Provenance of sands from the confluence of the Amazon and Madeira rivers based on detrital heavy minerals and luminescence of quartz and feldspar
Introduction
There is great debate about the evolution of the Amazonian river system, especially with regard to its onset to a transcontinental west to east drainage (Campbell et al., 2006, Figueiredo et al., 2009, Shephard et al., 2010, Sacek, 2014). Provenance questions are in the core of this debate because Andean sediments reaching the present Amazon river mouth could be the fingerprint for the development of a transcontinental river. Using sediment characteristics to deduce changes in a river system requires the understanding of factors controlling the production (source), transport and accumulation (sink) of sediments. Under this approach, sediments and sedimentary rocks are products of a source-to-sink system (Allen and Allen, 2005) operating on different temporal and spatial scales. Determining sediment provenance and reworking (i.e., number of burial-erosion cycles during sediment transport) can be considered two fundamental tasks to describe the Amazonian river system under a source-to-sink perspective (Blatt, 1967, Pettijohn et al., 1972, Pettijohn, 1975, Everett and Rye, 2003, Allen, 2008, Carter et al., 2010, Marsaglia et al., 2010, Wolinsky et al., 2010). Sediment provenance and reworking are mutually related and are essential to define proxies able to record spatial and temporal variations in production, transport and deposition of sediments.
Large rivers from tropical settings such as the Amazon are part of the major active source-to-sink systems around the world. The Amazon watershed and its subaqueous fan represent the biggest active source-to-sink system in terms of sediment load in South America (Latrubesse et al., 2005). The Amazon river accounts for more than 70% of the sediment load reaching the South Atlantic ocean (Meade et al., 1985). In this context, the Madeira river stands out as one of the major tributaries of the Amazon river, from which approximately 2.76 × 109 tonnes/year of suspended sedimentary load is delivered to the Amazon river (Meade et al., 1985, Agência Nacional de Águas (Brazilian National Water Agency)., 2014). The Madeira river drains terrains with distinct elevation, climate, vegetation cover and land use and it supplies around 15% of the Amazon river waters (Latrubesse et al., 2005). Factors such as catchment size and intense channel migration dynamics (i.e., meandering, avulsion) make large rivers subjected to complex changes in sediment sources, storage and reworking through time (i.e., Stouthamer and Berendsen, 2001, Slingerland and Smith, 2004). Temporary sediment storage in stabilised bars and floodplains can promote mixing of sediments from different sources and produced under distinct climate or tectonic conditions. Sedimentary reworking within the fluvial system can occur, for example, by means of erosion of ancient bars and abandoned meanders. In large rivers, the control in sediment mixing and reworking goes beyond the action of autocyclic phenomena of meandering or channel avulsion. Allocyclic factors, such as tectonics and climate changes, are also important, especially on millennial timespans (Stouthamer and Berendsen, 2007). Therefore, the study of sediment provenance, mixing and reworking in large tropical rivers is necessary to understand how sediment properties may record allocyclic changes. This is still poorly understood for Amazonian rivers, especially with regard to the sand supply. Most previous studies of the Amazonian rivers focus on the suspended load using geochemical approaches (Meade, 1994, Martinelli et al., 1993, Filizola, 1999, Bouchez et al., 2011, Govin et al., 2014). This paper investigates provenance and sedimentary reworking of sands of the Madeira river supplied to the Amazon river. For this, we performed heavy mineral analysis (Morton and Hallsworth, 1994, Morton and Hallsworth, 1999) combined with optically stimulated luminescence (OSL) sensitivity (Pietsch et al., 2008, Sawakuchi et al., 2011, Sawakuchi et al., 2012) for provenance and sediment reworking analysis of sand from bars and the bottom channel in the confluence of the Madeira and Amazon rivers.
Section snippets
Physiography, geology and fluvial hydrology of the Amazon and Madeira rivers
The Amazon river, in its portion upstream of the Madeira river mouth, is formed by the coalescence of numerous drainages sourced in the Andean region of Peru, Ecuador and Colombia and flowing to the Amazonian plains (mainly in Brazil) (Meade et al., 1985), covering an area of about 3 million km2 (Fig. 1). In turn, the Madeira river has a drainage area of nearly 1.4 million km2, being one of the major tributaries of the Amazon river (Latrubesse et al., 2005). The upstream waters of the Madeira river
Rocks and sediments drained by the Amazon and Madeira rivers
About 70% of the Amazon catchment area (upstream to the Madeira river mouth) is located in Brazil and drains Cenozoic sediments and sedimentary rocks of the Amazonian plains since the “Precordillera” of Peru and Colombia, where the common lithologies are argillites, arenites, siltites and unconsolidated muds, sands and gravels (Reis et al., 2006), but with a primary origin from Andean rocks (Fig. 1). Despite the Negro river draining large areas of the Amazon Craton with exposed Precambrian
Sampling and measuring of channel bathymetry
The sampling area comprehends the downstream sector of the Madeira river channel and the Amazon main channel upstream and downstream of the Madeira river mouth (Fig. 1). Twenty-nine sediment samples were collected in the lower Madeira river and Amazon river around the Madeira river mouth, from exposed sand bars or in the mid portion of channels using a grab sampler. The Amazon river samples comprise 5 samples collected upstream and 13 samples collected downstream. The remaining 8 samples were
Bathymetric profiles
In the study area, the Amazon and Madeira rivers presented contrasting bathymetric profiles concerning depth, width, and bottom morphology (Fig. 2). Between profiles A1 and A2 (separated by 40 km), the Amazon river narrows its channel from 4700 m to 2500 m width, accompanied by increase of the mean depth from − 18 to − 33 m. In the A1 profile, there is a “central plateau” where the mean depth is − 8 m, probably corresponding to an subaqueous extension of the upstream emerged mid-channel bar (Onças
Sand source rocks
The sands from the Amazon river upstream of the Madeira river mouth stand out due to their high content of augite (24.1%) associated with hypersthene (27.6%) and brown (hastingsitic?) hornblende. This heavy mineral assemblage suggests a high proportion of basic igneous rocks in the sediment source area. Despite the presence of augite and (less) hypersthene, the Madeira river also has a high content of andalusite (26.1%) and associated metamorphic minerals, such as sillimanite, staurolite and
Conclusions
The heavy mineral analysis revealed an unexpected assemblage in the recent sands of the Amazon and Madeira rivers, enriched in unstable minerals like augite, hypersthene, hornblende and andalusite. This is in agreement with the relatively high content of feldspar grains in sands from both rivers. Considering that these two rivers drain tropical climate regions, supposedly unsuitable for the preservation of unstable minerals, our results indicate that the Amazon fluvial system is very competent
Acknowledgements
The authors would like to thank Johanna Méndez Duque, Luciana Nogueira and Thays Desirée Mineli by the careful preparation and measurement of the samples. We also like to thank Dr. Piotr Weckwerth and an anonymous reviewer by the useful comments on the first version of the manuscript, and we are grateful to Dr. Jasper Knight (Editor) by the detailed corrections in the text. D.R.N.Jr. thanks Conselho Nacional de Pesquisa Científica e Tecnológica (CNPq) for a postdoctoral fellowship (Grant
References (98)
- et al.
Plate tectonic significance of Middle Cambrian and Ordovician siliciclastic rocks of the Bavarian Facies, Armorican Terrane Assemblage, Germany — U–Pb and Hf isotope evidence from detrital zircons
Gondwana Research
(2010) - et al.
The Pan-Amazonian Ucayali Peneplain, late Neogene sedimentation in Amazonia, and the birth of the modern Amazon River system
Palaeogeography, Palaeoclimatology, Palaeoecology
(2006) - et al.
From mountain source to ocean sink — the passage of sediment across an active margin, Waipaoa Sedimentary System, New Zealand
Marine Geology
(2010) Provenance mixing in an intraoceanic subduction zone: Tonga Trench–Louisville Ridge collision zone, southwest Pacific
Sedimentary Geology
(1990)- et al.
Luminescence characteristics of quartz from the Southern Kenyan Rift Valley: dose estimation using LM-OSL SAR
Radiation Measurements
(2006) - et al.
U–Pb ages of detrital zircons from Permian and Jurassic eolian sandstones of the Colorado Plateau, USA: paleogeographic implications
Sedimentary Geology
(2003) Equivalent dose determination using single aliquots
Nuclear Tracks and Radiation Measurements
(1991)Distinguishing quartz and feldspar in single grain luminescence measurements
Radiation Measurements
(2003)- et al.
Controls of heavy minerals and grain size in a holocene regressive barrier (Ilha Comprida, southeastern Brazil)
Journal of South American Earth Sciences
(2011) - et al.
Characterisation of river bed and suspended sediments in the Rio Madeira drainage basin (Bolivian Amazonia)
Journal of South American Earth Sciences
(1999)