Scholarly article on topic 'Weathering Intensity in Lowland River Basins: From the Andes to the Amazon Mouth'

Weathering Intensity in Lowland River Basins: From the Andes to the Amazon Mouth Academic research paper on "Earth and related environmental sciences"

CC BY-NC-ND
0
0
Share paper
Keywords
{"Weathering rates" / "Erosion rates" / Floodplains / "Amazon Basin"}

Abstract of research paper on Earth and related environmental sciences, author of scientific article — Julien Bouchez, Jérôme Gaillardet, Friedhelm von Blanckenburg

Abstract Actively eroding mountains supply un-weathered material into large river basins. It is still not known whether these un-weathered minerals undergo significant chemical weathering during storage in continental alluvial deposits within the surrounding lowland areas. Here we use previously reported weathering and erosion fluxes of rivers from the Amazon Basin to assess this effect. We show that the fraction of total denudation (weathering plus erosion) occurring as dissolved export experiences only a slight increase during transfer through the lowlands. The overall low weathering intensity of Andean sediments throughout the Amazon basin is attributed to the fact that source rocks are recycled meta-sedimentary rocks.

Academic research paper on topic "Weathering Intensity in Lowland River Basins: From the Andes to the Amazon Mouth"

Available online at www.sciencedirect.com

ScienceDirect

Procedia Earth and Planetary Science 10 (2014) 280 - 286

Geochemistry of the Earth's Surface meeting, GES-10

Weathering intensity in lowland river basins: from the Andes to the

Amazon mouth

Julien Boucheza*, Jérôme Gaillardetb, Friedhelm von Blanckenburgc

aInstitut de Physique du Globe de Paris, Sorbonne Paris Cité, CNRS, 1 rue Jussieu 75238 Paris cedex 05, France bInstitut de Physique du Globe de Paris, Sorbonne Paris Cité, Institut Universitaire de France, 1 rue Jussieu 75238 Paris cedex 05, France cGFZ German Research Centre for Geosciences, Helmholtz Centre Potsdam, Telegrafenberg, 14473 Potsdam, Germany, and Department. of

Geosciences, Freie Universität Berlin

Abstract

Actively eroding mountains supply un-weathered material into large river basins. It is still not known whether these un-weathered minerals undergo significant chemical weathering during storage in continental alluvial deposits within the surrounding lowland areas. Here we use previously reported weathering and erosion fluxes of rivers from the Amazon Basin to assess this effect. We show that the fraction of total denudation (weathering plus erosion) occurring as dissolved export experiences only a slight increase during transfer through the lowlands. The overall low weathering intensity of Andean sediments throughout the Amazon basin is attributed to the fact that source rocks are recycled meta-sedimentary rocks.

©2014The Authors.Publishedby Elsevier B.V.This is an open access article under the CC BY-NC-ND license

(http://creativecommons.Org/licenses/by-nc-nd/3.0/).

Peer-review under responsibility of the Scientific Committee of GES-10

Keywords: Weathering rates, Erosion rates, Floodplains, Amazon Basin

1. Introduction

Two main weathering regimes have been distinguished from relationships between chemical weathering and physical erosion rates that have been reported over a variety of spatial scales, from soils to large river basins1-4. (1) At low denudation rates (D, the sum of erosion rate E and weathering rate W), weathering and denudation rates

* Corresponding author. Tel.: +33(0)1 83 95 78 37. E-mail address: bouchez@ipgp.fr

1878-5220 © 2014 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license

(http://creativecommons.Org/licenses/by-nc-nd/3.0/).

Peer-review under responsibility of the Scientific Committee of GES-10

doi: 10.1016/j.proeps.2014.08.063

correlate with each other. This regime is one of long residence time of solids in the Critical Zone, in which | weathering rates (W) are limited by the supply of fresh minerals, hence by D, and is therefore called "supply-limited"3 (broadly similar to a "transport-limited"5'6 regime5'6). (2) At higher D, the solid residence time in the Critical Zone shortens and W attains a plateau in the W-D space as it becomes limited by the combination of reaction kinetics4'7-9 and chemical equilibrium (as runoff is finite)10. This "kinetically limited"4 regime is broadly similar to a "weathering-limited"5'6 regime. At the river catchment scale, W and D remain decoupled8, especially as mass wasting processes in active mountain belts produce un-weathered river sediment, thereby increasing erosion (E) and therefore D, but not W. It is still uncertain whether this decoupling holds true in large rivers systems, a question raised by the observation that river-scale W does not exceed soil-scale W, even in rivers passing through lowland areas surrounding active mountain belts8.

This question has remained unanswered to date because most data and concepts relating W and D are derived from observations made on soil-covered hillslopes or on small rivers draining eroding landscapes. We still know little about the chemical processes that solutes and sediments are subjected to once they are transported by river networks to the oceans through vast lowland areas in which particles are deposited and stored before being re-introduced again into riverine transport. The most striking features of these lowland areas are active floodplains that are made up of an assembly of meandering river channels, associated banks and bars, shallow lakes and seasonally inundated river deposits11. Over longer time scales, sediment can also be deposited in "inactive" alluvial formations that are involved in riverine transport again only episodically through channel migration or river avulsion. During their residence in lowland alluvial formations, sediment is left to further react with water and atmospheric gases such as CO2 and O212-14. How do these reactions modify the W-D relationship during transfer of sediments through lowland areas of large river basins? We tentatively answer this question by examining literature data on W-D across the Amazon Basin, from the Andean outlet to the river mouth.

•80° -75° -70° -65° -60° -55° -50° Altitude (m)

Fig. 1. Map of the Amazon Basin and location of the gauging stations. Open circles represent rivers draining mostly the Andes, black circles rivers draining mostly lowland areas, and grey circles rivers with Andean headwaters but also draining a large lowland area.

2. Source of data, methods, and concepts

| The Amazon is the world's largest river in terms of drainage area and water discharge15, and ranks amongst

the first for dissolved and sediment fluxes16. This basin also features a geomorphic distribution typical of large rivers | draining to passive margins. Headwaters drain the high, tectonically active Andes with altitudes of up to more than 6000 m, and high E and W. Downstream, the main tributaries flow across a vast foreland and lowland area17,18. Gauging stations at which W and E were measured are located in Fig. 1.

Here we focus on silicate-derived W, although carbonate weathering has been reported to occur in the Andean foreland12,14. Accurate values of silicate-derived W rely on disentangling the contribution of atmospheric

inputs, carbonate and evaporite dissolution, and possibly of anthropogenic activities19. This was done for the

12 18 20

Amazon Basin only in a handful of studies Values for E were taken from multi-annual time series of sediment gauging performed within the framework of the HyBAm (Hydrology of the Amazon Basin, http://www.ore-hybam.org) program21-24. Similar rates were obtained when using denudation rates derived from cosmogenic, in sifw-produced 10Be [ref. 25], as these two methods usually agree within a factor of 2 [ref. 26]. Total denudation rates D were calculated from D = E + W. The W/D ratio can then be calculated. This ratio is a metric for weathering intensity and is analogous to a chemical depletion fraction27. This W/D ratio is actually independent of discharge and drainage area, unlike D, W, or E taken individually. We acknowledge that the W/D ratio might be affected by a downstream change in E, as lowland areas can be loci of net sediment deposition / erosion, at least over short time scales21. Results are shown in Tab. 1.

W and D are commonly plotted against each other as area-normalized fluxes (Fig. 2). From the outlet of a mountain belt to the mouth of a large river, several end-member scenarios can be hypothesized, that result in distinct vectors in the W-D space (Fig. 2): (1) as the drainage area becomes larger, if sediment and dissolved fluxes remain the stable, both W and D decrease at a constant W/D ratio; (2) if sediment is trapped in the lowland area21, D decreases but W remains constant (if the lowland area supplies sediment D can increase); (3) if sediments from the mountain range undergo weathering in the lowlands, W increases but D remain constant. The two latter scenarios assume that the change in drainage area is small enough such that it is not driving the shift in D and W.

Tab. 1: Gauging stations locations, characteristics and weathering and denudation rates. For lowland reaches of Andean-fed rivers, the stations at which contributing tributaries are gauged are listed (see section 4)

Location/ Area Discharge W Ref. E Ref. W/D Contributing

Basin River Station km2 m3 s-1 t km2 y-1 t km2 y-1 stations

Napo Napo FO 12358 1263 73.3 12 631 23 0.10

Napo Coca SEB 5291 449 75.6 12 837 23 0.08

Marañon Marañon BOR 114237 4975 23.2 12 890 24 0.03

Marañon Huallaga CHA 68741 3042 25.8 12 710 24 0.04

Beni Beni RUR 69980 2153 14.5 12 3140 21 0.00

Mamoré Mamoré PVI 7886 532 29.1 12 1150 21 0.02

Mamoré Grande ABA 59311 360 7.6 12 2310 21 0.00

Jutaí Solimoes RJut 65000 3000 21.6 20 27 20 0.44

Iça Solimoes Riça 150000 6000 25.7 20 133.8 20 0.16

Juruá Solimoes RJur 186500 13000 23 20 97.2 20 0.19

Japurá Solimoes RJap 331000 27000 20.4 20 87.2 20 0.19

Purus Solimoes RPur 358000 106000 20.2 20 76.9 20 0.21

Purus Purus LAB 226552 5552 11.6 12 311.0 22 0.04

Negro Negro RNeg 468000 39000 8.5 20 8.2 20 0.51

Negro Negro Mouth 280000 16070 0.6 18 13.9 22 0.04

Beni Orthon CAR 33485 475 9.8 12 55 21 0.15

Mamoré Itenez VG 354022 1630 1.9 12 5 21 0.28

Tapajós Tapajós Itoba 387000 10780 10 18 11.0 22 0.48

Napo Napo BEL 100035 6489 31.6 12 463 23 0.06 FO,SEB

Marañon Marañon SR 357255 16885 19.2 12 470 24 0.04 BOR,CHA

Ucayali Ucayali JH 352593 12090 18.9 12 570 24 0.03 ATA

Amazonas Amazonas TAM 722089 30148 18.7 12 560 23 0.03 JH,SR

Solimoes Solimoes VGr 990000 52000 91 20 510.7 20 0.15 BEL,TAM

Solimoes Solimoes Manacapuru 2148000 98750 20 18 187.5 22 0.10 BEL,TAM,LAB

Beni Madre de Dios MIR 124231 5602 21.7 12 570 21 0.04

Mamoré Mamoré GUA 618220 7916 4.9 12 95.8 21 0.05 ABA,PVI,VG

Beni Beni CE 276350 9779 15.6 12 680 21 0.02 RUR+MIR+CAR

Madeira Madeira Mouth 1325000 31250 8.9 18 184.4 22 0.05 GUA+CE

Madeira Madeira RMad 1306000 34000 16.5 20 225.9 20 0.07 GUA+CE

Amazon Amazon Obidos 4619000 169480 15 18 120.4 22 0.11 Man.,RMad,RNeg

Amazon Amazon Obi 4640000 175000 32.2 20 246.9 20 0.12 Man.,RMad,RNeg

Ü2 œ

"kinetic / ïimitatjôfi""

Fig. 1. Conceptual diagram of area-normalized weathering rates vs. denudation rates and possible end member scenarios of evolution of those rates from a mountain outlet to a large river mouth. (1) As the drainage area becomes larger both W and D decrease but the W/D ratio remains the same; (2) As sediment is trapped in the lowland area D decrease but W remains constant; (3) As sediments from the mountain range undergo weathering in the lowlands, W increases (up to the supply-limitation, where all soluble elements have been solubilised) but D remain constant. Note that the "kinetic-limitation" is represented here as a flat line8, but other forms of kinetic limitations have been reported7,9.

Denudation rate

3. Results

Rivers of the Amazon Basin can be separated into three groups (Tab. 1). (1) Rivers draining only (or mostly) Andean regions have high W and high D and low W/D ratios, i.e. lower than <10-1 and down to 10-2 (meaning that 1% of the denudational flux occurs as dissolved). (2) Rivers draining only (or mostly) lowland and foreland regions have lower D, and relatively high W/D ratios (between 10-1 and 100, but actually always lower than 0.5). (3) The last group of rivers drain both the Andes and lowland areas and have intermediate D and W/D (scattered around 10-1).

Drainage area o ~ 1 o4 km2

O -4.105 km2 aaia^if \ River data ■■*

» * № •V f a o

# • /

f # s 2> / \V

10° 10' 102 103 1 04 Denudation rate (t km'2 yr1)

« 0.4

« 0.2

H Lowland

| | Andean •

I | Mixed

0° n Q°

104 1 05 1 06 Basin Area (km2)

Fig. 3. (left) Weathering rates vs. denudation rates (both area-normalized) for river of the Amazon Basin from the Andean outlet to the mouth (Obidos). (right) Relationship between W/D ratio and basin area. For both panels, open circles represent rivers draining only the Andes, black circles rivers draining only lowland areas, and grey circles Andean rivers including a large lowland area. In the left panel the size of the symbol scales with drainage area and crosses are Andean river basins with drainage area < 15,000 km2. The area shaded for "soil data" is from cosmogenic nuclides and chemical depletion fractions and the area shaded for river data is from dissolved and suspended loads8.

When plotted in the W-D space, rivers of the lowland Amazon Basin with Andean headwaters have both

lower D and W than their pure Andean sources (Fig. 3), suggesting that the large increase of drainage influences the two rates (trend (1) in Fig. 1). However, large tributaries also plot on higher W/D ratios than pure Andean rivers (trend (2) in Fig. 1, Fig. 3). The smallest of the large tributaries have W/D ratios similar to those found in pure Andean basins (Fig. 3).

4. Possible causes for a limited downstream change in weathering intensity

Three hypotheses can be invoked to explain the downstream shift in W/D in Fig. 3: (1) contribution of high W/D lowland tributaries draining cratonic shield areas (2) contribution of high W/D lowland tributaries (Fig. 3) draining large, stable alluvial formations; (3) further weathering of sediments recently derived from mountainous areas during their storage in active floodplains. These two latter hypotheses are not mutually exclusive and actually reflect the same process but on different time scales: weathering of detrital sediments in continental alluvial formations14. For each river reach, the role of processes (1) and (2) can be partially corrected for by calculating the W/D ratio resulting from the mixture of the W and D fluxes of all tributaries contributing to this reach (listed in Tab. 1). This number can then be compared to the W/D ratio measured at the outlet of the reach, yielding a downstream change in W/D (Fig. 4) that ranges from -0.04 (Napo River reach) to +0.04 (Mamore and Upper Solimoes river reaches). This small shift results either from weathering of river sediment in the active floodplain channel or from the contribution of small lowland tributaries or ground water, all not gauged nor sampled. The contribution of these inputs can be assumed to scale with the "un-monitored" drainage area, calculated as the difference between the drainage area at the reach outlet and the sum of the tributary drainage areas. As shown in Fig. 4, the change in W/D does not correlate strongly with the "un-monitored" drainage area (nor does it with "un-monitored" discharge, not shown). This observation suggests that the contribution of small lowland tributaries does not control the small downstream change in W/D, which is therefore attributed to weathering of sediments in the active floodplain.

Fig. 4. Change in the W/D ratio as a function of the "un-monitored" drainage area. For a given river reach, the number on the Y-axis corresponds to the difference between the W/D ratio measured at the outlet and the W/D ratio calculated from the discharge-weighted average of the W and D fluxes of all tributaries (i.e. lowland and Andean-derived) contributing to the reach. The number on the X-axis corresponds to the difference between the drainage area at the reach outlet and the combined drainage areas of all tributaries contributing to the reach.

5. Limitation of weathering intensity by the meta-sedimentary source rock

Despite a slight shift in weathering intensity from the Andes to the Amazon mouth, the largest river in the world exports weathering products characterized by W/D ratios around 10-1. The only rivers actually reaching W/D ratio in the range 0.2-0.5 are pure lowland rivers, most likely because they drain non-eroding settings. If one considers now the Amazon Basin as a whole as a Critical Zone-type reactor, it is surprising that despite fairly long sediment residence times28,29, the weathering intensity at the Amazon mouth (10%) is far below that of supply-limited soil columns (50%). However, most of the soil data is derived from studies made on granitic rocks, of which up to 50% can be dissolved. By contrast, most continental sediment-producing orogens are made up of meta-sedimentary rocks, which have already lost soluble elements during previous weathering cycles30,31. Therefore, low

W/D ratios of denudational fluxes of many mountain ranges might result from the fact that meta-sedimentary source rocks can attain only a low weathering intensity. In support of this explanation we note that rivers of the northern part of the Andes (e.g. Napo river), draining a larger proportion of volcanic and andesitic rocks, display higher W/D ratios than southern tributaries31,32. To summarise, as most river sediment contained in the Earth's largest rivers are derived from meta-sedimentary source rocks, weathering intensity cannot increase significantly with residence time, even if this "Critical Zone" has the size of the Amazon Basin.

References

1.Gaillardet J, Dupré B, Louvat P, Allègre CJ. Global silicate weathering and CO2 consumption rates deduced from the chemistry of large rivers.

Chem. Geol. 1999;159:3-30.

2. Millot R, Gaillardet J, Dupré B, Allègre CJ. The global control of silicate weathering rates and coupling with physical erosion: insights from

rivers of the Canadian Shield. Earth Planet Sci Lett 2002; 196:83-98.

3. Riebe CS, Kirchner JW, Finkel RC. Erosional and climatic effects on long-term chemical weathering rates in granitic landscapes spanning

diverse cliamtic regimes. Earth Planet Sci Lett 2004; 224:547-562.

4. West AJ, Galy A, Bickle M. Tectonic and climatic control on silicate weathering. Earth Planet Sci Lett2005; 235:211-228.

5. Stallard RF. Tectonic, environmental, and human aspects of weathering and erosion: a global review using a steady-state perspective. Annu

Rev Earth Planet Sci 1995; 23:11-39.

6. Carson Ma and Kirkby MJ. Hillslope form and processes, 1972. Cambridge Univerisyt Press, 476 pp.

7. Gabet EJ and Mudd SM. A theoretical model coupling chemical weathering rates with denudation rates. Geology 2009; 37:151-154.

8. Dixon JL and von Blanckenburg F. Soils as limiters and pacmakers of global silicate weathering. C R Acad Sci 2012; 344:597-609.

9. West AJ. Thickness of the chemical weathering zone and implications for erosional and climatic drivers of weathering and for carbon cycle

feedbacks.Geology 2012; 40:811-814.

10. Maher K and Chamberlain CP. Hydrologic regulation of chemical weathering and the geologic carbon cycle. Science 2014; 343:1502-1504.

11. Nanson GC and Croke JC. A genetic classification of floodplains. Geomorphology 1992; 4:459-486.

12. Moquet JS, Crave A, Viers J, Seyler P, Armijos E, Bourrel L, Chavarri E, Lagane C, Laraque A, Casimiro WSL, Pombosa R, Noriega L, Vera A, Guyot JL. Chemical weathering and atmospheric/soil CO2 uptake in the Andean and Foreland Amazon basins. Chem Geol 2011; 287:1-16.

13. Lupker M, France-Lanord C, Galy V, Lavé J, Gaillardet J, Gajurel AP, Guilmette C, Rahman M, Singh SK, Sinha R. Predominant floodplain over mountain weatheing of Himalayan sediments (Ganga basin).Geochim Cosmochim Acta 2012; 84:410-432.

14. Bouchez J, Gaillardet J, Lupker M, Louvat P, France-Lanord C, Maurice L, Armijos E, Moquet JS. Floodplains of large rivers: weathering reactors or simple silos?Chem Geol 2012;332-333:166-184.

15. Callède J, Cochonneau G, Vieira Alves F, Guyot JL, Santos Guimarâes V, De Oliveira E. Les apports en eau de l'Amazone à l'Océan Atlantique. J Wat Sci. 2010; 23:247-273.

16. Milliman JD and Farnsworth KL. River discharge to the coastal oceans: a global synthesis, 2011. Cambridge University Press, 392 pp.

17. Gibbs RJ. Amazon River: environmental factors that control its dissolved and suspended load. Science 1967; 156:1734-1737.

18. Gaillardet J, Dupré B, Allègre CJ, Négrel P. Chemical and physical denudation in the Amazon River Basin. Chem Geol 1997;142:141-173.

19. Bouchez J and Gaillardet J. How accurate are rivers as gauges of chemical denudation? Geology 2014;42:171-174.

20. Mortatti J and Probst JL. Silicate weathering and atmospheric/soil CO2 uptake in the Amazon basin estimated from river water geochemistry: seasonal and spatial variations. Chem Geol 2003; 197:177-196.

21. Guyot JL, Filizola N., Quintanilla J, Cortez J Dissolved solids and suspended sediment yields in the Rio Madeira Basin, from the Bolivian Andes to the Amazon IAHS Publ 1996; 236:55-63.

22. Filizola N and Guyot JL. Sediment yields in the Amazon basin: assesment using the Brazilian national dataset. HydrolProc 2009; 23:32073215.

23. Laraque A, Bernal C, Bourrel L, Darrozes J, Christophoul F, Armijos E, Fraizy P, Pombosa R, Guyot JL. Sediment budget of the Napo River, Amazon basin, Ecuador and Peru. Hydrol Proc 2009; 23:3509-3524.

24. Guyot JL, Bazan H, Fraizy P, Ordonez JJ, Armijos E, Crave A. Suspended sediment yields in the Amazon basin of Peru: a first estimation.

IAHS Publ 2007; 314:1-8.

25. Wittmann H, von Blanckenburg F, Guyot JL, Maurice L, Kubik PW. From source to sink: preserving the 10Be-derived denudation rate signal of the Bolivian Andes in sediment of the Beni and Mamore foreland basins. Earth Planet Sci Lett 2009; 288:463-474.

26. WittmannH, von Blanckenburg F, Maurice L, Guyot JL, Filizola N, Kubik PW. Sediment production and delivery in the Amazon River basin quantified by in situ-produced cosmogenic nuclides and recent river loads. Geol Soc Am Bull 2011; 123:934-950.

27. Brimhall GH and Dietrich WE. Constitutive mass balance relations between chemical composition, volume, density, porosity, and strain in metasomatic hydrochemical systems: results on weathering and pedogenesis. Geochim Cosmochim Acta 1987; 51:567-587.

28. Dosseto A, Bourdon B, Gaillardet J, Allègre CJ, Filizola N. Time scales and conditions of weathering under tropical climate: study of the Amazon basin with U-series. Geochim Cosmchim Acta 2006; 70:71-89.

29. Dosseto A, Bourdon B, Gaillardet J, Maurice-Bourgoin L, Allègre CJ. Weathering and transport of sediment in the Bolivian Andes: time constraints from uranium-series isotopes. Earth Planet Sci Lett 2006; 248:759-771.

30. Dellinger M., Gaillardet J, Bouchez J, Calmels D, Galy V, Hilton RG, Louvat P, France-Lanord C. Lithium isotopes in large rivers reveal the cannibalistic nature of modern continental weathering and erosion. Earth Planet Sci Lett 2014; in press.

31. Moquet JS, Viers J, Crave A, Armijos E, Lagane C, Casimiro WSL, Pépin É, Pombosa R, Noriega L, Santini W, Guyot JL. Comparison between chemical weathering and physical erosion rates in Andean basins of the Amazon River, this issue.

32. von Blanckenburg F, Bouchez J, Wittmann H. Earth surface erosion and weathering from the 10Be (meteoric)/"Be ratio. Earth Planet Sci Lett 2012; 351-352:295-305.