Proceedings from the University of Washington

School of Oceanography Senior Thesis, Academic Year 2011-2012

Extent of the oxygen minimum zone in the eastern tropical North

Pacific
Erika A. Fee1
1
University of Washington, School of Oceanography,
Box 355351, Seattle, Washington 98195
*
[email protected]
Received June 2012

NONTECHNICAL SUMMARY

Off the west coast Mexico, in the Eastern Tropical North Pacific, there is a vast layer of water
with extremely low oxygen content. This low-oxygen layer, commonly referred to as an oxygen minimum
zone (OMZ), has been surveyed numerous times since the 1960s. According to a theory proposed in
1999, a rise in global temperatures will cause the volume of the OMZ to increase, becoming both
shallower and deeper in the water column. To determine whether or not there has been a change in the
vertical extent of the OMZ in the ETNP, I made oxygen concentration profiles along a transect from San
Diego, CA to Manzanillo, MX, and made a cross-sectional profile of oxygen concentrations along 110°W
to compare with past studies. The primary data presented in this paper was gathered during R/V Thomas.
G. Thompson cruise TN278 which took place 16 March to 23 April 2012. I found that the OMZ is
encroaching on coastal areas and has increased in size and intensity over the past forty years. It is
important to understand the extent and changes with time of the OMZ, as OMZs have a strong impact on
the carbon and nitrogen cycles through carbon sequestration and denitrification. The feedback effects that
change these cycles make OMZs important areas to study so that we may better predict future climatic
and oceanic conditions.
.

ABSTRACT

Off the west coast Mexico, in the Eastern Tropical North Pacific, there is a vast Oxygen
Minimum Zone (OMZ). An analysis of O2 concentration measurements in the ETNP made in 1972
identified a 600m thick layer of water with almost no detectable oxygen (Cline and Richards 1972).
According to a theory proposed in 1999, a rise in global temperatures will cause the volume of the OMZ
to increase, becoming both shallower and deeper in the water column (Cannariato and Kennett 1999). The
primary data used for this paper was gathered during R/V Thomas. G. Thompson cruise TN278 which
took place 16 March to 23 April 2012. To determine whether or not there has been a change in the
vertical extent of the OMZ, I collected water samples with Niskin bottles at ten stations along a transect
from San Diego, CA to Manzanillo, MX and gathered oxygen concentration data using a CTD sensor
while on board the R/V Thomas G. Thompson. Titrations were performed on the water samples using the
Winkler method (Winkler 1888) as modified by Carpenter (1965). The results of the titration were used to
calibrate the sensor readings, and corrected sensor profiles were compared to observations made during
past cruises. It is important to understand the extent and changes with time of the OMZ, as OMZs have a
strong impact on the carbon and nitrogen cycles (Devol and Hartnett 2001, Paulmier and Ruiz-Pino 2009)
through carbon sequestration and denitrification. The feedback effects that changes to these cycles may
have makes OMZs important areas to study so that we may better predict future climatic and oceanic
conditions.
1
 The oxygen minimum zone of the Eastern Kennett 1999). Heating of surface waters reduces
Tropical North Pacific is one of the better studied the solubility of oxygen and increases the strength
OMZ in the ocean. This zone has been surveyed of the thermocline, further contributing to OMZ
numerous times since the 1960s, with oxygen expansion (Helm et al 2011).
concentration measurements taken as far down as
1500m (Richards 1965). When this OMZ was The OMZ may also have a feedback effect
observed during PONCHO cruise TT-026 in on global climate change through its effects on the
March 1968 and PISCO cruises TT-035 and -037 carbon and nitrogen cycles. Reduced
in February and April 1969 aboard the RV Thomas decomposition of sinking organic matter in OMZs
G. Thompson, the OMZ was defined as a layer causes an increased amount of atmospheric CO2, a
extending from 200 m to 800 m where O2 greenhouse gas, fixed by primary productivity to
concentrations fell below 1 µg L-1 (Cline and be sequestered in the sediment underlying OMZ s
Richards 1972). A later study of oxygen (Devol and Hartnett 2001). Expansion in OMZs
minimum zones in the Atlantic and Pacific oceans also causes an increase in heterotrophic
showed that the ETNP is part of a layer of oxygen- denitrification and anammox reactions, which are
depleted water spanning the coast of Central and responsible for fixed nitrogen loss and have a
South America from approximately 20°N to 15°S negative feedback effect on oceanic primary
(Karstensen et al 2008). This same study defined productivity (Paulmier and Ruiz-Pino 2009),
the OMZ as a layer extending from 100m to 900m reducing the amount of atmospheric CO2
where O2 concentrations were less than 0.1 ml L-1. sequestered by the ocean. OMZs are responsible
for a significant fraction of global emissions of
Since becoming areas of particular trace greenhouse gases N2O and CH4, which have
interest, there have been many different units and more powerful positive radiative forcing effects
values used to define OMZs. In a recent paper by than CO2 (Wright et al 2012). Because of the
Jody Wright, the OMZ was described as a region impact OMZs have the cycling of carbon and
of suboxic and anoxic waters where oxygen nitrogen, it is important to have a better picture of
concentrations fell below the threshold at which the current extent of the OMZ and determine
alternative electron acceptors begin to replace the whether it has increased in size, and if it has, by
use of oxygen, < 20 µmol kg-1 (Wright et al 2012). how much.
Therefore, for this paper, the OMZ will be defined
as the region of the water column where oxygen METHODS
concentrations are ≤ 20 µmol kg-1. Hydrographic
surveys of this region suggest that the low oxygen The research cruise, TN278, took place 16
concentration results from a combination of high March to 23 April 2012 on board the R/V Thomas
primary production at the surface, strong thermal G. Thompson. Oxygen samples were taken at ten
stratification that inhibits, and poor ventilation stations along the first leg of the cruise transect
(Fiedler and Talley 2006). These same where depths exceeded 1000m (Fig. 1). Only the
hydrographic surveys also found that the OMZ is sensor data was used for the second leg of the
the thickest off the coast of Baja California in the cruise. Oxygen concentration measurements were
eastern tropical North Pacific. made at each station using a SBE 43 electrode
attached to the CTD, and water samples were
It has been theorized that OMZs fluctuate taken using Niskin bottles fired off at eight to ten
in size in response to global climate change. depths on the return trip. My colleague Miles Carl
Global temperature changes cause the strength and and I used about one liter of water from each
location of Pacific Intermediate Water ventilation Niskin bottle for duplicate and triplicate O2
and surface productivity to fluctuate, resulting in samples. We used Winkler titration (Winkler
an increase in the thickness of low oxygen layers 1888) as modified by the Carpenter method
during warmer periods and a decrease in thickness (Carpenter 1965) with the shipboard Metrohm 765
when the climate is colder (Cannariato and

2
Dosimat manual titrator in order to find the Station 2 was unavailable, while the samples for
oxygen concentration of the samples. Station 26 were contaminated.

Figure 1: TN278 cruise path from San Diego to Figure 2: Sensor values plotted against titration
Manzanillo. The orange stations are where bottle values.
samples were taken in addition to O2 sensor data.

TheFigure 1. This is a textmethod

Carpenter-Winkler box setwas
up to follow
used the
because,
if done correctly, this method has been shown to According to Figure 2, the sensor data
produce results accurate to within ±1% (Emerson needed to be corrected by a factor of 0.9452 in
et al 1999). The oxygen titration results were in order to reflect the actual oxygen concentration
mg-at L-1, while readings from the SBE 43 values. Once corrected, the O2 concentrations were
electrode were in µmol kg-1, so the titration results plotted against depth and density (Fig. 3; Fig. 4).
had to be converted in order to compare the two.
The equation used to convert to µmol kg-1 was: Figure 3: Oxygen sensor profiles with depth for
sample stations.

The equation used to get the density (ρ) was:

( )
( ) ( (
( ) )) ( (
( ) ))( )
Where z is depth (km), T is temperature (K), and S
is salinity (PSU).

RESULTS
The titration results were plotted against
sensor readings for every depth water samples
were taken using Excel in order to produce a
calibration factor for the sensor (Fig. 2). The data
from Stations 2 and 26 ended up being unusable.
The bottle data for the sample bottles used at

3
 Oxygen data from every CTD cast of
Figure 4: Oxygen sensor profiles with density for 1000m or more was evaluated using Ocean Data
sample stations. View (Fig. 5). Each point of data was extrapolated
out to produce a continuous cross-sectional profile
following the cruise track. To examine the changes
in the OMZ with time, a cross-sectional profile of
the second leg of the cruise following 110°W was
compared with cross-sections of data for the same
track (Fig. 6) made by previous publications
(Codispoti and Richards 1976, WOCE 2002).

Figure 5: Oxygen profile cross-section, for stations where depth exceeded 1000m, along the first leg of the
cruise. The color scale is constrained to oxygen concentrations ≤ 20 µmol kg-1.

4
 Figure 6: Oxygen profile cross-section of the second leg of the cruise (TN278), compared with cross-sections of oxygen
data from previous studies (Codispoti and Richards 1976, WOCE 2002). The color scales is the same as Figure 5.

DISCUSSION Comparisons of oxygen with density show a

similar trend of decreasing oxygen.
The sensor readings were more precise than the
titrations, but were unable to provide accurate
Looking at the cross-sectional, the
oxygen readings below a concentration of 1 µmol
northern most boundary of the OMZ is somewhere
kg-1 (Fig. 2). It was also found that the sensor
above 16°N, and it gets shallower at an average
tended to underestimate O2 values in the mixed
rate of ~0.2 m km-1 and deeper at an average rate
layer, returning values that were about 90%
of ~0.16 m km-1 (Fig. 5). Of particular concern is
saturation. Despite these two discrepancies, the
that the OMZ is encroaching upon the continental
general trend of O2 concentrations along the first
slope and coastal waters at the south-east end of
leg of the cruise transect was a progressively
the transect. This area is experiencing an
shallower and deeper OMZ the further south and
extremely shallow (~40m) OMZ where oxygen
east one travels along the transect (Fig. 3).
concentrations are < 5 µmol kg-1 throughout much

5
of the water column with an especially thick oxygen profile for Station 13 shows an inversion
anoxic layer close to the surface. As 50% of between ~175 m (1027 kg m-3) and ~350m (1029
coastal marine animals cannot survive in waters kg m-3), and the O2 concentration values are
where the oxygen concentrations are < 70 µmol greater than those of the previous station over the
kg-1, this “shoaling” of the OMZ, coupled with same range (Fig. 3, Fig. 4). Given Station 13’s
increasing ocean acidification, can have an close proximity to the continental shelf break, it is
extremely negative impact on marine animal possible that this inversion represents upwelling of
populations and local fishing (Seibel 2010). oxygenated water at this location. Comparing the
Determining the horizontal extent of the OMZ oxygen profile with a salinity profile appears to
would provide a better understanding of how support the possibility of upwelling (Fig. 7).
coastal areas are being affected. ADCP data did not go deep enough to provide
current directions over the depth range of the
Station 13 is an exception to the overall inversion that would confirm upwelling.
trend of a progressively shallower OMZ. The

Figure 7: Profiles of salinity and oxygen for Station 13. Both show an inversion between 175
m and 350 m.

Comparing the cross-section of the second The changes in the OMZ in the ETNP suggest that
leg of the cruise following 110°W to observations this percentage has likely increased and will
of the OMZ along the same transect made by continue to increase. The oxygenated surface layer
Codispoti and Richards (1972) and WOCE Global has been compressed, decreasing habitat available
Data (WOCE 2002), there have been some to marine life, especially high-performance species
significant changes in the OMZ with time. Over like tuna, and increasing the distance organisms
the past forty years, the OMZ has gotten have to travel during diel vertical migration in
shallower, deeper, and more oxygen deficient. In order to reach oxygenated waters (Stramma et al
2009, an estimated 8% of the ocean’s volume 2012, Wright et al 2012).
contained less than 20 µmol O2 kg-1 (Seibel 2010).

6
CONCLUSIONS margin oxygen-minimum zone during the
past 60 k.y. Geology 27:
These findings of an increase in size and
975 – 978.
intensity of the OMZ in the ETNP are consistent
with past studies of OMZs. Assuming the
Carpenter, J.H., 1965a. The accuracy of the
increasing size of the OMZ in the ETNP is
Winkler method for dissolved oxygen.
representative of expansion of OMZs worldwide,
Limnol. Oceanogr. 10: 135 – 140.
the ocean is becoming increasingly oxygen
deficient as global temperatures increase. The
Carpenter, J.H. 1965b. The Chesapeake Bay
positive and negative feedback effects carbon and
Institute technique for the Winkler
nitrogen cycling in the OMZs have on global
dissolved oxygen method. Limnol.
climate are complex and proportional to the size of
Oceanogr. 10: 141-143.
the OMZ (Devol and Hartnett 2001, Paulmier and
Ruiz-Pino 2009, Wright et al 2012). In order to
Cline, J. D., and Richards, F. A. 1972. Oxygen
understand the overall feedback effect OMZs have
deficient conditions and nitrate reduction
on global climate, a better understanding of these
in the eastern tropical North Pacific ocean.
processes as well as the current and potential
Limnol. Oceanogr. 17: 885 – 900.
future extent of global OMZs is crucial.
Codispoti, L. A., and Richards, F. A. 1976. An
Increasing oxygen stress and vulnerability
analysis of the horizontal regime of
to overfishing resulting from the increase in the
denitrificaiton in the eastern tropical North
size of OMZs will have negative consequences for
Pacific. Limnol. Oceanogr. 21: 379 – 388.
the populations of marine organisms. This is
especially true for species such as tuna which have
Devol, A. H., and Hartnett, H. E. 2001. Role of the
a high oxygen demand and are the targets of
oxygen-deficient zone in transfer of
commercial fishing (Stramma et al 2012).
organic carbon to the deep ocean. Limnol.
Compression of the oxygenated surface layer and
Oceanogr. 46: 1684 – 1690.
shoaling of the OMZ greatly reduce the amount of
habitat available to marine life and will negatively
Emerson, S., Stump, C., Wilbur, D., and Quay, P.
impact population sizes and biodiversity.
1999. Accurate measurement of O2, N2,
Identification and careful management of
and Ar gases in water and the solubility of
particularly severe OMZs is necessary to maintain
N2. Marine Chemistry 64: 337 – 347.
sustainable populations of marine animals and
avoid future food shortages.
Fiedler, P. C., and Talley, L. D. 2006.
Hydrography of the eastern tropical
ACKNOWLEDGEMENTS Pacific: A review. Prog. Oceanogr. 69:
I thank my colleague Miles Carl for his 143 – 180.
assistance in gathering data, my advisers Al Devol
and Rick Keil for their inspiring ideas and helpful Helm, K. P., Bindoff, N. L., and Church, J. A.
discussion, Rachel Horak and Will Kammin for 2011. Observed decreases in oxygen
providing the additional data I needed to complete content of the global ocean. Geophys. Res.
my project, and my classmates, the marine techs Lett. 38, L23602.
and the crew of the RV Thomas G. Thompson for
their assistance at sea. Karstensen, J., Stramma, L., and Visbeck, M.
2008. Oxygen minimum zones in the
eastern tropical Atlantic and Pacific
REFERENCE LIST oceans. Prog. Oceanogr. 77: 331 – 350.
Cannariato, K. G., and Kennett, J. P. 1999.
Climatically related millennial-scale Paulmier, A., and Ruiz-Pino, D. 2009. Oxygen
fluctuation in strength of California minimum zones (OMZs) in the modern
ocean. Prog. Oceanogr. 80: 113 – 128.

7
Richards, F. A. 1965. Physical, chemical, and
productivity data from a survey of the
Caribbean Sea and the northeastern
tropical Pacific Ocean. Univ of WA
Technical Report 249.

Seibel, B. A. 2010. Critical oxygen levels and

metabolic suppression in oceanic oxygen
minimum zones. J. Exp. Biol. 214: 326 –
336.

Stramma, L., Prince, E. D., Schmidtko, S., Luo, J.,

Hoolihan, J. P., Visbeck, M., Wallace, D.
W. R., Brandt, P. and Körtzinger, A.
2012.Expansion of oxygen minimum
zones may reduce available habitat for
tropical pelagic fishes. Nature Clim.
Change 2: 33 – 37.

Winkler, L.S. 1888. The determination of

dissolved oxygen. Ber. Dtsche. Chem.
Ges. 21: 2843-2855.

WOCE Data Products Committee, 2002: WOCE

Global Data, Version 3.0, WOCE
International Project Office, WOCE
Report No. 180/02, Southampton, UK

Wright, J. J., Konwar, K. M., and Hallam, S. J.

2012. Microbial ecology of expanding
oxygen minimum zones. Nat. Rev.
Microbiol 10: 381 – 394.