Reconsidérer les flux continent-océan Catherine Jeandel& Eric Oelkers (2014) CNRS, LEGOS, Observatoire Midi-Pyrénées, Toulouse University CNRS, GET, Observatoire.

Reconsidérer les flux continent-océan

Catherine Jeandel& Eric Oelkers (2014)

CNRS, LEGOS, Observatoire Midi-Pyrénées, Toulouse UniversityCNRS, GET, Observatoire Midi-Pyrénées, Toulouse University

Seminaire LEGOS, 7 novembre 2013

Thank you to: K. Tachikawa, F. Lacan, B. Peucker-Ehrenbrink, M. Jones, C. Pearce, M. Grenier

Weathering Processes….

Chemical Weathering

Ocean

Dissolved load

Mechanical Weathering or

Denudation

Particulate Transport

Global mass fluxes to the ocean

0

5

10

15

20

25

Dissolved flux(Gaillardet et al.,

2003)

Bedload Flux(Walling, 2006)

SuspendedMaterial Flux

(Syvitiski et al.,2003)

Ma

ss

flu

x t

o t

he

oc

ea

ns

(G

T/y

r) Solid flux = 30 times the dissolved one= 50 times the atmospheric one

eNd oceanic distribution follows the general circulation(in the modern ocean as in the sediment)But its concentration doubles only

Along the general oceanic circulation

-15

Lacan et al., Chem. Geol., 2012

Nd paradox

Field Observations (1)

4

CHUR144

143SAMPLE

144

143

Nd 101

Nd

Nd

Nd

Nd

-3.8

Nd concentration isonly doubling

Field Observations (2)Imbalanced Nd oceanic budget

Dust + river inputs : not sufficient

Missing term: river solid load and/or sediments deposited on the marginsLacan&Jeandel, 2001;Tachikawa et al, 2003

Field Observations (4)

« BOUNDARY EXCHANGE » - quantifiable using Nd IC- invisible with concentrations only

FR FA FBE

BE is THE major source term (>95% of the total : 1.1 1010 g(Nd)/an).

Continental inputs & margins: major for Nd(only 3% dissolution)

And for the other chemical elements?

Lacan &J eandel, 2001, 2005Van der Flierdt et al, 2004Carter et al, 2012Rickli et al, 2009, 2011Grasse et al, 2012Wilson et al, 2012Grenier et al, 2013…/…

Arsouze et al, 2009

Strontium

• Ocean 87Sr/86Sr is homogenous at 0.70916

• The primary controls were accepted as dissolved riverine input (87Sr/86Sr ≈ 0.7136) and hydrothermal exchange (87Sr/86Sr ≈ 0.7029)

• However, the unradiogenic flux is a factor of 3 too low to balance the inputs

• Therefore, part of the story is missing (as for Nd oceanic budget)

Field Observations (5)

Strontium Imbalance: hypotheses

• Current riverine fluxes may be elevated due to a recent glaciation (Vance et al., 2009)

• Subsurface weathering of volcanic islands are not included (Allègre et al., 2010)

• Volcanic particulate riverine material could also contribute unradiogenic Sr

Field Observations (6)

Sampling in Borgarfjörður estuary(Morgan Jones’ current work)

Field Observations (7)

Field evidence of particulate dissolution in the ocean: Borgarfjordur Estuary (1)

0

2000

4000

6000

8000

10000

12000

0 2 4 6 8 10 12 14

Distance, km

[Cl]

pp

m

0.704

0.705

0.706

0.707

0.708

0.709

0.71

0 2 4 6 8 10 12 14

Distance, km

Sr8

7/8

6 e

stu

ary

wa

ter

Estuary water as a function of salinity Sr87/Sr86 as a function of salinity

Field Observations (8)

Field evidence of particulate dissolution in the ocean: Borgarfjordur Estuary (2)

-0.0001

0

0.0001

0.0002

0.0003

0.0004

0.0005

0.0006

0 5 10 15

Distance, km

Sr8

7/86

( fl

uid

mix

ing

mo

del

- e

stu

ary

wat

er)

Sr87/Sr86 in estuary water as a function of river water fraction

Sr87/Sr86 attributable to particulate dissolution

0.704

0.705

0.706

0.707

0.708

0.709

0.71

0 0.2 0.4 0.6 0.8 1

Fraction river content of estuary water

Sr8

7/8

6 e

stu

ary

wa

ter

50% of Sr from dissolution of particulates

Field Observations (9)

As for Nd, dissolution of lithogenic material is occuring at the land/ocean interface

Silica budget (Mediterranean Sea) 106 Moles/y

Durrieu de Madron et al, 2009

51,000-127,000

288,000- 424,000

Siout = 2 to 8 x (Si in)Total ext. inputs = 25,000-126,000

Si missing = 35,000-348,000

Total dissolved river input = 24,300-118,700

Total Atmospheric input =709-7,350

Field Observations (11)

Source of the Si missing?Range required: 3.5 1010 à 3.4*1011 moles/y

Total river solid discharge = 0.73 109 T/y (Ludwig et al , 2003)

Dissolution of only 1% of this flux

8.3 1010 mol/y SiO2

Same result considering

Release from 1% of the first 10 cm of sediment deposited on the margins (consistent with Tréguer and de la Rocha, 2012)

Field Observations (12)

Fe Particulate/Dissolved flux

Particulate transport dominatesthe flux of most metals to the ocean(Oelkers et al, 2011)

The release in seawater of even a tiny fraction (~1%) of this material deposited on the shelf/margins mayimpact oceanic isotopic & element budgets (Jeandel et al, 2011, Jeandel&Oelkers, 2014)

Fe

Nd

Si

Sr

Field Observations (14)First Intermediate conclusion

Th

Particulate/dissolved flux

Questions

• Processus: Desorption or Dissolution?

• Which phase (Fe-Mn coating) or mineral (primary or secondary)?

• Kinetic?

Batch experiments(Jones et al, 2012a,b; Pearce et al, 2013)

Sediment samples were mixed with open ocean seawater (water of southern origin, eNd = -9.6)

• Marine top core from Kerguelen plateau (station C1, eNd = -1.4)

• Estuarine sediment from SW Iceland (eNd = +7.2)

• Riverine bedload Sediment from SW Iceland (eNd = +7.5)

30 l SW1g/l4 monthsSr, REE, Si aliquots taken weekly

Nd IC aliquots taken every 2 weeks

Batch reactors

-9.6

-1.4 to +7.5

Experiments (1)

Jeandel et al; Traces and tracers, Liège, 2011

Hours

-10

-5

0

5

10

0 1000 2000 3000 4000

Duration (Hours)

N

dExp A

Exp B

Exp C

1 month

Initial Sw value = -9.6

-1.4 Kerguelen

Iceland Riv. 7.2

Hours

Iceland Est.

ISOTOPENd results (Pearce et al, 2013)

0

10

20

30

0 1000 2000 3000 4000

Duration (Hours)

Nd

(pm

ol/k

g)

Exp A

Exp B

Exp C

CONCENTRATION

IC: Nd rapid release (0.2 to 2% of Nd contained in the KER basalt)

Cc: Nd scavenging after release: secondary phases =

REE phosphatemineral rhabdophane (REE(PO4)nH2O) ?

Experiments (4)

Silica release?

0 1000 2000 3000 4000 5000 6000 70000.000

1.000

2.000

3.000

4.000

5.000

6.000

7.000

BR1a (Riverine) - 5 °C BR13a (Estuarine) - 5 °C BR1a (Riverine) - 25 °C BR13a (Estuarine) - 25 °C

Time (hrs)

[Si]

(p

pm

)

SW Initial concentration = 0.3 ppm

1 month

5°CSi(OH)4 x 10

25°CSi(OH)4 x 20

Experiments (6)

Experiments (7)

Experiments (8)

OlivinePlagioclase

Second Intermediate conclusion

* Particulate material is chemically reactive, olivine & plagioclase being most susceptible to alteration

* Dissolution of lithogenic material followed by precipitation of secondary phases, is strongly suspected

* Trace elements (Nd, Sr, Mn, Ba, Ni ) & major ones (Si, Mg) are implied

* Kinetic is rapid (weekly to monthly scale)

Jeandel et al; Traces and tracers, Liège, 2011

Other elements that are released(similar experiments not shown here)

10 100 1000 100000.1

1.0

10.0

100.0

1000.0

10000.0

SiLiBMgKCaMnNiRbSrBa

Rela

tive

con

cent

ratio

n in

sea

wat

er

Si

Ba

Ni

Mn

Conclusions & Consequences (1)

Conclusions

Weathering of the particulate load (suspended, bedload, margin sediments)can significantly affect the oceanic budget of isotopes & elements

This processus -previously unconsidered- might help to balancethe marine Sr, Nd, Si cycles and likely Th, Ba, Mg…Ni, Cu, Zn, Mn: Cameron, Little, Vance (all published in 2014)

The weathered fraction includes a significant lithogenic component (feldspath and olivine being among the most sensitive)

Kinetic of this process is rapid (within a month)

Followed by secondary phase precipitation: might affect more the isotopic than the concentration budgets

Conclusions & Consequences (2)

Consequences

Marine Sr cycle: differences in the reactivity of basalticand crustal particles suggest that solid load weatheringmight help account for part of the « missing Sr flux »to the ocean. Impact on the geological reconstitutions?

Marine Nd, REE cycles: consistent with the « Boundary Exchange » hypothesis. Impact on its use as paleo-tracer?

Marine Th isotopes: solid load weathering likely perturbates the coastal 232Th/230Th. Impact on its use as particle dynamic tracer?

Conclusions & Consequences (3)

Consequences

Marine Si concentrations and isotopes: This additional flux increases the Si input term, diminishes the Si residence time by 35 to 50% (at steady-state; Jeandel et al, 2011; Tréguer & de la Rocha, 2012)

Impact on the biological pumpImpact on the CO2 modern cycleImpact on the climatic models

Conclusions & Consequences (4)

(many) Remaining Questions

Other elements?Clearly, one should reconsider the land to ocean fluxes of all the elements, mostly those of the essential marine micro-nutrient

Particle speciation effect?Working on this will allow the best modelling of the impact ofsuch release on a global scale, taking into account the geochemical particularities of watersheds and margins…

Thank you!