O2 Cycle by yurtgc548

VIEWS: 3 PAGES: 42

• pg 1
```									Changes in O2 over Earth’s History

1
Annual Cycle in
Atmospheric O2

Barrow 71ºN
Samoa 14ºS
C. Grim 43ºS

(1 ppm O2 = 5 per meg)

2
Oxygen Isotopes
16O   = 99.74%, 17O = 0.05%, 18O = 0.21%

d18O (‰) = [(18O/16O)sample/(18O/16O)std –1]*1000

d17O (‰) = [(17O/16O)sample/(17O/16O)std –1]*1000

17Δ (per   meg) = [d17O – 0.518*d18O]*1000

Standards: SMOW or AIR

(d18O of O2 in air = +23.5 ‰ vs SMOW)
3
Molecular O2 Cycle
Important Processes
- photosynthesis, respiration
- air-water gas exchange, mixing, circulation
Photosynthesis (O2 is from the water molecule)
CO2 + 2H2O* +light  CH2O + *O2

Respiration
H2O + *O2 + CH2O  CO2 + 2H2O*
Isotope KIE during Photosynthesis
Fractionation effect during
photosynthesis by Synechocystis

(Guy et al., 1993)                   (Helman et al., 2005)

Little or no fractionation during photosynthetic
production of O2 (<1‰)                                   5
d18O of Precipitation Globally

Mean d18O of precipitation ~ -4 ‰ (assuming mean temp = 15ºC)
6
d18O (‰) of Surface Ocean

7
Respiration KIEs
for d18O of O2

8
Atmospheric Dole Effect
• d18O of O2 in air is +23.5 ‰ (vs SMOW)
• At steady-state, the d18O of O2 produced by photosynthesis has
to equal the d18O of O2 consumed by respiration.
(18O/16O)water* photo = (18O/16O)O2air* resp

• Since photo = 1.000, then
resp = (18O/16O)water / (18O/16O)O2air

For marine photosynthesis: resp = 1.000 / 1.0235 = 0.9770
For terrestrial photosynthesis: resp = 1.008 / 1.0235 = 0.9849
For 50/50 split: resp = 1.004 / 1.0235 = 0.9809                  9
Variations in d18O-O2 over glacial cycles

(Petite et al., 1999)   10
Processes Affecting Concentration and d18O of
dissolved O2 in Surface Layer
Air-Water O2 Gas Exchange

g

Inflow                              Respiration

Organic Carbon         Turbulent Mixing,
Export (= P – R)       Entrainment, Upwelling,   11
Eddies, etc.
Effects of Respiration, Photosynthesis and
Gas Exchange on d18O and O2

O2 Concentration 
- Photosynthesis decreases d18O
- Respiration increases d18O
- Gas Exchange drives d18O toward equilibrium (24.2 ‰)   12
d18O-O2 in Amazon Lakes and Rivers

Lakes= squares; Amazon R. = circles; Tributaries = triangles
13
R/P of Amazon Lakes and Rivers

14
Lakes= squares; Amazon R. = circles; Tributaries = triangles
Diurnal Cycles in O2 and d18O in Lakes
Tonle Sap Lake, Cambodia                                          Flooded Forest Pond, Canada
0                                                         22
d 18O (‰) (vs AIR)

-2                                                        20
-4                                                        18
-6                                                        16

O/Ar
-8                                                        14
-10                                                        12
-12                                                        10
-14                                                         8
6:00 AM   12:00 PM   6:00 PM    12:00 AM   6:00 AM   12:00 PM

d18O      O/Ar

Tonle Sap provides                  To detect a diurnal O2 and
75% of fish harvested
in Cambodia (D.                     d18O cycle typically high
Lockwood, unpub                     rates of photosynthesis,
data)                               low gas exchange rates and
15
shallow water body.
Seasonal Cycle in O2 and d18O in Mekong R.

4.00                                            240                             16                                             40000
O2sat
14                                             35000
3.00                                            230

Dissolved O 2 (uM)
or % saturation
d18O (‰)                                                              12                                             30000

Discharge (m3/s)
2.00                                            220
d1818Ovs. air (‰)
d O vs. AIR (‰)

10                                             25000

R:P
1.00                                            210   [O2] (µM)                 8                                              20000

[O2] (µM)                                                          6                                              15000
0.00                                            200
4                                              10000
d18O
-1.00                                           190                             2                                              5000
=
R:P O21
Rainy          Dry        Rainy                                       0                                              0
-2.00                                           180                       J- A- S- O- N- D- J- F- M- A- M- J- J- A- S- O- N-
J J A S O N D J F M A M J J A S O N D                       O2sat 05 05 05 05 05 05 06 06 06 06 06 06 06 06 06 06 06

16
d18O-O2 in Oligotrophic Surface Ocean

ALOHA Surface Layer
0.70

0.60

0.50
del18O(‰)

0.40

0.30

0.20

0.10
0.0      0.2     0.4      0.6       0.8        1.0       1.2     1.4    1.6      1.8    2.0
Biological O2 Sat'n (%)

Mar 06    May 06    Jun 06      Jul 06         Aug 06    Oct 06   Nov 06    'Feb 07
'Mar 07   'May 07   'Jun 07     'Dec 06        Equil

The overall range (variability) in d18O (0.3 ‰) and O2
saturation (1%) is much smaller than in freshwater systems
because of lower photosynthesis rates and higher air-sea
gas exchange rates.                                                                                                  17
Diurnal d18O-O2 Cycle in Coastal Ocean

(vs AIR)

Sagami Bay, Japan (Sarma, 2005)    18
Depth Trends in d18O and O2sat at ALOHA

d 18O vs AIR (per mil)                                 Fraction O2sat
-1.0   0.0     1.0    2.0   3.0   4.0   5.0                 0.760   0.840     0.920        1.000
0                                                            0

50                                                           50

100                                                          100
Depth (m)

Depth (m)
150                                                          150

200                                                          200

250                                                          250

Feb 02                                                 Feb 02
300             Oct 02                                       300       Oct 02
Feb 03                                                 Feb 03
350                                                          350

19
d18O vs O2sat Trend in Thermocline of the
Pacific Ocean
Pacific Ocean (1972-78)
20
18
del18O (o/oo vs AIR)

16
14
12
10
8
6
4
2
0
0.00   0.20      0.40     0.60     0.80   1.00   1.20
O2/O2sat

Red = Rayleigh predicted KIE of 0.9945 for respiration                          20
Calculated KIEfor Respiration

-5
Pacific Ocean
-8
KIE respiration (‰)

-11

-14

-17

-20

-23
0   1000   2000     3000      4000    5000       6000
Depth (m)

Assumes open system at steady-state.                          21
Oxygen Cycle: Use of Triple Isotopes
• A mass independent reaction during ozone
production in the stratosphere causes an
anomalous isotopic composition of atmospheric
O2 (and CO2).
• This O2 isotopic anomaly is a very useful
tracer to estimate photosynthesis (productivity)
rates on land and in aquatic systems (ocean,
lakes, rivers, etc.).
• Potentially, this method could make a
significant impact on our understanding of the
ocean’s biological pump
22
Anomalous d17O and d18O Composition of
Atmospheric O2 and CO2
Lab Experiments                     Field Measurements

2O2 + energy  O3 + O(1D)               O(1D) + CO2  CO2 + O
Result: Small amount of O2 (CO2) in stratospheric has an
23
anomalously low (high) 17O/18O. This O2 mixes into troposphere.
Isotopic Notation for 17O Anomaly
• Express the 17O/16O anomaly using 17Δ notation
17Δ =   (d17O – 0.518*d18O)*1000
• Units are per meg, 1 per meg = 1 ‰ / 1000
• AIR is the standard and has a 17Δ = 0 per meg
• Since air is depleted in 17O/16O, most other species
will have positive 17Δ values on this scale
• The coefficient of 0.518 was chosen to equal the
slope of d17O vs d18O observed during respiration.
(Luz and Barkan, 2000)
24
Slope of d17O vs d18O during Respiration

25
17Δ   of O2 in water equilibrated with Air

(Luz and Barkan, 2003)           (Sarma et al, 2006)
26
17Δ   of Photosynthetic O2
Lab Experiments        17Δ (per   meg vs AIR)
Marine Plankton               244±20; 252±5
Sea of Galilee Plankton       159±10
Puget Sound Plankton          ~ 200

27
Ocean Range of 17Δ Values

Purely Photosynthetic O2           249 per meg

Half Photo + Half Atmos O2                 130 per meg

16 per meg
Purely Atmospheric O2

Measuring 17Δ yields a direct estimate of the
28
proportion of O2 from air and photosynthesis.
Measured 17Δ in the Surface Ocean
17Δ (per   meg)
•   Oligotrophic N. Pacific (Quay)            20-40
•   Oligotrophic N. Atlantic (Luz)            30-50
•   Southern Ocean (Hendricks)                20-50
•   Equatorial Pacific (Hendricks, Juranek)   50-90
•   Sagami Bay (Sarma)                        80-100
•   California Current System (Munro)         25-100
•   Sea of Galilee (Luz and Barken)           100-140

29
Near Hawaii                                                                  Near Bermuda
 17 (per meg)                                                                17 (per meg)

0
0
0
0
0

0
0
0
0
0
20
40
60
80
10
12
14
16
18
20
40
60
80
10
12
14
16
18

0
0

0                                                                            0

50                                                                           50

100                                                                          100

depth (m)
depth (m)

150                                                                          150

200                                                                          200
21-Feb-02                                                               13-Mar-00
250                                                                          250                             10-Jul-00
25-Feb-03
13-Jun-01                                                               16-Oct-00
300                                                                          300
6-Oct-02                                                                 25-Jan-01
350                                                                          350

L. Juranek (U.Washington)                                                         B. Luz (Hebrew U.)          30
Mixed Layer O2 and 17Δ*O2 Budget

• dO2/dt = kam*Sol*pO2atm – kam*Sol*pO2ml + Photo – Resp
-where kam = air-sea gas transfer rate and Sol=O2 solubility
• d(17Δ*O2/dt) = kam*Sol*pO2atm*17Δair – kam*Sol*pO2ml*17Δdiss +
Photo*17Δphoto – Resp*17Δdiss
-assume respiration doesn’t change the 17Δ of the dissolved O2
-ignore mixing and advection fluxes for now
• Substituting for kam*Sol*pO2ml yields an expression for gross
Photo:
Photo = kam*pO2atm*Sol*(17Δair – 17Δdiss)/(17Δphoto – 17Δdiss)

31
Estimating gross Photosynthesis rates from 17Δ

• If one estimates air-sea O2 gas transfer rates (kam)
from wind speed measurements, then one can calculate
the gross Primary Production (PPg) rate from a single
measurement (17Δ of dissolved O2)

PPg = kam* O2sat * (17air – 17diss)
(17diss – 17photo)

32
Advantages of 17Δ-PP over 14C-PP Method
a. In situ PP rates not in vitro PP rates
-there are no bottle effects.
b. Much simpler field method
-no time consuming bottle incubations
c. Integrates over the lifetime of O2 in the mixed layer
-typically 10-20 days (i.e., 50-100m / 5m/d)
d. Measures gross PP rates
-not an ambiguous rate between gross and net PP
-recycling of 14C-labeled OC in the bottle and use
of non-14C labeled CO2 during photosynthesis yield
biases in PP rates that are difficult to quantify     33
a. Measures gross PP rate integrated over the mixed layer depth,
not the photic layer depth.
b. Uncertainty of method is significant and depends primarily on
uncertainty of gas exchange rate (30%) and 17Δ measurement.
c. Need to convert from O2 production to organic carbon
production
-a 10-20% reduction for Mehler reaction and photorespiration
-divide O2 production by the Photosynthetic Quotient (PQ) of
~1.1 (NH4 based PP) to ~1.4 (NO3 based PP)
d. In some situations, upwelling, mixing or entrainment can bias
the 17Δ in the mixed layer causing an overestimation of gross PP.
34
17Δ   Gross PP rates in the Surface Ocean
Gross PP (mg C m-2 d-1)

•   Oligotrophic N. Pacific (Juranek) 800 - 1500
•   Oligotrophic N. Atlantic (Luz)    300 - 1000
•   Southern Ocean (Hendricks)        600 - 3000
•   Equatorial Pacific (Juranek)      1000 - 2000
•   Sagami Bay (Sarma)                1500 - 3000
•   California Current System (Munro) 100 - 3000
•   Sea of Galilee (Luz&Barkan)       1600 – 16000
•   Global Ocean (at 1 gmC/m2/d)      130 PgC/yr

35
Comparison of 17O-PPg versus bottle 14C-PP

BATS and HOTS = 1.6±0.4; CalCOFI = 2.7±1.6   36
Estimating the ratio of net to gross PP
• Photo = kam*pO2atm*Sol*(17Δair – 17Δdiss)/(17Δphoto – 17Δdiss)

• dO2/dt = kam*pO2atm*Sol*(1 – pO2/pO2atm) + Photo – Resp
-assuming net community productivity (NCP) = gross
Photosynthesis – total Respiration and substituting for
kam*pO2atm*Sol yields:
NCP/ Photo = (O2/O2atm – 1)* (17Δphoto – 17Δdiss) / (17Δair – 17Δdiss)

• the NCP/PPg ratio yields an estimate of the efficiency of organic
carbon recycling in the ocean
-if all photosynthetically produced organic carbon was      37
respired to CO2 in the mixed layer then NCP/PPg = 0
Estimates of NCP/PPg from 17Δ and O2/Ar
Measurements

38
Ratio of NCP/PPg in Surface Ocean
-at HOT and BATS:              0.13±0.03
-Southern Ocean:               0.17±0.13
-Equatorial Pacific:           0.12±0.11
-California Current            0.16±0.12

• Coastal Ocean has NCP/PPg ratio that is similar
to open oligotrophic ocean. (Unexpected).
• Could be our most accurate estimate of the
efficiency of ocean’s biological pump.
39
Estimates of Carbon Export (NCP) Rates
-at HOT and BATS:              10±5 mmols C m-2 d-1
-in the Southern Ocean:        13±4
-in the Equatorial Pacific:    6.9±6.2
-California Current (CalCOFI): 14±10

-Globally, at 10 mmols m-2 d-1, yields 16 Pg C/yr
(higher than previous estimates of 6-10 Pg C/yr)

40
Future of 17Δ and O2/Ar Ocean Research

• Improved ability to detect PP events.
• Applicable to obtain large scale synoptic surveys
of ocean PP rates.
• Improve resolution of short spatial and
temporal scale variability in marine PP in certain
regions (e.g., coastal).
• Useful for validation of satellite PP rates.

41
Basin Scale Trends in
17Δ-PPg in Pacific Ocean

(using a container ship as
sample collection platform)

5000                                                    August 2005
100
C-based sat PP

 GPP and sat. PP (mg C m d )
4500

-2 -1
90         Aug 2004                                                                                                                                    VGPM sat PP
80         Feb 2005                                                                          4000                                                      in situ GPP
Aug 2005                                                                                                                                    Chl*10^4
70                                                                                           3500
 17O (per meg)

60                                                                                           3000
50                                                                                           2500
40                                                                                           2000
30                                                                                           1500
20
1000
10
500
17

0
-40        -30   -20   -10          0   10   20   30
0
-45   -35   -25   -15        -5       5     15     25          35
Latitude
Latitude along transect               42

```
To top