Application of radiogenic isotopes SrNdPb and Useries to
Application of radiogenic isotopes (Sr-Nd-Pb) and Useries to erosion Dr. Anthony Dosseto GEMOC, Macquarie University 26 -Apr-2005
I. Radiogenic isotopes Sr-Nd-Pb
I. Radiogenic isotopes A. Strontium
• 87 Rb -> 87 Sr (T 10 yr) = 4. 97 x 10 1/2 • 86 Sr stable, non-radiogenic • 87 Sr/86 Sr records Rb-Sr fractionation on time scales > 100 Ma • Rb and Sr do not fractionate significantly during weathering • 87 Sr/86 Sr in river waters and sediments reflect different source contributions
I. Radiogenic isotopes A. Strontium 1. In the dissolved load of rivers
Sr-isotopes in waters • Sr is soluble • Abundant in carbonates (Ca, Mg, Sr similar behavior) • 87 Sr/86 Sr carbonates = 0. 708 • Carbonates are weathered ~ 10 times faster than granites • 87 Sr/86 Sr in waters dominated by carbonates (Gaillardet et al. , 1999)
Sr-isotopes in waters • Ratios in the dissolved load lower than in particles • Contribution of carbonates to the dissolved load but not to the suspended load (Allegre et al. , 1996)
Sr-isotopes in waters • With major, trace elements, Sr-isotopes used to quantify contribution of different lithologies (Gaillardet et al. , 1997)
I. Radiogenic isotopes A. Strontium 2. In the suspended load of rivers
• Suspended particles = silicates • 87 Sr/86 Sr particles provide information on the silicic bedrock… the continental crust
Sr-isotopes in suspended particles (Allegre et al. , 1996)
I. Radiogenic isotopes B. Neodymium
Nd-isotopes • 147 Sm -> 143 Nd (T 10 yr) = 4. 97 x 10 1/2 • 144 Nd stable, non-radiogenic • 143 Nd/144 Nd records Sm-Nd fractionation on time scales > 100 Ma • Sm and Nd do not fractionate significantly during weathering • 143 Nd/144 Nd in river waters and sediments reflect composition of the bedrock
Nd-isotopes • Nd is insoluble and, in the dissolved phase, linked to colloids • In particles, as Sr-isotopes, 143 Nd/144 Nd provide information on the continental crust, the substratum of the basin
Nd-isotopes
I. Radiogenic isotopes C. Lead
Pb-isotopes • 238 U -> … -> 206 Pb (T • • 9 yr) = 4. 51 x 10 1/2 235 U -> … -> 207 Pb (T = 7. 13 x 108 yr) 1/2 232 Th -> … -> 208 Pb (T = 1. 41 x 1010 yr) 1/2 204 Pb stable, non-radiogenic 206 Pb/204 Pb, 207 Pb/204 Pb record U-Pb fractionation, 208 Pb/204 Pb Th-Pb fractionation on time scales > 100 Ma
Pb-isotopes • Th and Pb are insoluble and do not fractionate during weathering => 208 Pb/204 Pb particles = bedrock, continental crust • In oxidizing conditions, U is more soluble than Pb • 206 Pb/204 Pb and 207 Pb/204 Pb are modified by weathering only on time scales > 100 Ma
Pb-isotopes • Pb-isotopes in particles can be used: – to constrain the U-Th-Pb composition of the continental crust eroded – to identify different sources contributing to a river
II. Uranium-series A. What is it?
Uranium-series ? What is it ?
Uranium-series ? What is it ?
Uranium-series ? What is it ?
Uranium-series • For a system closed for more than 1 Ma, all the radioactive systems are in secular equilibrium : Parent activity = daughter activity lp. Np = lf. Nf where l and N represent the radioactive constant and number of atoms respectively. e. g. , (226 Ra/230 Th) = 1 (parentheses denote activity ratios)
Uranium-series Daughter/parent activity ratio • Geological processes induce fractionations within the U-series = radioactive disequilibria • e. g. , (226 Ra/230 Th) > or < 1 Secular equilibrium Time elapsed since fractionation (yr)
Uranium-series Daughter/parent activity ratio • Once disequilibrium is produced, every system goes back to secular equilibrium by radioactive decay. • This phenomena is characterized by a time scale function of the daughter nuclide half-life Secular equilibrium Time elapsed since fractionation (yr)
Uranium-series • Several radioactive systems studied: – 238 U-234 U (1 Ma) – 238 U-230 Th (300 ka) – 230 Th-226 Ra (10 ka) • Various time resolutions to study a single process
Why use U-series to study erosion? • How long to develop soil profiles ? • How long to transfer sediments from mountains to the ocean ? • Weathering withdraws CO 2 from the atmosphere… How long would it take to absorb a sharp increase in atm. CO 2 ? • Need of a tool time-sensitive on time scales characteristics of erosion (1 – 1000 ka)
Why use U-series to study erosion? • U-series fractionate during weathering • Radioactive disequilibria are time dependent • They record weathering-related fractionation up to 1 Ma • Bedrock age > 1 Ma secular equilibrium bedrock composition and initial conditions for weathering are known
II. Uranium-series B. Behaviour of U-series isotopes during weathering
Mobility of radionuclides on Earth’s surface • • In oxidizing conditions, U 6+ -> mobile Reducing conditions, U 4+ -> immobile During weathering: U 6+ In organic-rich sediments or soil horizons, in aquifers with reducing conditions: U 4+ • Ra: family of Ca, Mg, Ba -> Ra 2+, mobile • Th: immobile, except in presence of organic matter (good complexant)
How do U-series fractionate during weathering? • U and Ra are generally more soluble than Th • Waters enriched in U and Ra • Residues of weathering (sediments, soils) depleted in U and Ra over Th
How do U-series fractionate during weathering?
How do U-series fractionate during weathering? • When 238 U -> 234 Th + a 234 Th can be ejected from the solid (recoil effect) • 234 Th -> 234 U + 2 b- (T 234 Th = 24 j) • Residual solids depleted in 234 U over 238 U
II. Uranium-series C. Application of U-series to erosion 1. Time scale for the development of soil profiles
Soil age and weathering rates v. Formation and evolution of weathering profiles v. U/Th Dating of pedogenic concretions v. Depth variation of U-series nuclides in weathering profiles
Dating of pedogenic concretions Closed-system evolution of an initial Th-free material : 234 U/238 U-230 Th/234 U AGE OF THE SYSTEM Presence of initial detrital U and Th in the samples ? CORRECTION PROCEDURES
Correction procedure Mixture of two isotopically homogenous end-members with one containing only U isotopes at the time of deposition (230 Th/232 Th) (234 U/232 Th) 20 15 10 5 0 0 5 (238 U/232 Th) 10 3 2 1 0 0 5 10 15 20 (234 U/232 Th) Slope : (234 U/238 U) and (234 U/230 Th) of the Th-free endmember
Determination of weathering rates …. and reality ! Theory. . . (230 Th/ 238 U) 5 10 15 0. 5 1. 0 1. 5 2 Radioactive decay 0 (230 Th/ 238 U) 0 WEATHERING RATE 0. 5 1. 0 1. 5 0 0. 5 1. 0 1. 5 2 5 5 10 10 15 15 Burkina-Faso (Dequincey et al. , 2002) 20 U-Th fractionation 20 Modelling of U series nuclides mobility within weathering profiles
• U loss from the bedrock to the top • Progressive U accumulation from the top to a maximum depth ZF U(ppm) (230 Th/238 U) Cameroon (Boulad et al. , 1977)
comparison of 238 U-234 U and 234 U-230 Th chronometers 2. 0 e (234 U/238 U) Radioactive decay 1. 5 u h 0 T 23 eq ilin - 4 U 23 U accumulation zone 1. 0 "Forbidden zone" 0. 5 0. 0 U leaching zone 0. 5 1. 0 1. 5 (230 Th/238 U) 2. 0
238 U-234 U-230 Th method African lateritic profile : Kaya toposequence, Burkina-Fasso (Dequincey et al, 2002) (234 U/238 U) (230 Th/ 238 U) (234 U/238 U) 0. 5 1. 0 1. 5 0. 9 1. 0 1. 1 0 1. 0 5 5 0. 9 0. 5 10 10 15 15 1. 0 (230 Th/238 U) 1. 5 Recent. U gain Recent. U loss Recent U gains and losses in each level
• Dequincey et al. (2002) : time scale for erosion > 100 ka for African laterites
II. Uranium-series C. Application of U-series to erosion 2. Source of radionuclides in rivers
Origin of radionuclides in rivers Example of the Upper Rhine system Riotte et Chabaux, 1999 Durand et al. , 2003
Rainwater contribution Variation of (234 U/238 U) during the year at one sampling station During the year Flood event Secular equilibrium Negligible effect of rainwater on (234 U/238 U) signature of waters
Lithological control on (234 U/238 U) Groundwater Combining U-Sr signature allows quantification of these fluxes
Origin of (234 U/238 U) in river water (Durand et Chabaux, EUG-EGS-AGU, Nice 2003) Deep water influence on U-Sr budget
Deep groundwater (234 U/238 U) 4 Lauter 3 Superficial groundwater 2 1 Alluvial groundwater 0. 708 0. 710 0. 712 0. 714 0. 716 87 Sr/86 Sr different levels of water rocks interactions (surface weather groundwaters. . . ) contributing to the chemical fluxes of river
II. Uranium-series C. Application of U-series to erosion 2. Time constraints on erosion processes at a basin scale: The Amazon basin
Study of the Amazon basin a) Presentation of the study b) Time scale of erosion for highland rivers c) Transfer time of sediments from the Andes to the tropical plain d) Is erosion in the Andes in steady-state ? e) Conclusions
The Amazon basin • The world’s biggest watershed (6. 106 km 2) • Average discharge: 20. 106 m 3/s (Molinier et al. , 1995) (1 st source of fresh water to the oceans) • Sediment flux: 600 -800 Mt/yr (Filizola, 2004) (3 rd biggest flux)
The Amazon and main tributaries « Timescale and conditions of weathering under tropical climate: Study of the Amazon basin with U-series » (2005) A. Dosseto, B. Bourdon, J. Gaillardet, C. Allègre, N. Filizola
Morphostructures of the basin
Bolivian Amazon Rio Madeira basin « Timescale and conditions of chemical weathering in the Bolivian Andes and their foreland basin » (in prep) A. Dosseto, B. Bourdon, J. Gaillardet, C. Allègre, L. Maurice-Bourgoin
From the Andes to the tropical plain
Study of the Amazon basin a) Presentation of the study b) Time scale of erosion for highland rivers
Time scale of erosion for the Amazon and Solimões • Amazon/Solimões : (230 Th/232 Th) similar to average continental crust • Erosion of pristine crust over time scale < 10 ka (Th-U fractionation too recent to have modified (230 Th/232 Th) ratios)
Continuous weathering (Vigier et al. 2001) • Continuous leaching of soluble elements (U, Ra) • Duration of weathering Residence time of particles
Time scale of erosion for the Amazon and Solimões • Solimões and the Amazon at Óbidos : T ~ 4 -6 ka
Time scale of erosion for the Amazon and Solimões • Amazon before Madeira (Amazon 07) : T = 70 ka • Higher value because contribution of Rio Negro particles (T ~ 500 ka)
Time scale of erosion for Rio Madeira • Madeira : T ~ 150 ka
Time scale of erosion for Rio Madeira • Multi-step erosion – Production of sediments during 1 st event – Storage for T > 300 ka (230 Th decay) / Formation of sedimentary rock – Erosion of the sedimentary rock during 2 nd event
Time scale of erosion for Rio Madeira • Bedrock currently eroded : sedimentary rock or pristine crust ? • Reproduce 238 U-230 Th-226 Ra disequilibria of particles : no assumption on bedrock composition
Time scale of erosion for Rio Madeira • Bedrock U/Th < continental crust • Current erosion of bedrock depleted in U over a previous erosion cycle
Time scale of erosion for Rio Madeira • Multi-step erosion / Recycling of sediments throughout the history of the basin
Summary – Time scale of erosion for highland rivers • Solimões/Amazon system – Erosion of a pristine crust – T = a few ka • Madeira – Erosion of sedimentary rocks – Recycling of sediments (multi-step process) – T ~ 20 ka (at the confluence with the Amazon)
Study of the Amazon basin a) Presentation of the study b) Time scale of erosion for highland rivers c) Transfer time of sediments from the Andes to the tropical plain
• 230 Th excess in particles, increases from the Andes to the plain • Cannot result from the contribution of plain tributaries • Increasing transport time of suspended particles
Transfer time of sediments through the Madeira basin • Transfer time of suspended particles ~ 15 ka T = 3 -4 ka (Andes) 20 ka (confluence with the Amazon) • Short time scale of erosion in the Andes, T = 3 -4 ka, comparable to what we get for Solimões/Amazon
Study of the Amazon basin a) Presentation of the study b) Time scale of erosion for highland rivers c) Transfer time of sediments from the Andes to the tropical plain d) Is erosion in the Andes in steady-state ?
• Steady-state erosion : Mr = Mw. TDS + Mp+ Ms [U]r. Mr/Mw = [U]d + [U]p. P + [U]s. Ms/Mp. P
Steady-state erosion ? • 230 Th-238 U can be used to predict the suspended particle concentration, P, compatible with a steady-state erosion
Steady-state erosion in the Andes ? No ! • Rivers draining the Andes : observed concentrations >> predicted • They export more sediments than predicted by steadystate erosion
Why ? • Using major and trace elements, Gaillardet et al. (1997) made the same observation earlier • They suggest 2 possible explanations : – The soils in the Andes are currently destroyed more rapidly than they are produced – They consider that the bedrock composition is an average continental crust. If considering bedrock depleted in soluble elements (eroding sedimentary rocks), the discrepancy between predicted and observed particle concentrations can be resolved
• With U-series, the bedrock composition is known : secular equilibrium • The discrepancy between predicted and observed particle concentrations must be explained by the destruction of Andean soils
• Recent phenomena (<10 ka), because of the low thickness of the soils in the Andes • Observation of short time scales for erosion in the Andes supports this hypothesis of a recent change in erosional regime • Where comes this change from ?
Climatic perturbation modification of the erosional regime • Several palaeoclimatic studies have shown an increase in precipitation since ~ 5 ka (Rigsby et al. , 2003; Servant et Servant-Vildary, 2003; Tapia et al. , 2003) • This increase may explain why the amount of sediments exported is higher than predicted by steady-state erosion
Climatic control on erosion • Erosion in the Andes out of equilibrium = response to recent climatic changes • In agreement with theoretical work of Whipple (2001): – Rapid climatic fluctuations (1 -100 ka) may prevent the accomplishment of steady-state; – The time scale of tectonic perturbation (10 Ma) is large compared to the response time of erosion
Study of the Amazon basin a) Presentation of the study b) Time scale of erosion for highland rivers c) Transfer time of sediments from the Andes to the tropical plain d) Is erosion in the Andes in steady-state ? e) Conclusions
Conclusions (1/3) • Solimões/Amazon sytem : erosion of pristine continental crust over T ~ few ka • Madeira : – Erosion of bedrock depleted in soluble elements (sedimentary rock) – Existence of more than 1 erosion cycle
Conclusions (2/3) • It is possible to constrain the transfer time of suspended matter within a watershed • Example of Rio Madeira : T ~ 15 ka
Conclusions (3/3) • Erosion in the Andes – T = a few ka – out of equilibrium • Observations related to recent climatic changes — Erosion responds rapidly to high frequency external forcing processes
- Slides: 85