 42
OCEANACIDIFICATION
climate change that reduce CO2 inputs to the atmosphere will also reduce
ocean acidification. These include increasing energy efficiency, shifting
energy sources from fossil fuels to nuclear and renewables, and imple
menting carbon capture and storage technologies (Pacala and Socolow,
2004). Similarly beneficial would be carbon management approaches that
remove CO2 from the atmosphere through biological sequestration on
land (e.g., afforestation, soil conservation) or industrialscale geochemical
approaches (Stephens and Keith, 2008). But geoengineering solutions
designed to slow climate warming without reducing atmospheric CO2
concentration, such as injection of sulfate aerosol precursors into the
stratosphere (Crutzen, 2006), will not reduce ocean acidification (Wigley,
2006; Boyd, 2008). On a regional scale, in coastal and estuarine waters
where acidification in surface waters may result partly from pollution
such as acid rain or in bottom waters from eutrophication induced by
excessive nutrient inputs, limiting emissions of air or water pollutants
may be effective as a mitigation strategy.
Management strategies designed to sequester CO2 in the ocean could
potentially exacerbate ocean acidification in intermediate or deep waters.
Iron fertilization of surface waters has been suggested as a potential
approach for boosting primary production in regions that are ironlimited,
thus increasing the export of organic carbon to the subsurface as dis
cussed in the next chapter (Boyd et al., 2007). Critics of iron fertilization
have questioned its efficiency at sequestering CO2 and pointed out the dif
ficulty in predicting its ecological consequences. Ocean acidification could
affect the efficiency of iron fertilization and its potential consequences
by modifying the biological availability of iron in surface seawater (see
section 2.2). If effective, iron fertilization would increase the rate of pen
etration of CO2 into intermediate waters, thus accelerating acidification
in those water masses. A similar effect would result from direct injec
tion of CO2 into intermediate or deep ocean waters. Enhanced deep sea
acidification could also occur as a result of leakage from subseabed CO2
sequestration (Blackford et al., 2009) either in sediments (House et al.,
2006) or bedrock (e.g., oil and gas fields, salt domes, etc.) (Caldeira and
Wickett, 2005).
The effectiveness of direct ocean CO2 injection techniques could theo
retically be enhanced and the resulting acidification minimized by first
reacting CO2 with a base, neutralizing the carbonic acid, and producing
primarily bicarbonate. Alternatively a base could be added directly to
seawater. Most proposed schemes use carbonate rock (e.g., limestone)
as the base and differ mostly in the techniques used to accelerate CaCO3
dissolution (Rau and Caldeira, 1999; Caldeira and Rau, 2000; Golomb et
al., 2007; Harvey, 2008; Rau, 2008). Sodium hydroxide (NaOH) could also
be produced from water electrochemically with the coproduced hydro
OCR for page 43
CHEMISTRYOFSEAWATER
chloric acid (HCl) being neutralized by silicate rocks (House et al., 2007).
Given that neutralization of CO2 requires an equivalent amount of base
(1:1 molar ratio), the logistics and resource demands for neutralizing a
significant fraction of the 2 Pg C (~1014 moles CO2) per year taken up by
the ocean are likely to be prohibitive. But such mitigation strategies might
be feasible on a local or regional scale.
OCR for page 44
