Ocean Acidity

Identification

1.	Indicator Description

This indicator shows recent trends in acidity levels in the ocean at three key locations. The indicator also
presents changes in aragonite saturation by comparing historical data with the most recent decade.
Ocean acidity and aragonite saturation levels are strongly affected by the amount of carbon dissolved in
the water, which is directly related to the amount of carbon dioxide (C02) in the atmosphere. Acidity
affects the ability of corals, some types of plankton, and other creatures to produce their hard skeletons
and shells. This indicator provides important information about an ecologically relevant effect
associated with climate change.

Components of this indicator include:

•	Recent trends in ocean C02 and acidity levels (Figure 1)

•	Historical changes in the aragonite saturation of the world's oceans (Figure 2)

2.	Revision History

April 2010: Indicator published.

May 2014: Updated Figure 1 with data through 2012 for two sampling locations; updated Figure 2
with data through 2013.

August 2016: Updated Figure 1 with data through 2014 for Hawaii and 2015 for Bermuda; updated
Figure 2 with data through 2015.

Data Sources

3. Data Sources

Figure 1 includes trend lines from three different ocean time series: the Bermuda Atlantic Time-Series
Study (BATS); the European Station for Time-Series in the Ocean, Canary Islands (ESTOC); and the Hawaii
Ocean Time-Series (HOT). Data from these stations have been used in numerous peer-reviewed
publications—for example, Bates et al. (2014).

Figure 2 contains aragonite saturation (Qar) calculations derived from atmospheric C02 records from ice
cores and observed atmospheric concentrations at Mauna Loa, Hawaii. These atmospheric C02
measurements are fed into the Community Earth Systems Model (CESM), maintained by the National
Center for Atmospheric Research (NCAR). CESM is a dynamic ocean model that computes ocean C02
uptake and the resulting changes in seawater carbonate ion (C032 ) concentration and Qar over time.

Technical Documentation: Ocean Acidity

1


-------
4. Data Availability

Figure 1 compiles pC02 (the mean seawater C02 partial pressure in piatm) and pH data from three
sampling programs in the Atlantic and Pacific Oceans. Raw data from the three ocean sampling
programs are publicly available online. In the case of Bermuda and the Canary Islands, updated data
were procured directly from the scientists leading those programs. BATS data and descriptions are
available at: http://batsftp.bios.edu/BATS/bottle/. ESTOC data can be downloaded from:
www.eurosites.info/estoc/data.php. although the ESTOC surface buoy stopped transmitting data after it
was damaged in July 2010, and it had not yet been repaired or replaced as of July 2016. HOT data were
downloaded from the HOT Data Organization and Graphical System website at:

http://hahana.soest.hawaii.edu/hot/products/products.html. Additionally, annual HOT data reports are
available at: http://hahana.soest.hawaii.edu/hot/reports/reports.html.

The map in Figure 2 is derived from the same source data as NOAA's Ocean Acidification "Science on a
Sphere" video simulation at: http://sos.noaa.gov/Datasets/list.php?category=Ocean (Feely et al., 2009).
EPA obtained the map data from Dr. Ivan Lima of the Woods Hole Oceanographic Institution (WHOI).

Methodology

5. Data Collection

Figure 1. Ocean Carbon Dioxide Levels and Acidity, 1983-2015

This indicator reports on the pH of the upper 5 meters of the ocean and the corresponding partial
pressure of dissolved C02 (pC02). Each data set covers a different time period:

•	BATS data used in this indicator are available from 1983 to 2015. Samples were collected from
two locations in the Atlantic Ocean near Bermuda (BATS and Hydrostation S, at 31°43' N, 64°10'
W and 32°10' N, 64°30' W, respectively). See: http://bats.bios.edu/bats location.html.

•	ESTOC data are available from 1995 to 2009. ESTOC is at (29°10' N, 15°30' W) in the Atlantic
Ocean.

•	HOT data are available from 1988 to 2014. The HOT station is at (23° N, 158° W) in the Pacific
Ocean.

At the BATS and HOT stations, dissolved inorganic carbon (DIC) and total alkalinity (TA) were measured
directly from water samples. DIC accounts for the carbonate and bicarbonate ions that occur when C02
dissolves to form carbonic acid, while total alkalinity measures the buffering capacity of the water,
which affects the partitioning of DIC among carbonate and bicarbonate ions. At ESTOC, pH and alkalinity
were measured directly (Bindoff et al., 2007).

Each station followed internally consistent sampling protocols over time. Bates et al. (2012) describe the
sampling plan for BATS. Further information on BATS sampling methods is available at:
http://bats.bios.edu. ESTOC sampling procedures are described by Gonzalez-Davila et al. (2010). HOT
sampling procedures are described in documentation available at:

Technical Documentation: Ocean Acidity

2


-------
http://hahana.soest.hawaii.edu/hot/hot jgofs.html and:
http://hahana.soest.hawaii.edu/hot/products/HOT surface C02 readme.pdf.

Figure 2. Changes in Aragonite Saturation of the World's Oceans, 1880-2015

The map in Figure 2 shows the estimated change in sea surface Qar from 1880 to 2015. Aragonite
saturation values are calculated in a multi-step process that originates from historical atmospheric C02
concentrations that are built into the model (the CESM). As documented in Orr et al. (2001), this model
uses historical atmospheric C02 concentrations based on ice cores and atmospheric measurements (the
latter collected at Mauna Loa, Hawaii).

6. Indicator Derivation

Figure 1. Ocean Carbon Dioxide Levels and Acidity, 1983-2015

At BATS and HOT stations, pH and pC02 values were calculated based on DIC and TA measurements
from water samples. BATS analytical procedures are described by Bates et al. (2012). HOT analytical
procedures are described in documentation available at:
http://hahana.soest.hawaii.edu/hot/hot jgofs.html and:

http://hahana.soest.hawaii.edu/hot/products/HOT surface C02 readme.pdf. At ESTOC, pC02 was
calculated from direct measurements of pH and alkalinity. ESTOC analytical procedures are described by
Gonzalez-Davila et al. (2010). For all three locations, Figure 1 shows in situ measured or calculated
values for pC02 and pH, as opposed to values adjusted to a standard temperature.

The lines in Figure 1 connect points that represent individual sampling events. No attempt was made to
generalize data spatially or to portray data beyond the time period when measurements were made.
Unlike some figures in the published source studies, the data shown in Figure 1 are not adjusted for
seasonal variability. The time between sampling events is somewhat irregular at all three locations, so
moving averages and monthly or annual averages based on these data could be misleading. Thus, EPA
elected to show individual measurements in Figure 1.

Figure 2. Changes in Aragonite Saturation of the World's Oceans, 1880-2015

The map in Figure 2 was developed by WHOI using the CESM, which is available publicly at:
www.cesm.ucar.edu/models. Atmospheric C02 concentrations were fed into the CESM, which is a
dynamic ocean model that computes ocean C02 uptake and the resulting changes in seawater
carbonate concentration over time. The CESM combines this information with monthly salinity and
temperature data on an approximately 1° by 1° grid. Next, these monthly model outputs were used to
approximate concentrations of the calcium ion (Ca2+) as a function of salt (Millero, 1982), and to
calculate aragonite solubility according to Mucci (1983). The resulting aragonite saturation state was
calculated using a standard polynomial solver for MATLAB, which was developed by Dr. Richard Zeebe of
the University of Hawaii. This solver is available at:

www.soest.hawaii.edu/oceanographv/faculty/zeebe files/C02 System in Seawater/csys.html.
Aragonite saturation state is represented as Qar, which is defined as:

Qar = [Ca2+][C032-]/K'Sp

Technical Documentation: Ocean Acidity

3


-------
The numerator represents the product of the observed concentrations of calcium and carbonate ions.
K'Sp is the apparent solubility product, which is a constant that is equal to [Ca2+][C032~] at equilibrium for
a given set of temperature, pressure, and salinity conditions. Thus, Qar is a unitless ratio that compares
the observed concentrations of calcium and carbonate ions dissolved in the water with the
concentrations that would be observed under fully saturated conditions. An Qar value of 1 represents
full saturation, while a value of 0 indicates that no calcium carbonate is dissolved in the water. Ocean
water at the surface can be supersaturated with aragonite, however, so it is possible to have an Qar
value greater than 1, and it is also possible to experience a decrease over time, yet still have water that
is supersaturated.

For Figure 2, monthly model outputs were averaged by decade before calculating Qar for each grid cell.
The resulting map is based on averages for two decades: 1880 to 1889 (a baseline) and 2006 to 2015
(the most recent complete 10-year period). Figure 2 shows the change in Qar between the earliest
(baseline) decade and the most recent decade. It is essentially an endpoint-to-endpoint comparison, but
using decadal averages instead of individual years offers some protection against inherent year-to-year
variability. The map has approximately 1° by 1° resolution.

7. Quality Assurance and Quality Control

Quality assurance and quality control (QA/QC) steps are followed during data collection and data
analysis. These procedures are described in the documentation listed in Sections 5 and 6.

Analysis

8.	Comparability Over Time and Space

Figure 1. Ocean Carbon Dioxide Levels and Acidity, 1983-2015

BATS, ESTOC, and HOT each use different methods to determine pH and pC02, though each individual
sampling program uses well-established methods that are consistent over time.

Figure 2. Changes in Aragonite Saturation of the World's Oceans, 1880-2015

The CESM calculates data for all points in the Earth's oceans using comparable methods. Atmospheric
C02 concentration values differ in origin depending on their age (i.e., older values from ice cores and
more recent values from direct atmospheric measurement); however, all biogeochemical calculations
performed by the CESM use the atmospheric C02 values in the same manner.

9.	Data Limitations

Factors that may impact the confidence, application, or conclusions drawn from this indicator are as
follows:

1. Carbon variability exists in the surface layers of the ocean as a result of biological differences,
changing surface temperatures, mixing of layers as a result of ocean circulation, and other
seasonal variations.

Technical Documentation: Ocean Acidity

4


-------
2.	Changes in ocean pH and mineral saturation caused by the uptake of atmospheric C02 can take
a long time to spread to deeper waters, so the full effect of atmospheric C02 concentrations on
ocean pH may not be seen for many decades, if not centuries.

3.	Ocean chemistry is not uniform throughout the world's oceans, so local conditions could cause a
pH measurement to seem incorrect or abnormal in the context of the global data. Figure 1 is
limited to three monitoring sites.

4.	Although closely tied to atmospheric concentrations of C02, aragonite saturation is not
exclusively controlled by atmospheric C02, as salinity and temperature are also factored into the
calculation.

10.	Sources of Uncertainty

Figure 1. Ocean Carbon Dioxide Levels and Acidity, 1983-2015

Uncertainty measurements can be made for raw data as well as analyzed trends. Details on uncertainty
measurements can be found in the following documents and references therein: Bindoff et al. (2007),
Bates et al. (2012), Dore et al. (2009), and Gonzalez-Davila et al. (2010).

Figure 2. Changes in Aragonite Saturation of the World's Oceans, 1880-2015

Uncertainty and confidence for CESM calculations, as they compare with real-world observations, are
measured and analyzed in Doney et al. (2009) and Long et al. (2013). Uncertainty for the approximation
of Ca2+ and aragonite solubility are documented in Millero (1982) and Mucci (1983), respectively.

11.	Sources of Variability

Aragonite saturation, pH, and pC02 are properties of seawater that vary with temperature and salinity.
Therefore, these parameters naturally vary over space and time. Variability in ocean surface pH and
pC02 data has been associated with regional changes in the natural carbon cycle influenced by changes
in ocean circulation, climate variability (seasonal changes), and biological activity (Bindoff et al., 2007).

Figure 1. Ocean Carbon Dioxide Levels and Acidity, 1983-2015

Variability associated with seasonal signals is still present in the data presented in Figure 1. This seasonal
variability can be identified by the oscillating line that connects sampling events for each site.

Figure 2. Changes in Aragonite Saturation of the World's Oceans, 1880-2015

Figure 2 shows how changes in Qar vary geographically. Monthly and yearly variations in C02
concentrations, temperature, salinity, and other relevant parameters have been addressed by
calculating decadal averages.

12.	Statistical/Trend Analysis

This indicator does not report on the slope of the apparent trends in ocean acidity and pC02 in Figure 1.
The long-term trends in Figure 2 are based on an endpoint-to-endpoint comparison between the first

Technical Documentation: Ocean Acidity

5


-------
decade of widespread data (the 1880s) and the most recent complete 10-year period (2006-2015). The
statistical significance of these trends has not been calculated.

References

Bates, N.R., M.H.P. Best, K. Neely, R. Garley, A.G. Dickson, and R.J. Johnson. 2012. Detecting
anthropogenic carbon dioxide uptake and ocean acidification in the North Atlantic Ocean.
Biogeosciences 9:2509-2522.

Bates, N.R., Y.M. Astor, M.J. Church, K. Currie, J.E. Dore, M. Gonzalez-Davila, L. Lorenzoni, F.E. Muller-
Karger, J. Olafsson, and J.M. Santana-Casiano. 2014. Changing ocean chemistry: A time-series view of
ocean uptake of anthropogenic C02 and ocean acidification. Oceanography 27(1):121-141.

Bindoff, N.L., J. Willebrand, V. Artale, A, Cazenave, J. Gregory, S. Gulev, K. Hanawa, C. Le Quere, S.
Levitus, Y. Nojiri, C.K. Shum, L.D. Talley, and A. Unnikrishnan. 2007. Observations: Oceanic climate
change and sea level. In: Climate change 2007: The physical science basis (Fourth Assessment Report).
Cambridge, United Kingdom: Cambridge University Press.

Doney, S.C., I. Lima, J.K. Moore, K. Lindsay, M.J. Behrenfeld, T.K. Westberry, N. Mahowald, D.M. Glober,
and T. Takahashi. 2009. Skill metrics for confronting global upper ocean ecosystem-biogeochemistry
models against field and remote sensing data. J. Marine Syst. 76(l-2):95-112.

Dore, J.E., R. Lukas, D.W. Sadler, M.J. Church, and D.M. Karl. 2009. Physical and biogeochemical
modulation of ocean acidification in the central North Pacific. P. Natl. Acad. Sci. USA 106:12235-12240.

Feely, R.A., S.C. Doney, and S.R. Cooley. 2009. Ocean acidification: Present conditions and future
changes in a high-C02 world. Oceanography 22(4):36-47.

Gonzalez-Davila, M., J.M. Santana-Casiano, M.J. Rueda, and O. Llinas. 2010. The water column
distribution of carbonate system variables at the ESTOC site from 1995 to 2004. Biogeosciences 7:1995-
2032.

Long, M.C., K. Lindsay, S. Peacock, J.K. Moore, and S.C. Doney. 2013. Twentieth-century ocean carbon
uptake and storage in CESMl(BGC). J. Climate 26(18):6775-6800.

Millero, F.J. 1982. The thermodynamics of seawater. Part I: The PVT properties. Ocean Phys. Eng.
7(4):403-460.

Mucci, A. 1983. The solubility of calcite and aragonite in seawater at various salinities, temperatures,
and one atmosphere total pressure. Am. J. Sci. 283:780-799.

Orr, J.C., E. Maier-Reimer, U. Mikolajewicz, P. Monfray, J.L. Sarmiento, J.R. Toggweiler, N.K. Taylor, J.
Palmer, N. Gruber, C.L. Sabine, C.L. Le Quere, R.M. Key, and J. Boutin. 2001. Estimates of anthropogenic
carbon uptake from four three-dimensional global ocean models. Global Biogeochem. Cy. 15(l):43-60.

Technical Documentation: Ocean Acidity

6


-------