Navigation – Plan du site

Analysis of geomorphic dynamics in ancient quartzite landscape using phytolith and carbon isotopes, Espinhaço Mountain Range, Minas Gerais, Brazil

Analyse de la dynamique géomorphologique dans un ancien paysage de quartzite de la chaîne de montagnes de l'Espinhaço (Minais Gerais, Brésil), à l'aide des phytolithes et des isotopes du carbone
Cristina Helena Ribeiro Rocha Augustin, Heloisa Helena Gomes Coe, Karina Ferreira Chueng et Jenifer Garcia Gomes
p. 355-376

Résumés

Bien que les paysages développés sur quartzite au Brésil s'étendent sur de grandes surfaces, les connaissances sur les processus et les mécanismes impliqués dans leur formation sont peu avancées. Des recherches récentes ont identifié des signatures karstiques associées à la dissolution de la silice. Cependant, il existe d'autres formes de relief et de matériaux couramment rencontrés dans le domaine quartzitique de la chaîne de montagnes de l'Espinhaço, localisée dans le sud-est du Brésil. L'objectif principal de cette recherche est d'analyser les processus géomorphologiques associés aux matériaux recouvrant les surfaces faiblement inclinées et relativement plates qui se trouvent toujours près des affleurements de quartzite, dans la zone d'étude. C'est une étape importante pour tenter de comprendre le développement des types de relief dans les régions quartzitiques des zones tropicales humides. L'approche géomorphologique a été privilégiée, et, sur les sites identifiés, deux profils de sol ont été décrits et échantillonnés pour des analyses en laboratoire. L'étude a été complétée par l'analyse de phytolithes retrouvés dans le sol, des analyses du δ13C et des datations par 14C-AMS (MRT) de la matière organique du sol (MOS). Les profils relativement peu profonds montrent une séquence d'horizons similaires, variant d’une couche sableuse avec beaucoup de matière organique en surface à un matériau caillouteux immédiatement en-dessous. Les horizons superficiels sont constitués principalement de sable fin et d’argile, ce qui est fréquent dans les milieux tropicaux humides. Les deux profils présentent un niveau de pierres angulaires à subangulaires, formé de galets de quartz se trouvant juste au-dessus de la quartzite altérée dans les horizons C, où les valeurs les plus anciennes des datations 14C de la MOS ont été trouvées. Compte tenu de la durée moyenne du temps de résidence (MRT), on peut supposer que les premiers dépôts du profil P2 datent de plus de 10 000 ans (10506-10230 ans cal BP) et que le premier profil (P1), situé sur le versant supérieur, présente une MOS (MRT) plus récente (5919-5664 ans cal BP). Ceci suggère que les processus d'érosion ont été soit plus intenses, soit plus récents, sur cette partie du versant et que les horizons plus profonds peuvent peut-être avoir déjà été en surface, lorsque le matériau s’est déposé. En outre, la grande quantité de phytolithes indéterminables en raison de leur altération, ainsi que la prédominance des types les plus robustes, peuvent être interprétées comme l'évidence d'une érosion intense et d'un transport du matériau par le ruissellement superficiel, suggérant que ce matériau est colluvial et que sa source correspond à l’affleurement de quartzite proche. Après son dépôt sur la « rampe colluviale », terme qualifiant cette topographie associée à une très faible pente, le matériau est resté relativement préservé, ce qui a permis l'occurrence des processus pédogénétiques, aboutissant à la mise en place de néosols quartzo-sableux. La rampe elle-même semble être le résultat d'une très faible altération et remobilisation du matériau, associée à un fort contrôle exercé par les macro-structures mésoprotérozoïques plates de la roche sous-jacente. Cette étude est l’une des premières utilisant les analyses de phytolithes combinées à celles des isotopes du carbone à des fins d'analyse géomorphologique, avec des résultats très intéressants. Leur pertinence souligne leur potentiel pour les études géomorphologiques.

Haut de page

Notes de la rédaction

Article soumis le 15 janvier 2014, accepté le 6 août 2014.

Texte intégral

We would like to thank: the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) for granting founds to participate in the 8th International Conference on Geomorphology (IAG-Paris); the Fundação de Apoio do Estado de Minas Gerais (FAPEMIG: CRA – APQ-01652-09), and the Fundação de Amparo à Pesquisa do Rio de Janeiro (FAPERJ: APQ1 E-26/111.722/2010), for financial support for the realization of this research.

Introduction

1One of the most common features occurring in the ancient massive of Serra do Espinhaço Meridional, in the State of Minas Gerais, Brazil, are the quartzite outcrops. They form a set of outcrops interspersed with low inclined surfaces giving them a plateau like landscape (Saadi, 1995). The relatively flat surfaces present stretches which vary in length from a dozen meters to kilometres long, and are covered by a thin layer of sandy material with a depth between 0.5 and 2 m (Rocha, 2011; Augustin et al., 2011). Sometimes there is just loose sand and at others pedogenic processes are evident with the development of Quartz-sand Neosols and Spodosols near streams or swamp areas (Rocha, 2011). These relatively flat surfaces occur indiscriminately at different topographic levels at a lower position on the plateau, and are always associated with the quartzite outcrops in the area upon which L. King (1956) has identified fragments of different types of cycles of pedimentation since the Jurassic, corresponding to the surfaces of Gondwana, Post-Gondwana and South American. It is not yet clear which processes and mechanisms are involved in the development of these sandy formations. They could either be: a) in situ, of an eluvial nature, resulting from the dissolution or break down by chemical and physical weathering of the quartzite from the underlying rock; or b) colluvium, resulting from the accumulation of material eroded and transported by dispersed and/or concentrated runoff. The position and field characteristics of the sampled profiles do not show evidence that the material could be fluvial. If the sandy formations are eluvium, their evolution is similar to that reported in tropical areas by several authors (Thomas, 1994). The relatively thin depth of the material can be related to the low rates of chemical weathering of quartzite, especially that occurring in a representative part of the Espinhaço mountain range with up to 95% of pure quartz (Fogaça and Scholl, 1979), associated with low rates of erosion. On the other hand, if they are colluvium, they are the result of slope erosion and transport of material from the already weathered rock outcrops. The answer to that question is important because it can help us to understand the dynamics of the landscape in areas of relatively quiet tectonic conditions underlain by quartzite. Although siliciclastic rock is often responsible for the occurrence of higher topography in Brazil, the relief is not well known and has been very little studied. An understanding of the processes involved in the formation of some of its features can undoubtedly enhance the comprehension of how these landforms evolve. This is especially important when it is considered that the processes and mechanisms associated with the formation of relatively flat surfaces occurs in ancient tectonically quiet areas, even at heights above 1000 m that are common to various parts of the emerged lands of the Earth (Peulvast and Sales, 2002; Tucker and Hancock, 2010; Ollier, 2010; Twidale, 1997, 1999, 2014).

2The main objective of this research is to analyse the development of the sandy material encountered on the relatively flat surfaces interspersed with quartzite outcrops in the higher topography of the Espinhaço mountain range. It aims to contribute to a better comprehension of the evolution of the landscape in paleosurfaces of this type of lithology. Although the evolution of ancient landscapes is referred to as temporally long (Twidale, 1999, 2000; Bishop, 2007) we can assume that they also evolve through a pattern of processes such as those operating in the area today. The fact that a significant part of the sandy formations has gone through pedological processes tends to increase the complexity when attempting to identify their nature and environmental conditions. To help in our objectives, pedogeomorphic interpretation and field description of the material, as well as the surroundings, were carried out along with soil sampling for granulometry, phytolith, and 13C/14C isotope analyses. The granulometry analysis is often used by researchers from different fields that deal with forms and processes of earth surfaces. The interpretation of the results is not always straightforward, but we expected that with the help of other tools we would be able to enhance the comprehension of the nature of the material in the study area.

3Phytoliths are microscopic particles formed as a result of absorption of dissolved silicon (Si) from soils by roots, followed by deposition and biomineralization. These biosilica structures can be found inside the cells of many living plants. Furthermore, they are usually found within the finest sand fractions, where the whole plant or tissue parts have decomposed (Piperno, 1988). After being recovered from soils, these microfossils show configurations that can be associated with their original vegetation. Phytoliths can be classified by taxonomy or by their physical characteristics, and they are generally studied by analyses of their assemblages. After phytolith morphotypes are identified, calculation of phytolith indices helps track some important environmental parameters, such as density of trees and adaptation to drought or water stress (Fredlund and Tieszen, 1994; Bremond et al., 2005a; Bremond et al., 2005b; Messager et al., 2010; Coe et al., 2012a). Even though some phytoliths can percolate downwards through the soil profile, statistical analysis of phytolith assemblages can still provide valuable information on the vegetation patterns. As D.A. Krauss et al. (1997) pointed out, as long as the soil composition is relatively uniform with depth, it is safe to assume that the percolation rate for a given phytolith morphotype will not vary significantly.

4Stable isotopes of carbon (12C and 13C) can also be used as a proxy of biological processes in plants. Carbon isotopic discrimination during photosynthesis provides insight into photosynthetic metabolism in plants as well as its environmental influences. For instance, plants that utilize the Calvin-Benson photosynthetic pathway, which tends to prefer the lighter isotope, are referred to as C3 and have δ13C values from -32 to -22‰. C4 plants use the Hatch-Slack cycle and have δ13C values in the range of -9 to -17‰ (Gleixner et al., 2002; Gordon and Goñi, 2003; Killops and Killops, 2005). Consequently, the 13C/12C ratio in soil organic matter (SOM) should remain close to the ratio of the original vegetation. However, due to organic matter (OM) heterogeneity in soils and sediments, δ13C values will depend on the OM original sources (Hedges and Parker, 1976). In forest ecosystems δ13C values of soil OM (SOM) are generally less negative than the source vegetation. Studies show an increase of 1 or 2‰ during decomposition of vegetable tissue (Martinelli et al., 2009).

5Information regarding palaeovegetation changes can ultimately be related to the possible time of occurrence by radiocarbon (14C) dating of SOM. Although charcoal fragments and humin are the most appropriate fractions for dating SOM (Pessenda et al. 2001), the lack of sufficient amounts in some cases has been forcing researchers to rely on 14C dates obtained on total SOM. Because SOM is composed of a complex mixture of OM of different origins at various stages of decomposition, its age is normally significantly younger than dates obtained from more stable organic compounds found in soil (Pessenda et al. 2001). Despite the use of phytoliths for pedogenic analysis having increased recently (Coe, 2009; Coe et al., 2012b; Coe et al., 2013; Gomes, 2012), this is one of the first applications of such combined techniques for geomorphology studies. In this sense, this research is innovative in that it uses this proxy for geomorphic interpretation with emphasis on the Quaternary.

Natural setting

6The study area is located in the Espinhaço Meridional (below 17°S) mountain range in the central-eastern part of the state of Minas Gerais, Brazil (fig. 1), between the geographical coordinates 18°25’30’’S and 43°48’15’’W (Profile 1: P1; fig. 1), and 18°25’31’’S and 43°48’26’’W (Profile 2: P2; fig. 1).

Fig. 1 – Location of the study area. It occurs in the Espinhaço massif in southeast Brazil, characterized by the dominance of quartzite from the Espinhaço Supergroup.
Fig. 1Localisation de la zone d'étude. Massif d'Espinhaço, dans le sud-est du Brésil, caractérisé par la prédominance de quartzite du Supergroupe Espinhaço.

Fig. 1 – Location of the study area. It occurs in the Espinhaço massif in southeast Brazil, characterized by the dominance of quartzite from the Espinhaço Supergroup.Fig. 1 – Localisation de la zone d'étude. Massif d'Espinhaço, dans le sud-est du Brésil, caractérisé par la prédominance de quartzite du Supergroupe Espinhaço.

7The mountain range corresponds to the external portion of the western half of the Araçuaí-West Congo Neoproterozoic orogen that “fringes the São Francisco craton to the east and is separated from the crystalline core of the orogen by the Abre Campo geophysical discontinuity” (Peres et al., 2004, p. 771), and it is formed by three main geological units. The quartzite from the Espinhaço Supergroup is stratigraphically positioned on top of the depositional sequences that were metamorphosed, folded and faulted during the Brasiliano techno-thermal event (ETT), in the Neoproterozoic (600-530 Ma). Schist can be found below, and this was also deformed by the same event and the remobilized crystalline basement, exposed by erosion in the Gouveia Depression. The most recurrent geomorphic features of the Espinhaço Meridional are scarps, faults, and residual syncline and anticline structures with a SW-NE direction following the main deformational direction (Saadi, 1995; Augustin et al., 2011). They form sequences of ridges, known as “sierras”, eroded by systems of small rivers that explore the faulting and deformational features to dissect the inherited geological background (Augustin, 1995), exposing quartzite outcrops. This explains why the drainage always follows the main geological lineaments (Fonseca and Augustin, 2014). The ruptile characteristics of the quartzite favoured the fracturing and faulting of these rocks, generating elevations and scarps that formed, after millions of years of denudation, today’s sequence of sierras at altitudes above 1000m (Augustin et al., 2011; Fonseca and Augustin, 2014), in which the main features are the quartzite outcrop, separated by flat surfaces of different extension, one of which is the object of this research. This area is located on the eastern border of a syncline named “Serra do Engenho” (fig. 2).

Fig. 2 – Location of P1 and P2 on the low inclined surface near the quartzite outcrops of the Serra do Engenho syncline, in the Espinhaço massif.
Fig. 2L'emplacement de P1 et P2 sur la surface basse inclinée près des affleurements de quartzite du synclinal de la “Serra do Engenho”, dans le massif Espinhaço.

Fig. 2 – Location of P1 and P2 on the low inclined surface near the quartzite outcrops of the Serra do Engenho syncline, in the Espinhaço massif. Fig. 2 – L'emplacement de P1 et P2 sur la surface basse inclinée près des affleurements de quartzite du synclinal de la “Serra do Engenho”, dans le massif Espinhaço.

The inclination is gentle and ends at a valley wall. Grassland is predominant and riparian cover follows a creek before it cuts the portion of the ramp near the main scarp of the syncline.
L'inclination est douce et se termine sur un versant de vallée. La prairie domine et une végétation riveraine borde un ruisseau avant qu'il ne coupe la partie de la rampe située près de l'escarpement principal du synclinal.

8The vegetation, although classified as being predominantly composed of rupestrian grassland conditioned by its lithology, soil and climate, is actually composed of a wide variety of phytophysiognomic aspects, some of which are typical of the Cerrado biome. The Cerrado is characterized by the presence of low tortuous crooked twisted trees, with cracked or furrowed trunks with thick bark and rigid coriaceous leaves. It may present variations in the predominance of plant strata, and the tree cover ranges from 5% to 70% (Medina, 2004). According to A.F.B. Alvarenga (1985) and C.H.R. Augustin (1995) there are four types of vegetation cover in the region: (a) the rupestrian grasslands, typical of rock outcrops, in areas above altitudes of 1000 m; (b) altitudinal vegetation, occurring on flat surfaces, also above 1000 m, associated with the presence of sandy material, and shallow acidic soils low in nutrients. Most of them are Litholic Arenosols or Spodosols, interspersed with wetter areas, forming swamps, which may be permanent, where peats appear, or are periodically flooded; (c) there is also riparian vegetation along the streams especially in the domain of granite-gneiss and metabasic rocks; (d) cerrado vegetation, which is very degraded, appearing between altimetric elevations of 800 to 1000 m, in which there is the dominance of a grass stratum, intercalated by herbaceous shrub and a few low trees. On the small study plateau there is the presence of rupestrian and a mixture of altitudinal vegetation and degraded cerrado, in which grass and a few scattered low shrubs (80-100 cm high) predominate (fig. 2).

9The climate is, according to the Köppen classification, Mesothermic, Cwb (or intertropical). The climate is characterized by humid summers with an average annual rainfall of 1400 mm, there being 81 to 100 rainy days concentrated between October and March when there is an average annual temperature of 28°C, and dry winters from April to September, having an average temperature of 18°C (Augustin, 1995; Almeida Abreu and Renger, 2002). Due to the high altitudes, the annual average temperature does not reach more than 22°C (Nimer, 1989).

Material and Methods

Sites and samples

10The study area (fig. 1 and fig. 2) was chosen because it is representative of similar landforms characteristic of the Espinhaço Mountain Range. For the identification of the area, photo-interpretations of the 1997 Cruzeiro do Sul photos, at a scale of 1:25,000, were used, with the help of a glass stereoscopic device.

11The area is composed of a low-inclined surface adjacent to a quartzite outcrop, sloping to the southwest towards a small stream well embedded in metabasic rock (fig. 2).

12The sampling points were identified along a transect after the measurement of declivity using 2 m poles placed at regular 20 m intervals from the upper portion of the low-inclined slope to where the inclination becomes higher, indicating the immediate influence of the valley, further down the slope (fig. 2). The direct reading of declivity was measured with a Suunto clinometer and geomorphological sites (GS) were identified. GS was first described by R. Bourne (1931, apud C.H.R. Augustin, 1979), later refined geomorphologically by D. Lintton (1951) and R.L. Wright (1973) and applied by C.H.R. Augustin (1979, 1995). It can be definied as portions of the slope surface that present a regular inclination correlated to the occurrence of relatively homogenous internal and external characteristics of the material and limited by breaks of slope, which may be associated with changes in these characteristics.

13We assume the material in the profiles to be regolith, or soil in the broader sense. It implies that we could describe the profiles as a geomorphic remnant of the processes occurring on the slope, not only as the result of pedogenesis. The soil pits were opened at the centre of the two identified Geomorphological Sites (GS). They showed the presence of different types of materials described following the procedures of R.D. Santos et al. (2005) and 8 samples out of a total of 17 were collected for laboratory analysis. The material corresponds to 4 sequences identified at P1 and 4 at P2 (fig. 3). The collected samples were subjected to granulometry, MO, phytolith, isotope and 14C analyses.

Fig. 3 – Soil profiles from P1 (A) and P2 (B).
Fig. 3Les profils de sol étudiés : P1 (A) et P2 (B).

Fig. 3 – Soil profiles from P1 (A) and P2 (B). Fig. 3 – Les profils de sol étudiés : P1 (A) et P2 (B).

Both present a sandy organic superficial horizon with signs of laterization, and sequences of angular quartz pebbles below. In P2 the stony layer follows a sandy horizon rich in OM. In both profiles the lower horizons are formed by weathered quartzite with different levels of laterization.
Les deux profils présentent un horizon superficiel organique sablonneux avec des signes de latérisation et des séquences de galets de quartz angulaires au-dessous. Sur le profil P2, la couche de pierres suit un horizon sablonneux riche en matière organique. Sur les deux profils, les horizons inférieurs sont formés de quartzite altérée avec différents niveaux de latérisation.

14Samples of the vegetation cover were also collected within a range of 50 m2 of the sampling points to serve as modern phytolith assemblages for reference. They represent the main types currently present in the study area of rupestrian grassland, rupestrian grassland under “arnica”, grasses, wood, woodland with palms, grassland with few trees, riparian woodland, grassland and grassland with many trees (fig. 2).

Laboratory analyses

Granulometry of the material

15The granulometry of the 8 samples from the 2 profiles was analyzed following the procedures of the Empresa Brasileira de Agropecuária – EMBRAPA (1997) at the Geomorphological Laboratory of the Geosciences Institute at the Federal University of Minas Gerais (UFMG). The other 9 samples correspond to the superficial surface collected under the main types of identified vegetation cover mentioned above, for specific phytolith analysis.

Phytolith analyses

16The methodology for the phytolith analyses consisted of laboratory procedures for the extraction of phytoliths, microscopy for quantification, classification and calculation of phytolith indices from the collected material. Phytolith extraction was performed in the sedimentology laboratory of LAGEMAR – UFF and in LabGeo of UERJ/FFP. Phytoliths were extracted from 20 g of dry soil after the dissolution of carbonates, organic matter oxidation, iron oxide removal, and granulometric and densimetric (2.35) separation. The identification and counting of phytolith assemblages were then carried out using an optical microscope at 400x magnification. We counted at least 200 phytoliths with a diameter greater than 5 μm and taxonomic significance (classified) as well as those without taxonomic significance (unclassified) due to dissolution or fragmentation. Results are presented as percentages of the total of classified phytoliths according to the classification of C. Twiss (1992) augmented by S.C. Mulholand (1989), G.G. Fredlund and L.L. Tieszen (1994), R. Kondo et al. (1994), A. Alexandre et al. (1997), D. Barboni et al. (1999) and F. Runge (1999), and according to the ICPN (International Code for Phytolith Nomenclature 1.0; Madella et al., 2005). After counting, the phytolith indices D/P, Bi, Iph and Ic were calculated.

17For the chemical extraction and counting, phytoliths were extracted from 20 g of dry soil, slightly crushed, sieved at 2 mm, and processed through the following stages (Kelly, 1990): (1) dissolution of carbonates using HCl (1N); (2) oxidation of organic matter using H2O2 (30%) at 90ºC; (3) solubilization of iron oxides using sodium citrate (Na3C6H5O7) and sodium hydrosulfite (Na2S2O4); (4) removal of the clay fraction (< 2 mm) by decantation; (5) densimetric separation of opal particles in a heavy liquid of sodium polytungstate (density of 2.3 g/cm3). Particles were dried and weighed after extraction and the Phytoliths were then mounted on microscope slides in glycerin for 3D observation and in Entellan for counting at 400x magnification. At least 200 phytoliths with taxonomic significance were counted.

18To identify phytolith types, they were classified into eleven types following the classification of C. Twiss (1969, 1992) improved and completed by the phytolith shape descriptions of S.C. Mulholland (1989), G.G. Fredlund and L.L. Tieszen (1994), R. Kondo et al. (1994), A. Alexandre et al. (1997), C.A.E. Strömberg (2004) and J. Mercader et al. (2009), and named using the International Code for Phytolith Nomenclature 1.0 (Madella et al., 2005) as: (1) Globular echinate, which is produced by Arecaceae (Kondo et al., 1994; Runge, 1999; Runge and Fimbel, 1999; Vrydaghs and Doutrelepont, 2000) but can also be found in Bromeliaceae (Piperno, 1985); (2) Globular granulate, produced by tropical trees at low elevation (Scurfield et al., 1974; Kondo et al., 1994; Alexandre et al., 1997; Bremond et al., 2008); (3) Globular smooth, which appears to have several origins. It has been reported in the epidermis of leaves and in the ray or parenchyma cells of dicot twigs and wood (Kondo et al., 1994); (4, 5) Cuneiform and Parallelepiped Bulliform cell types produced in grass epidermis (Twiss, 1969; Kondo et al., 1994) but which can also be found in Cyperaceae subject to hydric stress (e.g., Grigore et al., 2010); (6) Acicular type from micro-hair and prickles produced in grass epidermis (Palmer et al., 1985; Kaplan et al., 1992); (7) Bilobate and (8) Cross short cell types produced in great quantities by the Panicoideae grass subfamily (Twiss, 1969; Mulholland, 1989; Fredlund and Tieszen, 1994; Kondo et al., 1994); (9) Saddle, which is produced in high quantities by the Chloridoideae grass subfamily (Twiss, 1969; Mulholland, 1989; Fredlund and Tieszen, 1994; Kondo et al., 1994); (10) Elongate originating from the silicified long cells of grass epidermis and (11) Blocky and Tabular types reported in woody taxa (Strömberg, 2004; Mercader et al., 2009), which are not easy to distinguish from the Parallelepiped bulliform types from grasses and were merged into a single morphological group. Phytoliths without taxonomic significance (unclassified phytoliths) because of their original shape, or due to dissolution or fragmentation, were gathered into an unclassified category. Types are presented as percentages of the sum of classified phytoliths. Counting of the same slide by two persons gave reproducibility (SD) of ± 2%. After counting, the phytolith indices D/P, Bi, Iph and Ic were calculated.

19Three phytolith indices were calculated as follows: (1) the D/P index is the ratio of the dicotyledonous phytolith type (Globular granulate) over the sum of Poaceae phytolith types (Short cells + Acicular + Cuneiform bulliform; Bremond et al., 2005a); (2) The hydric stress index Bi, previously named Fs (Bremond et al., 2005b; Messager et al., 2010) is the ratio of the Cuneiform bulliform type over the sum of Poaceae phytolith types (Short cells + Acicular + Bulliform cuneiform), expressed as a percentage; (3) the Ic (climatic) index is the ratio of the phytolith types characteristic of grasses from temperate cold regions and high intertropical elevations (Rondel + Trapeziform polylobate + Trapeziform short cell) over the sum of short cell phytolith types of Poaceae (Rondel + Trapeziform polylobate short cell + Trapeziform short cell + Saddle + Cross + Bilobate short cell). The aridity index Iph (Fredlund and Tieszen, 1994; Bremond et al., 2005b) is the ratio of bilobate type over the sum of short cell phytolith types, but given the low amount of Bilobate type, this index was not calculated in this study.

Stable isotope analyses

20Elemental and isotopic soil organic matter (SOM) analyses were performed by the Laboratory of Isotope Ecology CENA / USP, with a detection limit of 0.03%. After sieving at 2 mm and removal of carbonates, the samples were weighed in tin capsules using an analytical balance. The analyses of the elemental composition of OM concentrations were carried out in a Carlo Erba elemental analyzer, model EA 1110, coupled to a Finigan Delta Plus mass spectrometer, allowing the simultaneous determination of organic carbon concentration and isotope signatures. Results are expressed as δ13C relative to PDB (Pee Dee Belemnite) defined as δ13C (‰ - parts per thousand) = ([R sample / R standard] -1) x 1000. The samples were analyzed with a precision of 0.2‰. The limit of detection for C was 0.03%.

Radiocarbon Analyses

2114C-AMS dating of soil organic matter (SOM) was performed at the Institute of Physics of the Fluminense Federal University (LAC-UFF), Brazil, which runs a NEC 250kV Single Stage Accelerator Mass Spectrometry system. Before measurement, the SOM fraction was chemically separated using 1N HCl at 90ºC to remove carbonates. This procedure was repeated until the samples stopped foaming and the supernatant was clear, after which the SOM fraction was rinsed with deionised water to neutralize pH and dried at 90ºC. Pre-treated samples were then combusted in sealed evacuated pre-baked quartz tubes at 900ºC, cryogenically cleaned, and converted to graphite using zinc and titanium hydrate, with iron catalyst in Pyrex tubes in a muffle oven at 520ºC for 7 h as described in X. Xu et al. (2007).

22In order to verify the distribution of 14C ages of different chemical soil fractions (Pessenda et al., 2001), a refractory C fraction (humin) was extracted from the topsoil horizon, following the conventional acid-base-acid (ABA) protocol (Santos and Ormsby, 2014). The ABA pre-treatment can be summarized as repeated baths with 1N HCl and 1N NaOH at ~70°C, with the base washes repeated until the labile extract of humic acid is fully removed. Upon pre-treatment, clean samples were combusted to CO2, cryogenically cleaned, and reduced to graphite by hydrogen reduction, following established protocols (Santos et al., 2007a, b). This procedure and 14C-AMS measurement was carried out at UCI, as we anticipated that a graphite target with a very low carbon mass (<< 0.050 mgC) would be produced. Carbonates were removed from the soil, using just 1N HCl rinses as described earlier. All 14C results were corrected using the on-line δ13C AMS values of the respective measured graphite aliquots obtained during the respective 14C-AMS measurements by both facilities. Control and background samples, subjected to the same chemical procedures as unknown samples, were also processed to graphite and measured. Age calibrations were performed using CalibBomb (http://calib.qub.ac.uk/​CALIBomb/​) and the Southern Hemisphere (SHCal04) dataset (McCormac et al., 2004) in the OxCal4 software (Bronk Ramsey, 2009).

Results

Soils profiles

23The two soil profiles are localized on a gently inclined surface (3o). Profile 1 (P1) is situated in the highest portion of the slope, about 30 meters from one of the quartzite outcrops and Profile 2 (P2) was taken from a slightly concave element at a distance of 120 m down slope from P1, in the central section of the slope (fig. 2). Both profiles present “horizons” in which it is possible to identify sequences of different materials (fig. 3). The following “horizons” were identified in P1: A1: 0-25 cm, organic rich sand material; A2: 25-40 cm, laterized sand; 40-55 cm, angular to subangular pebbles with 0.3 to 1.5 cm diameters (stony layer); and C: below 55 cm, highly weathered quartzite with laterization though still showing its structure (fig. 3). In P2 the following “horizons” were described: A1: 0-30 cm: organic rich sand material; 30-45 cm, sandy horizon with a higher quantity of organic matter (organic layer); 45-60 cm, angular to subangular pebbles, with diameters from 0.2 to 1.0 cm (stony layer); C: below 60 cm, highly weathered quartzite with laterization in which it is possible to identify mottling (fig. 3). In both profiles the presence of organic matter as well as a stony layer can be observed (fig. 3, tab. 1 and tab. 2). The strong chemical weathering process is responsible for the laterization of all horizons, especially in the C horizon corresponding to the underlying quartzite, with the presence of the colour red at the bottom of both profiles.

Granulometric results

24The granulometric results (tab. 1) show that the dominant diameter of the material in both profiles is fine sand (80-67%) and that the presence of clay is very low (˂ 5%).

Tab. 1 – Results from the granulometric analysis of the material from the two studied profiles.
Tab. 1Résultats des analyses granulométriques du matériel des deux profils étudiés.

Granulometry (%)

Samples

Regolith

Depth (cm)

Coarse sand

Medium sand

Fine sand

Total sand

Silt

Clay

P1/A1

0-25

1

3

76

80

17

3

P1/A2

25-40

1

2

79

82

16

2

P1/Stony layer

40-55

4

2

70

77

21

2

P1/C

+ 55

0

5

69

74

24

1

P2/A1

0-30

1

5

78

84

13

3

P2/A2 organic

30-45

1

2

67

71

25

5

P2/Stony layer

45-60

6

2

69

77

20

3

P2/C

+ 60

0

1

67

69

30

2

25The percentage of total sand (P1: 80-74%; P2: 84-69%), as well as the amounts of fine sand, tend to decrease with depth (P1: 76-69%; P2: 78-67%). It is interesting to observe that the percentage of silt tends to increase in the deeper horizons (P1: 17-24%; P2: 13-30%), where the weathered rock is encountered, while clay decreases (P1: 3-1%; P2: 3-2%), albeit irregularly, because it actually goes up in the second layer of P2 (5%). The percentage of pebbles, with diameter between 2 and 4 cm, is lower in P1 (3%) than in P2 (8%) and occurs at different but relatively similar depths of 40-55 cm in P1, and 45-60 cm in P2. The pebbles are angular to subangular, with a predominance of angular (fig. 3). The sequences of the angular/subangular pebbles found in both profiles are formed of pure milky quartz despite the differences in depth and thickness (tab. 1). SOM occurs in the upper horizon of both profiles but is more concentrated in the second horizon of P2 (fig. 3 and tab. 2). This is reflected by the colour of the horizons: P1 (A1-5YR 5/4; A2 5YR 5/8; stony layer: 5YR 5/6 and C: 5YR 5/8), P2 (A1: 5YR 4/3; A2: 5YR 2,5/1; stony layer: 5YR 3/1and : 5YR 6/6). The colours are also related to the differences in laterization along the profiles, especially at the bottom. The proportion of coarse sand is higher in the stony layers, but the percentages are still relatively low for stony horizons (P1: 4%; P2: 6%). That is especially meaningful considering that the total amount of sand is higher in both profiles in the superficial horizons (tab. 1).

26The analysis of soil organic carbon stocks shows that they are higher in P2 than in P1, with the highest concentration in the organic horizon (fig. 3 and tab. 2). Carbon stocks in the soil follow the normal pattern of decreasing with depth, except for in the organic horizon of profile 2, where the stock is more than 3 times higher than the surface horizon (fig. 3 and tab. 2).

Tab. 2 – Phytolith, carbon and isotopic results.
Tab. 2Résultats des analyses de phytolithes et isotopiques pour les profils étudiés.

Tab. 2 – Phytolith, carbon and isotopic results. Tab. 2 – Résultats des analyses de phytolithes et isotopiques pour les profils étudiés.

Isotopic analysis - 14C-AMS dating

27The MRT ages of soil organic matter (SOM) are assumed to be the minimum mean ages (14C dating) of deposition in these soils. The results of carbon isotope of six samples and the calibrated ages are presented in table 3.

Tab. 3 – 14C-AMS results for the study profiles – Serra do Engenho-Espinhaço Meridional.
Tab.3Résultats des datations 14C-AMS pour les profils étudiés – Serra do Engenho-Espinhaço Meridional.

Samples

Depth

(cm)

Calibrated Ages

(yr cal BP)

Mean Calibrated Ages

(yr cal BP)

P1 A1

0 - 25cm

629 - 519

574

P1 A2

25 - 40cm

2,335 - 2,152

2,243

P1 C

+ 55cm

5,919 - 5,664

5,792

P2 A2 organic

30 - 45cm

315 - 151

233

P2 Stony layer

45 - 60cm

439 - 150

295

P2 C

+ 60cm

10,506 - 10,230

10,368

28There was insufficient material for 14C dating of the stony layer of P1. The results show the variation of 14C-AMS dating of soil organic matter (SOM) in the different horizons in both profiles (tab. 3, and fig. 3). The highest values, considering the MRT, are found at the top of the weathered horizons, especially in P2 (10,506 - 10,230 yrs cal BP), which places it at the end of the Pleistocene/beginning of Holocene (GSA Geologic Time Scale, 2011). The older SOM of P1 (5,919 – 5,664 yrs cal BP) is more or less half the values of the older SOM of P2. The upper horizons in both profiles are relatively younger, especially in the superficial horizon of P2 (315 – 151 yrs cal BP).

Phytolith analysis

Quantification

29The result of the quantification of phytoliths is presented in table 2, figure 4 and figure 5.

Fig. 4 – Phytolith and isotopic analyses of Profile 1.
Fig. 4Analyses phytolithiques et isotopiques du profil 1.

Fig. 4 – Phytolith and isotopic analyses of Profile 1. Fig. 4 – Analyses phytolithiques et isotopiques du profil 1.

Fig. 5 – Phytolith and isotopic analyses of Profile 2.
Fig. 5Analyses phytolithiques et isotopiques du profil 2.

Fig. 5 – Phytolith and isotopic analyses of Profile 2. Fig. 5 – Analyses phytolithiques et isotopiques du profil 2.

30It shows that the stock of phytoliths of P1 follows a normal trend of decreasing with depth, with more phytoliths in the surface horizons than in deeper ones. P2 shows a very large increase in the stock of phytoliths in the organic horizon. Profile 2 presents lower stocks than Profile 1, with the exception of the organic horizon, where the value is very high (almost 5%).

Classification

31It is possible to observe from the results of the quantification of phytoliths (tab. 2) that in Profile 1 only 16% of the counted particles were classified in the A1 horizon (tab. 2 and fig. 5). This number increases in the A2 horizon, with 44% of the phytoliths being classified, and then decreases with depth (36 and 35%). In Profile 2 (tab. 2, fig. 4 and fig. 5), in horizon A1, 56.1% were classified phytoliths, with a significant increase of classified phytoliths in the organic horizon (85.6%), before once again decreasing with depth, as in profile 1, reaching only 9% in the deepest horizon.

Types

32Considering the types of phytoliths, we observed in Profile 1 (fig. 4, fig. 5, fig. 6 and tab. 2) that among the classified phytoliths the most common types are the blocky (from 42 to 68%) and the parallelepiped bulliform (from 4 to 32%), followed by globular granulate (from 6 to 32%) and elongate (5 to 11%).

Fig. 6 – Photos of main types of observed phytoliths in P1 and P2 horizons.
Fig. 6 – Photos des principaux types de phytolithes observés dans les échantillons des profils 1 et 2.

Fig. 6 – Photos of main types of observed phytoliths in P1 and P2 horizons. Fig. 6 – Photos des principaux types de phytolithes observés dans les échantillons des profils 1 et 2.

a: bulliform parallelepiped; b: bulliform cuneiform; c: elongate; d: polylobate; e: bilobate; f: cross; g, h: rondel; i: trapeziform; j, k, l: acicular; m: globular granulate; n, o: globular echinate.
a : bulliforme parallélépipédique ; b : bulliforme cunéiforme ; c : allongé ; d : polylobé ; e : bilobé ; f : en croix ; g, h : rondelle ; i : forme en trapèze ; j, k, l : aciculaire ; m : sphérique granuleux ; n, o : échinulé sphérique.

33Bulliform cuneiform and especially the short acicular cell types were not very representative. In Profile 2 (fig. 4, fig. 5, fig. 6 and tab. 2) among the classified phytoliths, we observed mainly blocky (from 13 to 57%), bulliform paralellepipedal (from 9 to 31%), elongate (from 12 to 27%), bulliform cuneiform (from 7.6 to 13.4%), bilobate (from 1.5 to 13%), acicular (from 0.7 to 11%) and globular granulate (from 2 to 6%). Globular echinate, cross, rondel, and trapeziform polylobate were less representative. In both profiles, among phytoliths of Poaceae, which are predominant in all samples, the assemblages are always composed mainly of larger phytoliths (bulliform, elongate, acicular), and the small phytoliths (short-cells) are rare or absent. However, the few short-cells observed are all from C3 grasses (bilobate, cross) or from grasses adapted to altitude (rondel, trapeziform, polylobate; fig. 4, fig. 5, fig. 6 and tab. 2).

Phytolith indices

34The phytolith indices show that the D/P ratios (density of tree cover) in the surface horizons are characteristic of open vegetation cover. In Profile 1 the D/P presents much higher values (0.2 to 1.8) than in Profile 2 (0.02 to 0.12) as can be seen in table 2, figure 4 and figure 5. In Profile 1, the D/P varies between horizons (1.3 in the A1 horizon and 0.2 in A2). The highest D/P (1.8) is found in the stony layer, returning to the value of the surface in the deepest horizon. In Profile 2, the index varies little, except for in the organic horizon, which has the lowest D/P of the profile (0.02). Although current tree density in P1 is higher than in P2, both have values typical of savannah formations (Alexandre et al., 1997, Barboni et al., 1999) with a predominance of grasses similar to the current vegetation (tab. 2, fig. 4 and fig. 5). The Bi index (water stress) in the surface horizons is high and similar in both profiles (tab. 2, fig. 4 and fig. 5), but the variations in Profile 1 are more significant. The index increases in the A2 horizon (from 65 to 88%), decreases and reaches its lowest value in the stony layer (46%), returning to a value close to that of the surface (63%) in the deepest horizon (tab. 2, fig. 4 and fig. 5). This does not apply to Profile 2, where the index does not vary much, increasing with depth (from 61 to 77%), with a small decrease in the organic horizon, where it reaches the minimum value of 57% (tab. 2, fig. 4 and fig. 5). In both profiles, the Iph (aridity) and Ic (climate) indices could not be calculated, because the extremely low percentages (less than 5%) or absence of short cell phytoliths makes them statistically insignificant.

Discussion

35The occurrence of relatively flat surfaces near quartzite outcrops in the ancient massif of Serra do Espinhaço in southeast Brazil raises questions about the processes involved in the origin of the sandy material covering them. Due to the dominance of a warm humid climate in the area, it may be possible that their development has taken place in situ (eluvium). However, given that siliciclastic rocks are fairly resistant to chemical weathering (Asumadu et al., 1988; Brichta et al., 1980; Turner et al., 2003; Heather et al., 2008; Thomas, 1994), it may also be the product of the accumulation (colluvium) of chemically and physically weathered material, eroded from outcrops and upper portions of the slope and transported by disperse and/or concentrated runoff, mainly, but not only, sheet wash. In this case they are similar to the erosive/depositional “colluvial ramps” reported by M.R. Meis et al. (1979) in other lithologies in Brazil. Understanding the nature of the material is important because it can help to clarify how these features evolve. The study profiles show evidence of pedogenic processes which may exert influence on the interpretation of the results and make it more difficult to identify the nature of the material in the soil profiles. This analysis is based on the characteristics of the profile, material type and granulometry, with the help of results from phytolith and isotopic analysis (δ13C) and the 14C-AMS dating of SOM, as well as direct observation of the profile in the field, which includes their geomorphologic and geological context.

Granulometry and characteristics of the profiles

36Both soils are classified as Quartz-sand Neosols, although P2 shows a concentration of SOM in A2, which could be identified as B organic horizon as it is usually associated with Spodosols (fig. 3 and tab. 2). Because chemical analyses were not carried out to identify the presence of phosphorus, commonly reported in the lower horizons of Spodosols (Oliveira et al., 1992), and there were no other features detected that characterize such soils, like the presence of Bhs or Bs horizons before the “E” albic horizon, we assume that it is not a Spodosols. The absence of organic and aluminium colloid cementation (Oliveira et al., 1992) reinforces this interpretation. Difficulty in identifying Spodosols in Brazil is common, as reported by A.P. Oliveira et al. (2010, p. 848), since “existing studies on the class of Spodosols in Brazil are still insufficient to establish analytical criteria for a definition of the spodic B horizon and, consequently, for the classification of the Spodosols that occur here”. The soil profiles have a similar sequence of horizons, except for the organic one in P2 (fig. 3, tab. 1 and tab. 2). They also show relatively low variability of granulometry values (tab. 1) when compared with each other, which can be assumed to be an indication of pedo-geomorphological processes acting concomitantly and affecting the entire slope.

37The dominant diameter of the material in both profiles is that of the fine sand, while the presence of clay is very low (tab. 1). This is to be expected because the dominant quartzite rocks belong to the “Galho do Miguel” Formation (Espinhaço Supergroup), which is mainly formed of fine pure quartz mineral that is rarely micaceous (Fogaça, 1997). Coarser granulometry is rare (maximum 5%). Variations of the granulometry along the horizons in each profile (tab. 1) indicate, however, differences which may be representative of the processes acting on the slope, meaning the origin of the soil material may not be only from pedogenic processes. The percentages of coarser sand, which would be expected to be higher in the deeper horizons, due to chemical weathering usually being less effective with depth, are represented in P1 and P2 only in the stony layers, with 4 and 6%, respectively, but otherwise show no variation and are even absent in the weathered rock (0%). This differs from the weathering in situ model, where remnants of the weathered rock are supposed to be present near the weathering front (Büdel, 1970; Thomas, 1994). Also, while medium grain sand is more concentrated in the weathered rock in the C horizon of Profile 1 (5%), in Profile 2 it occurs higher up in the A1 horizon (also 5%), separated from the underlying weathered rock by two other layers of very different natures, one stony and the other organic rich. It is also very low (1%) in the weathered rock C horizon. The fine sand, predominant in all horizons in both profiles, also shows percentage differences. In P1 it is higher in A2 (79%) and in the stony layer (70%) immediately below, with the lowest values occurring in the weathered quartzite (69%), which is interesting because it shows no direct association with the mineral granulometry (fine sand) from the underlying quartzite, which could suggest an allochthon source of material. In P2, the amount of fine sand is greater in the superficial horizon (A1 = 78%), decreases in A2 (organic layer = 67%), increases again in the stony layer (69%) and presents the same amount as A2 in the weathered rock C horizon, reinforcing the interpretation that in both profiles the material comes from the outcrops and upper slope.

38When considering silt distribution along the horizons (tab. 1), it is important to notice that it tends to increase with depth in both profiles, suggesting a distribution pattern commonly encountered in humid tropical soils (Thomas, 1994). However, because silt is in the range of fine sands and found everywhere, it is not a very good indicator of pedo-geomorphological processes and this increase can be the result of vertical translocation rather than in situ weathering, which could explain the higher variability of this fraction along the P2 profile. This is especially true if one considers the bigger granulometry of the sand in the upper horizons that tends to promote translocation of finer particles. This has to be considered also in the case of the clay particles. Although the percentage of clay is very low in both profiles, due to its scarcity in the composition of the underlying rock, it is possible to observe that P2 presents a higher percentage than P1 (tab. 1). It should be remembered that in the development of a soil left relatively undisturbed for some time, it is to be expected that clay migrates downwards so that the highest amount occurs in the second horizon, as is the case in the A2-organic layer (5%).

39Considering that the mineral composition of the material is the same in the upper and lower portions of the profiles, the concentration of smaller diameters of fine sand and clay in superficial horizons (tab. 1) may also be associated with either chemical weathering of the rock in situ, which tends to be more effective near the surface in humid tropical conditions (Duchaufour, 1982; Thomas, 1994), or the deposition of finer weathered material from upper slope transported by sheet erosion (De Ploey and Poesen,1989).

40The presence of stony layers in both profiles (fig. 3 and tab. 1), which are formed of very rough angular quartz milk that is thicker in P2 (4%) than in P1 (6%), may be an indicator of possible material origin in the profiles. They could be stone-lines, resulting from in situ breaking down of quartz veins and whose mobilization occurs through bioturbation (Johnson, 1989; Peacock and Fant, 2002). We assume in this case that there were two quartz veins. This is because of the distance between the profiles, which is more than 200 m, and is also due to the fact that it would be difficult to have such an extensive intrusion of thermal solution in the quartzite at similar depths in this type of folded strata. According to Thomas (1994), the stone-lines could also be of alluvial or colluvial origin, resulting from creeping or sheet erosion. However, there is no evidence of alluvial incision or deposition along the transect between the 2 profiles. The same is also true for evidence of creeping as there are no inclined irregular shrubs or trees, besides the fact that loose sand is not usually the medium in which creeping takes place. There are, nevertheless, indicators of sheet wash such as the presence of organic matter at different stages of decomposition behind the grass clumps and low shrubs, which leads us to assume that runoff is the main process taking place in the study area, although there are shallow riling channels too. Sand is also a very favourable material for infiltration which could occasionally diminish the effectiveness of the overland flows, leading to deposition. Even so, its intensity can increase in situations of saturation of the superficial horizons, even in sandy ones (Marchioro and Augustin, 2007).

Granulometry and AMS 14C dating

41When the results of granulometry (tab. 1) are put together with the calibrated ages of the AMS 14C dating of OM (tab. 3), it is possible to detect sharp boundaries between the different horizons in both profiles, even taking into consideration that it is a bulk 14C age, which means that it should be interpreted with caution because it can only represent a mean residence time (MRT) of the OM in the soil sample (Martin & Johnson, 1995).

42Table 3 shows that the older SOM, when considering the MRT, is found at the top of the weathered rock in both profiles, the oldest occurring in the top of the weathered rock in the deepest horizon of P2 (10,506-10,230 yr cal BP). Given that it may be the result of translocation of more recent material from upper horizons, one can assume that it may be even older. This could mean that either this layer was already exposed allowing deposition before the development of the subsequent ones, or that translocation took place accumulating the older material lower down. Furthermore, if P1, which is located in the upper slope nearer the outcrops, shows younger 14C dating for the SOM (5,916-5,664 yrs cal BP) than P2 at similar depths (60 cm), then there is an indication that this material has being renewed because erosion that had taken away the older material so that it is no longer on the upper slope.

43This interpretation can be reinforced by the data observed in table 3. It is clear that the upper horizons from both profiles, including the stony layer in P2, have more recent SOM (MRT) that is no older than approximately 700 years (P1: 639-519 yrs cal BP, and P2: 315-151 yrs cal BP). It could be associated with intense erosion that stripped away the “older” material and with the deposition of new SOM from the upper slopes. P1 also shows higher 14C values for the SOM (MRT) in the second horizon (2,335-2,152 yrs cal BP) than P2 (439-50 yrs cal BP) at a similar depth, corresponding to the organic (A1) and to the stony (A2) layers. This is interesting because if the above assumption is correct, the erosion was more effective at a certain time in P2, which is positioned further away from the quartzite outcrops than P1. It is possible to assume that P1, although presenting younger SOM (MRT) 14C dating in the deepest horizon, shows more stable environmental conditions for the development of the profile, with less sharp time boundaries (tab. 3).

Carbon stocks and the characteristics of the soil profiles

44Carbon stocks (tab. 2) are higher in P2 than in P1, with the highest concentration in the organic horizon (tab. 1 and tab. 2), which shows relatively recent SOM (MRT) (439-150 yrs cal BP). This occurs because in P2 the processes of OM accumulation were more intense. This may be related to both denser vegetation cover and climatic conditions that facilitated SOM decomposition in the past, or to its location in a slightly concave segment of the slope, which could enhance deposition of fine material. When this stock is seen in the context of the 14C dating, even considering that it represents a mean residence time of the SOM in the soil sample, it shows that the mobilization and transport by runoff have being occurring for a long time, and that when the superficial flow reduces velocity it tends to accumulate the material in small depressions. Although the interpretation of these values has to be made cautiously, it is acceptable to infer that there is a time asymmetry in the development of the horizons in both profiles, which may be better associated with changes promoted by geomorphic processes than by pedology.

Analysis of phytoliths

45For this analysis, it has to be taken into consideration that at a given soil depth, the phytolith assemblage results from a balance between erosion, deposition, continuous translocation and selective dissolution processes (Alexandre et al., 1997a). Therefore, soil phytolith analysis can provide paleoenvironmental information (e.g., Piperno and Becker, 1996; Blecker et al., 1997; Alexandre et al., 1997a) only if the four controls mentioned above can be assessed. In the present case, pedological processes were inferred from field and petrographical observations.

46Potential Holocene changes in tree cover density were discussed taking into account pedogenesis, changes with depth in phytolith assemblages, and especially changes in the D/P phytolith index indicative of tree cover density (Alexandre et al., 1997b; Bremond et al., 2005a; Barboni et al., 1999). Because soil organic matter (SOM) is composed of a complex mixture of organic molecules of different origins at various stages of decomposition, its bulk 14C analysis can only represent the mean 14C concentration of the soil sample.

47Taking this into account, we can assume that the results of the phytoliths and carbon isotopes (fig. 4, fig. 5 and tab. 2) provide evidence and helped to interpret the material/geomorphic processes acting on the slope. In both profiles, even in the surface horizons and in almost all modern assemblages collected in their vicinity, phytoliths are profoundly broken or weathered, with a greater ratio of unclassified than classified phytoliths in most samples, especially in Profile 1. This makes it difficult to identify the phytolith types and calculate the phytolith indices, but in our view highlights the intensity of both the current and previous surface processes in the study area.

48The results also show a predominance of large phytoliths, blocky, bulliform parallelepiped and elongate morphotypes, as they are more resistant to erosion or weathering, with low proportions or even an absence of the small short cell phytoliths, since they are less resistant to these processes (fig. 4, fig. 5, fig. 6 and tab. 2). These results indicate an intense mobilization and /or weathering of the material in the horizons above the weathered quartzite. Together with the interpretation of field observations, granulometry, 14C dating and 13C analysis, it is possible, however, to assume that runoff is responsible for the mobilization and transport of the phytoliths, and has been leading to their intense destruction and with only the more robust types surviving.

49It is also possible to see that the small phytoliths are more frequent in P2 than in P1, with the highest concentration in the organic horizon (fig. 4, fig. 5 and fig. 6). This occurs because in P2 processes of OM accumulation are more intense, as we can see through the analysis of soil organic carbon stocks (tab. 2). In P1, due to the proximity of the outcrop and less vegetation cover (fig. 2), transport by superficial runoff was, and remains, more effective, making the accumulation of organic matter difficult on that part of the slope, so that it tends to get spread out, accumulating in small depressions along the low inclined surfaces as is the case of P2.

50The stock of phytoliths of P1 follows a normal trend of decreasing with depth. P2 shows a very large increase in the stock of phytoliths in the organic horizon, which present recent SOM 14C dating (315-151 yrs cal BP), considering that it is a MRT (tab. 2 and tab. 3). It is worth mentioning that an increase in the stock of phytoliths doesn’t mean higher density of vegetation cover, since grasses can produce up to five times more phytoliths than the dicotyledonous plants (Coe et al., 2012a), as shown by the D/P index, which decreases in the organic horizon (fig. 4, fig. 5 and tab.. 2). It can however be related to processes of SOM accumulation and decomposition which can be controlled to some extent by climatic conditions.

51Some other trends of phytolith indices were observed in the two profiles (Chueng, 2012). For instance, the D/P ratio shows that in the surface horizons (tab. 2), the indices are characteristic of open vegetation cover. Although current lower tree density in P1 is slightly higher than in P2, both have values typical of savanna formations (Alexandre et al., 1997, Barboni et al., 1999), from 0.8 to 1.8 in P1 and from 0.02 to 0.12 in P2, with a predominance of grasses similar to that of today (fig. 2, fig. 4 and fig. 5). In the literature, D/P values ranging from 0.2 to 1 are obtained for vegetation woodier than grasslands but less woody than dry forests (Bremond et al., 2005b; Barboni et al., 1999), while D/P values close to or higher than 10 are found for rainforests (Alexandre et al., 1997b, 2011). The higher tree density of P1 can be explained by the presence of more arboreal vegetation cover on the upper slope, even taking into account that the trees are still very scarce (fig. 2). Its presence is most probably due to concentrations of higher soil moisture as a result of the proximity of the quartzite outcrop. Although the vegetation overlying the profile is practically composed of grasses, woody phytoliths present in P1 assemblages may also have originated higher up the slope and been transported to the profile without generating organic deposition, due to the reasons pointed out previously. In P2, which is located circa 200 m further down in a small depression, transport of larger phytoliths is less intense, and the phytolith assemblages reflect the overlying vegetation; the D/P of the surface sediments over the profile being only 0.02.

52In the superficial horizons, the Bi index is high and similar in both profiles (fig. 4, fig. 5 and tab. 2). In the underlying horizons, at a depth of about 40 cm, just before the stony layers, it is possible to observe the largest phytolith stocks and the lowest D/P in both profiles, as well as an increase in the Bi index in P1, suggesting the occurrence of a dryer environment during the deposition. In the stony layer the index decreases in P1, reaching its lowest value (tab. 2), indicating a more humid environment than in the overlying horizon (fig. 4). The presence of wetter conditions may be the result of either the increase in rainfall index in the past, or a better infiltration capacity of the stony layers forming a microclimate specific to this point. The isotopic results indicate that there is a variation in the deepest horizon of both profiles, suggesting the occurrence of a more humid environment, with less open vegetation and a predominance of C3 plants (tab. 2). This coincides with both the highest SOM 14C dating values (MRT) of P1 (5,919-5,152 yrs cal BP) and P2 (10,506-10,230 yrs BP). In these deepest horizons below 60 cm, at the top of the weathered quartzite, the results show that the phytoliths are very broken and it is mainly the larger types that are encountered. This is especially true for P2, where only 9% of the phytoliths are classified, making the phytolith signal weak, with only those more resistant to dissolution or erosion remaining intact. In this case, the reconstitution of vegetation using the phytolith proxy is impaired. If we associate the phytolith signal to the isotopic analysis results (13C), P1 and P2 are similar (tab. 2, fig. 4 and fig. 5); they showed almost no variations along the profiles, and the values in these horizons, from -17‰ to -22‰ in P1 and from -19‰ to -23‰ in P2, are characteristic of more closed vegetation, suggesting a more humid environment (fig. 4, fig. 5 and tab. 2). Moreover, phytoliths enabled the inference that in this period the erosion processes were also very pronounced, due to higher energy levels resulting from the presence of more superficial running water in the environment.

53The deposition of organic matter and fine material elsewhere in the Espinhaço seems to reinforce the hypotheses that the annual average rainfall amounts remained basically unaltered forming a more humid pattern for very long periods during the later Last Glacial Maximum in the southern Espinhaço Mountain Range (Augustin, 1995). This author suggests that there is evidence for assuming that although annual mean rainfall quantities remain similar to those of today, the annual rainfall distribution may have been different during the Upper Pleistocene with a more uniform distribution along the year than today. She concludes that most changes can be related to the lower annual average temperature of approximately 8oC less when compared with nowadays, which could have created the conditions for deeper decomposition of the organic matter commonly found in the Espinhaço area. This is also the conclusion of L.F.P. Barros et al. (2011) for the area.

54However, these occasional variations in humidity which could have occurred affected probably the vegetation cover in the study area. For instance, there is higher tree density in P1 than in P2, what could be associated with transport of phytoliths of small trees that even today grow on the weathered outcrops, especially in the fractures. Because P2 is further away, the contribution of this material is not so significant. Isotopic analysis of the material from the lowest horizon (C) of the P2 profile indicates the presence of denser vegetation cover suggesting more humid conditions and dominance of C3 type of vegetation. This could also be credited to the slight climate variations or to local environmental conditions, such as the presence of a slightly depression in the P2, as well as the effect of the stony layer that increases water percolation that tends to accumulate below, near the contact with the underlying rock. We assume that the lower values of small phytoliths associated with the high percentage of unclassified ones (9%) pointed towards the occurrence of very active runoff, which has apparently been active in the area since the end of the Upper Pleistocene.

55The small percentage of identifiable phytoliths may also be associated with the occurrence of more effective chemical weathering occurring today. This weathering is observable through the intense red mottling in the quartzite indicating that leaching and laterization are taking place, probably also affecting the phytoliths. However, as discussed previously, other evidence presented seems to indicate the dominance of the erosion/deposition processes in the distribution and deposition of the sandy material. Even so, the relative climatic and tectonic stability has favoured pedologic processes.

56The analysis of the material found in the horizons indicates differences in the type and characteristics of the soil material as well as in relation to SOM 14C-AMS dating (MRT), 13C values and phytolith types and conditions. The horizon sequences in both profiles points to the occurrence of more than pedologic differentiation due to translocation, addition, subtraction or other processes involved in the individualization of the horizons. Therefore, we can assume that the horizons present evidence of being colluvial material, including OM that has been eroded, transported and deposited and in which pedogenesis is taking place.

SOM 14C-AMS results

57SOM 14C-AMS results (tab. 3), considering MRT show also that these processes are recent in geological terms, which reinforces the interpretation of the sandy material being of a colluvial nature, as it has been eroded and transported by runoff from quartzite outcrops on the upper slope, and deposited along these relatively flat surfaces. In this sense they correspond to the colluvial ramps studied by M.R. Meis et al., 1979) in other lithologies in Brazil. The relative thinness of the material in the profiles can be related to the slow effectiveness of physical and chemical weathering processes taking place mainly at the quartzite outcrops and supplying the lower slopes with material. The highest value of the SOM 14C (MRT) from both profiles occurs at the top of the weathered rock horizon (C) in P2 (10,506-10,230 yr cal BP), which was probably already exposed at the surface even before the formation of P1 C horizon (5,919-5,664 yr cal BP).

58The dominant presence of extremely broken phytoliths, and the fact that only the more robust types are present along both profiles seems to support this interpretation (fig. 4, fig. 5, fig. 6 and tab. 2). The presence in both profiles of stony layers of pebbles of quartz milk with more or less the same angularity and size at similar depths apparently indicates the formation of a stone-line. The SOM 14C (MRT) results of the stony layers (tab. 3) seems to indicate the occurrence of a short dryer period in the recent past between 439 and 151 yrs BP (MST). The absence of organic material below the stony layer of P1 strengthens this possibility (tab. 1). It could be the result of short distance displacement of the pebbles and their transport by concentrated and disperse runoff in the form of sheet wash and rill erosion, as discussed by J. De Ploey and J. Poesen (1989) rather than in situ residuum of the erosion of fine material above it. Therefore, this does not exclude the possibility of the occurrence of runoff since there is no evidence for a long period of very dry hot climatic conditions prevailing in the area (Augustin, 1995; Rocha, 2011; Barros et al., 2011).

59The smaller thickness of the stone layer in P1 compared to P2 can be either associated with the more effectiveness of runoff or to the presence of two different stony veins (fig. 2 and tab. 1). Another possible explanation is that the pebbles were formed by the weathering of the same quartz vein in situ as a result of downwearing processes. The problem with this interpretation is that it would have required the occurrence of a very extensive quartz vein, spreading across the length between the two sites.

60The combination of type and sequence of material and 14C dating point to the prevailing humid conditions, which could account for the presence of fine organic rich sand and of highly weathered quartzite underlying rock with evidence of intense laterization process (fig. 3 and tab. 3).

61The results also allow the assumption that the material found on the relative flat surfaces is formed by deposition of weathered quartzite outcrops and that is therefore of a colluvial nature, corresponding to the colluvial ramps encountered by other authors in different lithologies in Brazil. The “flat” character of the landform is apparently inherited from the structural tabular mega stratification of the “Galho do Miguel” formation (Fogaça and Scholl, 1979) dated from Mesoproterozoic (1,600-1,000 Ma).

62In the case of study material overlying the ramp at the Serra do Engenho, there is no sound evidence to assume that it could have been formed in situ by chemical and physical weathering. Although the model of evolution for extensive residual planation surfaces is very often credited to chemical weathering, these models are very often developed on the basis of observation and research conducted on granite-gneiss landscapes (Austin, 1970; Twidale, 1980, 1997, 1999, 2000, 2014; Johnson, 1993; Thomas, 1994). In this model, chemical alteration of the rock would form a deep regolith that eventually would be stripped away, exposing the lower weathered rock that would again go through weathering and regolith formation again and again.

63In quartzite lithology, however, dissolution of quartz seems to be a very slow process, as observed by R.A.L. Wray (2003) and other authors that have studied quartz weathering and dissolution (Triahi and Rajamai, 1999; Martini, 1979; Twidale, 1980; Turner et al., 2003; Schultz and White, 1999; Fabri et al., 2013). They are unanimous in pointing to the long time necessary to dissolve quartz under laboratory and natural conditions. This is also the conclusion of L. Piccini and M. Mecchia (2009) in relation to the development of the Auyan-tepui, in the sandstone of the Gran Sabana, in Venezuela. Considering the low response to chemical weathering of quartz, these authors suggest an average age of 30 Ma for the development of the Venezuelan Karst. Therefore, it can be assumed that in the study area, where the underlying quartzite rock presents up to 95% quartz in its composition (Fogaça and Scholl, 1979), chemical weathering is very slow. Even taking into account the presence of favorable conditions of humidity for long geological periods, pH below 6 and the presence of high ground water levels, which are believed to be essential for quartz chemical weathering (Schultz and White, 1999), it could take longer than the period covered by the 14C method.

64The dissolution model does not fit so well with the explanation for the material on the study ramp, partially due to the evidence provided by the type of material forming the profile sequences, which shows characteristics of erosion and transportation. Also 10Be results from the research of H.N. Barreto et al. (2013) indicate a very low rate of denudation (< 5 m/My) for the Espinhaço Mountain Range, which may reinforce the fact that the weathering rates as well as erosion are after all very low. For instance, the material found on the colluvial ramp is relatively thin (90 cm) for a deposition which occurred initially considering SOM MRT in the Upper Pleistocene/Earlier Holocene (10,506-10,230 yrs cal BP). This can be associated either with low weathering rates of the outcrop or with inefficiency in the transportation of the material downhill by runoff, due to the low inclination of the ramp, or even to the efficiency of erosion and transportation. In that case, the rate of incoming material from the outcrop is matched by the rate of erosion, which would maintain the depth of the material more or less constant.

65It is important, however, to point out that similar flat areas surrounding quartzite outcrops in the Espinhaço may have their development associated with other processes such as silica dissolution, which has to be further investigated since the development of karst features has already been identified in the Espinhaço region (Willems et al., 2004).

Conclusion

66The combined results of geomorphic analysis and field observation, granulometry, 14C, 13C and phytolith analysis seems to indicate that the material forming relatively flat surfaces in the Espinhaço massif is of a colluvium nature and its sources are quartzite outcrops localized on the upper slopes. Physical and chemical weathering processes generated material that has been further eroded and transported by different types of runoff and deposited together with OM along the relatively flat surface developed near quartzite outcrops in the study area. This corresponds to a colluvial ramp developed on siliciclastic rocks and is similar to those reported in other lithologies in Brazil.

67Physical and chemical weathering as well as removal of this material seem to be very slow since the thickness of the P2 deposited material, at the base of which 14C dating of the SOM (MRT) is at least 10.250 yr cal BP, does not reach 1m. This is also the conclusion from 10Be results showing the occurrence of a very slow rate of denudation in the Espinhaço Mountain Range (Barreto et al., 2013).

68Furthermore, the high rate of unclassified phytoliths could possibly be related to the effect of erosion, although the reconstruction of the vegetation in the area only through the proxy phytoliths in some cases, such as in the deepest horizons, showed limitations. However, even where the phytolith signal was weaker, it was complemented and supported by isotopic analyses. For other samples, such as the three upper horizons, the results of isotopic analyses showed no significant variations, while the phytolith results provided very interesting information for the reconstruction of paleoenvironments where the geomorphologic processes occurred.

6914C dating together with 13C and phytoliths allowed the assumption that there have not been very strong and meaningful alterations in the rainfall amount to account for significant changes in the deposition pattern of the material or even changes in the types of weathering or erosive processes during the time of the formation of the material of the colluvial ramps since the end of the Upper Pleistocene in the area. These results also indicate that the temperature during the latter Pleistocene and the beginning of Holocene may have been cooler, accounting for the complete decomposition of the organic material. The lack of evidence for important changes in rainfall amounts and the occurrence of lower temperatures during the upper Pleistocene and early Holocene coincide with other results for the area (Augustin, 1995), proposing that relative environmental and also tectonic stability allowed pedogenic processes to take place, which developed Quartz-sand Neosols.

70The combined analysis indicated enough evidence to suggest that the material encountered on the flat study areas in the Serra do Engenho in the Espinhaço Meridional mountain range is the result of deposition of sediments eroded and transported by runoff from nearby quartzite outcrops, where they are loosened by physical and chemical weathering. However, these results do not exclude the possibility that the development of similar relatively flat surfaces in the area is related to in situ chemical weathering.

71It can be also concluded that the phytolith and isotopic analyses applied to studies in Geomorphology, showed promise for the understanding of landscape evolution and environmental changes, especially when associated with other indicators in multiproxy analyses in order to obtain a more precise inference of these changes.

Haut de page

Bibliographie

Alexandre A., Bouvet M., Abbadie L. (2011) – The role of savannas in the terrestrial Si cycle: A case-study from Lamto, Ivory Coast. Global and Planetary Change 78, 162-169.

Alexandre, A., Meunier, J.D., Colin, F., Koud, J.M. (1997a) Plant impact on the biogeochemical cycle of silicon and related weathering processes. Geochimica et Cosmochimica Acta, 61(3), 677-682.

Alexandre A., Meunier J.D., Lézine A.M, Vincens A., Schwartz D. (1997b) – Phytoliths: indicators of grassland dynamics during the late Holocene in intertropical Africa. Palaeogeography, Palaeoclimatology, Palaeoecology 136 (1-4), 213-229.

Almeida Abreu P.A., Renger F.E. (2002) – Serra do Espinhaço Meridional: um orógeno de colisão do Mesoproterozóico. Revista Brasileira de Geociências 32 (1), 1-14.

Alvarenga A.F.B. (1985)Uso e ocupação do solo na porção Norte ocidental da Cidade de Gouveia, MG. TCC, Department of Geography, Geosciences Institute, Universidade Federal de Minas Gerais, Belo Horizonte, MG, Brazil, 61 p.

Asumadu K., Gilkes R.J., Armitage T.M., Churchward H.M. (1988) – The effects of chemical weathering on the morphology and strength of quartz grains-an example from S.W. Australia. Journal of Soil Science 39, 375-383.

Augustin C.H.R.R. (1979)A preliminary integrated survey of the natural resources near Alcantarilla, Southeast Spain. MSc Thesis, University of Sheffield, UK, 328 p.

Augustin C.H.R.R. (1985) – A Geografia Física: o levantamento integrado e avaliação de recursos naturais. Geografia Teorética 15 (29-30), 141-153.

Augustin C.H.R.R. (1995)Geoökologische Studien im südlichen Espinhaçogebirge bei Gouveia, Minas Gerais, Brasilien unter besonder Berücksichtigung der Landaschaftsentwicklung. PhD Thesis. Johann Wolfgang Goethe Universität, Germany, 294 p.

Augustin C.H.R.R., Fonseca B.M., Rocha L.C. (2011) – Mapeamento geomorfológico da Serra do Espinhaço Meridional: primeira aproximação. Geonomos 19 (2), 50-69.

Austin G.S. (1970) – Weathering of the sioux quartzite near New Ulm, Minnesota as related to cretaceous climate. Journal of Sedimentary Petrology 40 (1), 184-193.

Barboni D., Bonnefille R., Alexandre A. Meunier J.D. (1999) – Phytoliths as paleoenvironmental indicators, West Side Middle Awash Valley, Ethiopia. Palaeogeography, Palaeoclimatology, Palaeoecology 152, 87–100.

Barreto H.N., Varajão C.A.C., Braucher R., Bourlès D.L., Salgado A.A.R., Varajão A.F.D.C. (2013) – Denudation rates of the Southern Espinhaço Range, Minas Gerais, Brazil, determined by in situ-produced cosmogenic beryllium 10. Geomorphology 191, 1–13.

Barros L.F.P, Lavarini C., Lima L.S., Magalhães Júnior A. P. (2011) – Síntese dos cenários paleobioclimáticos do Quaternário Tardio em Minas Gerais/ sudeste do Brasil. Soc. & Nat. 23 (3), 371–386.

Bishop P. (2007) – Long-term landscape evolution: linking tectonics and surface processes. Earth Surf. Process. Landforms 32, 329–365.

Blecker, S.W., Yonker, C.M., Olson, C.G., Kelly, E.F. (1997) Paleopedologic and geomorphic evidence for Holocene climate variation, Shortgrass Steppe, Colorado, USA. Geoderma 76, 113-130.

Bourne R. (1931)Regional survey and its relation to stocktaking of the agricultural and forest resources of the British Empire. Oxford Forestry Memoirs 13. Clarendon Press, Oxford". 169 p.

Bremond L., Alexandre A., Hély C., Guiot J. (2005a) A phytolith index as a proxy of tree cover density in tropical areas: calibration with Leaf Area Index along a forest-savanna transect in southeastern Cameroon. Global and Planetary Change 45(4), 277-293.

Bremond L., Alexandre A., Peyron O., Guiot J. (2005b) Grass water stress estimated from phytoliths in West Africa. Journal of Biogeography 32, 311-327.

Bremond L., Alexandre A., Wooler M.J., Hély C., Willianson D., Schaëffer P., Majule A., Guiot J. (2008) Phytolith indices as proxies of grass subfamilies on East African tropical mountains. Global and Planetary Change 61, 209-224.

Brichta A., Paternoster K., Scholl W.U., Turinsky F. (1980) – Die Gruta do Salitre bei Diamantina, Minas Gerais, Brasilien, Kein “Einsturzloch”. Zeitschrift für Geomorphologie 24, 236-242.

Bronk Ramsey C. (2009) Bayesian analysis of radiocarbon dates. Radiocarbon 51 (1), 337–360.

Büdel J. (1970) Pedimente, Rumpflächen and Rückland-Steilhänge. Zeitschrift für Geomorphologie 14 (1057), 1-8.

Chueng K.F. (2012)Inferência da Cobertura Vegetal e das Condições Climáticas no Espinhaço Meridional, MG, durante o Quaternário através dos Indicadores Fitólitos e Isótopos de Carbono. TCC, Geography Department, Universidade do Estado do Rio de Janeiro, RJ, Brazil, 112 p.

Coe H.H.G. (2009)Fitólitos como indicadores de mudanças na vegetação xeromórfica da região de Búzios/Cabo Frio, RJ, durante o Quaternário. PhD Thesis, Geology Department, Universidade Federal Fluminense, Niterói, RJ,Brazil, 300 p.

Coe H.H.G., Alexandre A., Carvalho C.N., Santos G.M., Silva A.S., Sousa L.O.F., Lepsch I.F. (2012a) Changes in Holocene tree cover density in Cabo Frio (Rio de Janeiro, Brazil): Evidence from soil phytolith assemblages. Quaternary International 2, 1-10.

Coe H.H.G., Chueng K.F., Gomes J.G. (2012b) – Reconstituições da Vegetação e Inferências de Paleoclimas através da Utilização dos Indicadores Fitólitos e Isótopos de Carbono. Exemplos de Estudos no Brasil. Revista Geonorte 1 (4), 248-261.

Coe H.H.G., Gomes J.G., Chueng K.F. (2013) – Exemplos de reconstituições da vegetação e inferências de paleoclimas no Estado do Rio de Janeiro através da utilização de biomineralizações de sílica (fitólitos) e isótopos de carbono. Revista Tamoios 9, 1-21.

De Ploey J., Poesen J. (1989) – Experimental geomorphology and the interpretation of stone lines. In: Alexandre J. and Symoens J.J. (Eds): Stony-Lines. Journée d’étude, Bruxelles, 24, Académie Royale des Sciences d’Outre Mer. Geo-Eco-Trop 11, 75-82.

Duchaufour P. (1982) Pedology. London, George Allen & Unwin. 447 p.

EMBRAPA Centro Nacional de Pesquisa de Solos (Rio de Janeiro, RJ) (1997)Manual de métodos de análise do solo. 2 ed., Rio de Janeiro, 212 p.

Fabri F., Auler A., Calux A., Cassimiro R., Augustin C.H.R.R. (2013) – Geoespeleologia e principais aspectos espeleogenéticos das grutas Baixada das Crioulas I E II, Itambé do mato dentro, Minas Gerais. Geonomos 21(1), 56-62. www.igc.ufmg.br/geonomos.

Fogaça A.C. (1997) – Geologia da Folha Diamantina. In Grossi-Sad J.H., Lobato L.M., Pedrosa-Soares A.C., Soares-Filho B.S., Espinhaço Project in CD-ROM (text, maps and appendix). Belo Horizonte, COMIG – Companhia Mineradora de Minas Gerais, 1575-1665.

Fogaça A.C., Scholl W.U. (1979) – Estratigrafia na Serra do Espinhaço na região de Diamantina. In I Simp. Geol. Minas Gerais, Annais, Diamantina, MG. Brazil, 55-73.

Fonseca B.M., Augustin C.H.R.R. (2014) – Análise morfométrica de bacias de drenagem e sua relação com a estrutura geológica, serra do Espinhaço Meridional–MG. Revista Brasileira de Geomorfologia 15, (2), 153-172.

Fredlund G.G., Tieszen L.L. (1994) – Modern phytolith assemblages from the North American Great Plains. J. Biogeogr. 21, 321-335.

Gleixner G., Poirier N., Bol R., Balesdent J. (2002) – Molecular dynamics of organic matter in a cultivated soil. Organic Geochemistry 33 (3), 357-366.

Gomes J.G. (2012)Contribuição para a Compreensão da Gênese e Evolução de Solos na Bacia do Rio São João, RJ, através de análises fitolíticas. TCC, Geography Department, Universidade do Estado do Rio de Janeiro, RJ, Brazil, 137 p.

Gordon E. S., Goñi M. A. (2003) Sources and distribution of terrigenous organic matter delivered by the Atchalaya River to sediments in the northern Gulf of Mexico. Geochimica et Cosmochimica Acta 67(13), 2359-2375.

Grigore M.N., Toma C., Boscaiu M. (2010) Ecological implications of bulliform cells on halophytes, in salt and water stress natural conditions. Analele stiinþifice ale Universitãþii “Al. I. Cuza” Iasi, LVI, 5-15.

Geological Society of America (2011)www.geosociety.org/science/timescale/timescl.pdf.

Heather L.B., Sak P.B., Webb S.M., Brantley S.L. (2008) – Weathering of the Rio Blanco quartz diorite, Luquillo Mountains, Puerto Rico: Coupling oxidation, dissolution, and fracturing. Geochimica et Cosmochimica Acta 72, 4488–4507.

Hedges J.I., Parker P.L. (1976) – Land-derived organic matter in surface sediments from the Gulf of Mexico. Geochimica et Cosmochimica Acta 40, 1019-1029.

Johnson D.L. (1989) – Subsurface Stone lines, Stones Zones, Artifact-manuport Layers, and Biomantles Produced by Bioturbation via Pocket Gophers (Thomomys Bottae). American Antiquity 54, 370-389.

Johnson D.L. (1993) – Temperate landscapes with three tiered soils: toward a general theory of landscape evolution. Quaternary International 17, 67-78.

Kaplan L., Smith M.B., Sneddon L.A. (1992) Cereal grain phytoliths of Southwest Asia and Europe. Phytoliths systematics, emerging issues. In Rapp G.J., Mulholland S.C. (Eds.)., Advances in Archaeological and Museum Science, New York, 149-174.

Kelly E.F. (1990) Method for extracting opal phytoliths from soil and plant material. Workshop on Biotic Indicators of Global Change, University of Washington, June 1-3 1990, 1-11.

Killops S., Killops V. (2005) – Introduction to organic geochemistry. Malden, MA: Blackwell, 393 p.

King L. (1956) – A Geomorfologia do Brasil Oriental. Revista Brasileira de Geografia 18 (1), 147 p.

Kondo R., Childs C., Atkinson I. (1994)Opal Phytoliths of New Zealand. Manaaki Whenua Press, 85 p.

Krauss D.A., Kaplan L., Strauss E. (1997)The use of phytolith analysis in paleoenvironmental reconstruction and environmental management; a case study on Martha's Vineyard, MA. Proceedings of the Twenty-Fourth Annual Conference on Ecosystems Restoration and Creation, Hillsborough Community College, Plant City, May 1997, 116–131.

Linton D.L. (1951) – The delimitation of morphological regions. In Stamp L. and Wooldridge S.W. (Eds), Essays in Geography, London: Longman, 199-218.

Madella M., Alexandre A., Ball T. (2005) – International Code for Phytolith Nomenclature 1.0. Annals of Botany 96, 253-260.

Marchioro H., Augustin C.H.R.R. (2007) – Dimensões de parcelas experimentais: influência do escoamento superficial e erosão do solo em Gouveia/MG. Geografias, 3(2), 7-16.

Martin C.W., Johnson W.C. (1995) Variation in radiocarbon ages of soil organic matter fractions from late Quaternary buried soils. Quaternary Research 43(2), 232–237.

Martinelli LA., Ometto J.P.H.B., Ferraz S.E, Victoria L.R., Camargo P.B., Moreira M. Z. (2009)Desvendando questões ambientais com isótopos estáveis. Oficina de Textos. São Paulo. 144 p.

Martini J.E.J. (1979) – Karst in Black Reef Quartzite near Kaapsehoop, EasternTransvaal. Annals of the South African Geological Survey 13, 115- 128.

McCormac F.G., Hogg A.G., Blackwell P.G., Buck C.E., Higham T.F.G., Reimer P.J. (2004) – SHCal04 Southern Hemisphere Calibration, 0–11.0 cal kyr BP. Radiocarbon 46 (3), 1087-1092.

Medina B.M.O. (2004)Potencial Regenerativo dos solos da Serra do Cipó, Minas Gerais. MSc Dissertation. Instituto de Ciências Biológicas. Universidade Federal de Minas Gerais, Belo Horizonte, Brazil. 61 p.

Meis M.R., Machado M.B., Cunha S.B. (1979) – Upper Quaternary ramps: Rio Doce River Valley, southern easten Brazilian Plateau. Zeitschrift für Geomorphologie 23 (2), 131-151.

Mercader J., Bennett T., Esselmont C., Simpson S., Walde D. (2009) – Phytoliths in woody plants from the Miombo woodlands of Mozambique. Ann. Bot. 104, 91–113.

Messager E., Lebreton V., Marquer L., Russo-Ermolli E., Orain R., Miskovsky R.J., Lordkipanidze D., Despriée J., Peretto C., Arzarello M. (2010) – Palaeoenvironments of early hominins in temperate and Mediterranean Eurasia: new palaeobotanical data from Palaeolithic key-sites and synchronous natural sequences. Quat. Sci. Ver. 30, 1–9.

Mulholland S.C. (1989) – Phytolith Shape Frequencies in North Dakota Grasses: A Comparison to General Patterns. Journal of Archaeological Science 16, 489-511.

Nimer E. (1989)Climatologia do Brasil. 2 ed. Fundação IBGE. Rio de Janeiro, 421p.

Oliveira A.P., Ker J.C., Silva I.R., Fontes M.P.F., Oliveira A.P, Neves T.G. (2010) – Spodosols pedogenesis under barreiras formation and sandbank environments in the south of Bahia, Rev. Bras. Ciênc. Solo, 34(3), 847-860.

Oliveira J.B., Klinger T.J., Camargo M.N. (1992)Classes gerais de solos do Brasil – guia auxiliar para seu reconhecimento. 2 ed., FUNEP, 201p.

Ollier C. D. (2010) – Very deep weathering and related landslides. Geological Society, London, Engineering Geology Special Publication 23, 5-14.

Palmer P.G., Gerbeth-Jones S., Hutchison S. (1985) A scanning electron microscope survey of the epidermis of East African grasses. III Smithsonian Contribution to Botany 55, 1-136.

Peacock E., Fant D.W. (2002) Biomantle Formation and Artifact Translocation in Upland Sandy Soils: An Example from the Holly Springs National Forest, North-Central Mississippi, U.S.A. Geoarchaeology: an International Journal 17 (1), 91–114.

Peres G.G., Alkimim F.F., Jordt-Evangelista H. (2004) – The southern Araçuaí belt and the Dom Silvério Group: geologic architecture and tectonic significance. An. Acad. Bras. Ciênc. [online], 76 (4), 771-790. ISSN 0001-3765. http://dx.doi.org/10.1590/S0001-37652004000400011.

Pessenda L.C.R., Boulet R., Aravena R., Rosolen V., Gouveia S.E.M, Ribeiro A.S, Lamotte M. (2001) – Origin and dynamics of soil organic matter and vegetation changes during the Holocene in a forest–savanna transition zone, Brazilian Amazon region. Holocene 11 (2), 250–254.

Peulvast J.P., Sales V.C. (2002) – Aplainamento e Geodinâmica: Revisitando um problema clássico em Geomorfologia. Mercator – Revista de Geografia da UFC, 1(1), 113-150.

Piccini L., Mecchia M. (2009) – Solution weathering rate and origin of karst landforms and caves in the quartzite of Auyan-tepui (Gran Sabana, Venezuela). Geomorphology 106, 15–25.

Piperno D.R. (1985) Phytolith analysis and tropical paleo-ecology: Production and taxonomic significance of siliceous forms in new world plant domesticates and wild species. Review of Palaeobotany and Palynology 45(3-4), 185-228.

Piperno D.R. (1988) Phytoliths Analysis: an archaelogical and geological perspective. Academic Press, San Diego, 280 p.

Piperno D.R. and Becker P. (1996) Vegetacional History of a Site in the Central Amazon Basin Derived from Phytolith and Charcoal Records from Natural Soils. Quaternary Research 45(2), 202-209.

Rocha L.C. (2011)Evolução de longo-termo da Serra do Espinhaço Meridional: um estudo sobre o modelado do relevo. PhD Thesis, Instituto de Geociências, Universidade Federal de Minas Gerais, Belo Horizonte, M.G., Brazil, 179 p.

Runge F. (1999) – The opal phytolith inventory of soils in central Africa – quantities, shapes, classification, and spectra. Review of Palaeobotany and Palynology 107 (1-2), 23-53.

Runge F., Fimbel R. (1999) Opal phytoliths as evidence for the formation of savanna islands in the rain forest of Southeast Cameroon. In Heine K., Runge G.E.J. (Eds.), VXth INQUA Conference (15th: 1999: Durban South Africa). Palaeoecology of Africa and the Surrounding Islands. International Union for Quaternary Research, Tokyo, 171-189.

Saadi A. (1995) – A Geomorfologia da Serra do Espinhaço em Minas Gerais e de suas margens. Geonomos, 3 (1), 41-63.

Santos R.D., Lemos R.C., Santos H.G., Ker J.C., Anjos L.H.C. (2005)Manual de descrição e coleta de solo no campo. Embrapa, Sociedade Brasileira de Ciência de Solo, 92 p.

Santos G.M., Ormsby K. (2014) Behavioral variability in ABA chemical pretreatment close to the 14C age limit. Radiocarbon 55, 3–4.

Santos G.M., Southon J.R., Griffin S., Beaupre S.R., Druffel E.R.M. (2007a) Ultra small-mass AMS 14C sample preparation and analyses at KCCAMS/UCI Facility. Nucl. Inst. Methods Phys. Res. B 259, 293–302.

Santos G.M., Moore R.B., Southon J.R., Griffin S., Hinger E., Zhang D. (2007b) AMS 14C sample preparation at the KCCAMS/UCI Facility: status report and performance of small samples. Radiocarbon 49 (2), 255–269.

Schultz M.S., White A.F. (1999) – Chemical weathering in a tropical watershed, Luquillo Mountains, Puerto Rico III: Quartz dissolution rates. Geochimica et Cosmochimica Acta 63(3-4), 337–350.

Scurfield G., Anderson C.A., Segnit E.R. (1974) Silica in woody stems. Australian Journal of Botany 22, 211–229.

Strömberg C.A.E. (2004) Using phytolith assemblages to reconstruct the origin and spread of grass-dominated habitats in the great plains of North America during the late Eocene to early Miocene. Palaeogeography, Palaeoclimatology, Palaeoecology 207, 239-275.

Thomas T.F. (1994)Geomorphology in the Tropics. New York: John Wiley & Sons, 460 p.

Triahi J.K., Rajamai V. (1999) – In situ maturation of sediments within the weathering profiles: an evidence from REE behavior during weathering of Delhi quartzites. Current Science 76(9), 1255-1258.

Tucker G.E., Hancock G.R. (2010) – Modelling landscape evolution. Earth Surf. Process. Landforms 35, 28–50.

Turner B.F., Stallard R.F., Brantley S.L. (2003) – Investigation of in situ weathering of quartz diorite bedrock in the Rio Icacos basin, Luquillo Experimental Forest, Puerto Rico. Chemical Geology 202, 313-341.

Twidale C.R. (1980) – Origin of minor sandstone landforms. Erdkunde Band, 34, 219-224.

Twidale C.R. (1997) – Models of landscape evolution and the survival of paleoforms. Caderno Lab. Xeolóxico de Laxe Coruña, 22, 151-182.

Twidale C.R. (1999) Landforms ancient and recent: the paradox. Geog. Ann. 81A, 431–441.

Twidale C.R. (2000) Early Mesozoic (?Triassic) Landscapes in Australia: Evidence, Argument, and Implications. The Journal of Geology 108, 537–552.

Twidale C.R. (2014) Pediments and platforms: problems and solutions. Géomorphologie : relief, processus, environnement, 1, 43-56.

Twiss, C. (1969) Dust deposition and opal phytoliths in the Great Plains. Transactions of the Nebraska Academy of Sciences, XI (special issue): 73-82.

Twiss C. (1992) – Predicted world distribution of C3 and C4 grass phytoliths. In Rapp G.J., Mulholand S.C. (Eds.), Phytolith Systematics. New York: Plenum Press, 113-128.

Vrydaghs L., Doutrelepont H. (2000) Analyses phytolithariennes: acquis et perspectives. In Servant-Vildary S., Servant M. (Eds.), Dynamiques à long terme des écosystèmes forestiers intertropicaux. UNESCO, Paris, 389-398.

Willems L., Rodet J., Pouclet A., Rodet M.J., Hartert F., Compère P., Auler A.S. (2004) Carste em Quartzito da Região de Diamantina: Gruta do Salitre e Parque Estadual do Rio Preto, Minas Gerais. I Encontro Brasileiro de Estudos do Carste. Caderno de Resumos, 86. Poster. Belo Horizonte, MG.

Wray R.A.L. (2003) – Quartzite dissolution: karst or pseudokarst? Speleogenesis and Evolution of Karst Aquifers. Speleogenesis and Evolution of Karst Aquifers 2, 1-9.

Wright R.L. (1973) – An examination of the value of site analysis in the field studies in tropical Australia. Zeitschrift für Geomorphologie 17, 157-18.

Xu X., Trumbore S.E., Zheng S., Southon J.R., McDuffee K.E., Luttgen M., Liu J.C. (2007) – Modifying a sealed tube zinc reduction method for preparation of AMS graphite targets: reducing background and attaining high precision. Nucl. Inst. Methods Res. B 259, 320–329.

Haut de page

Annexe

Version française abrégée

Cette recherche contribue à la compréhension de l'évolution géomorphologique des surfaces relativement plates couramment rencontrées près des affleurements de quartzite de la « Serra do Espinhaço » dans l'État du Minas Gerais, au Brésil, par l'étude de la nature du matériau sableux qui les recouvre. L'étude de ces matériaux peut offrir une bonne explication de l'évolution des formes de relief dans les zones à substrat quartzitique, en particulier dans les environnements tropicaux. Bien qu'ils soient très récurrents dans le paysage brésilien, formant les montagnes les plus élevées, ils n'ont pas été très étudiés. Nous pensons que ce type d'études peut contribuer à améliorer l'intérêt des modèles d'évolution des massifs anciens dans les zones tropicales humides, généralement associés à l’altération chimique prépondérante des roches granito-gneissiques.

La zone échantillonnée est située à la limite ouest de la « Serra do Engenho », dans la partie sud de la chaîne de montagnes de l'Espinhaço (municipalité de Gouveia, Etat de Minas Gerais, fig. 1). Elle comprend une surface plane peu inclinée (3º, une « rampe ») pas très vaste (environ 300 m de long) mais représentative de ce type de surfaces. On a privilégié une approche géomorphologique pour l'analyse de deux sites sur lesquels deux profils de sol ont été décrits et échantillonnés (huit prélèvements) pour des analyses en laboratoire (fig. 2). En plus des informations géomorphologiques et la caractérisation granulométrique du matériau, des analyses de la matière organique, des phytolithes et des datations par 14C de la matière organique du sol (MOS) ont été réalisées. Ces analyses permettent de corréler les réponses de la paléovégétation aux conditions de l’environnement au moment de la formation et du dépôt des phytholithes. Ces types d’analyse ont été récemment appliquées à des études pédologiques. L'étude des phytolithes, complétée par des analyses isotopiques (δ13C et 14C), s'est avérée très utile pour la reconstitution paléoenvironnementale des processus géomorphologiques.

Le profil de sol P1 est situé dans la partie haute de la pente, à environ 30 mètres des affleurements quartzitiques. Le profil de sol P2 a été échantillonné dans un secteur légèrement concave (fig. 2). Ce sont deux néosols quartzo-sableux qui se structurent en quatre minces horizons constitués d'un matériau organique sableux, contenant des quantités variables de matière organique (fig. 3). Les horizons superficiels sont principalement constitués de sable fin et d'argile, ce qui est fréquent dans les environnements tropicaux humides. La base des deux profils est formée d'un niveau de cailloux de quartz angulaires à subangulaires, qui repose presque directement sur la quartzite altérée des horizons C, où on a retrouvé les valeurs de datations radiocarbone de la MOS (matière organique du sol) les plus anciennes. Compte tenu du temps moyen de résidence de la MOS, il est possible d'avancer que les premiers dépôts de P2 sont âgés de plus de 10 000 ans (10506-10230 ans cal BP), le premier profil (P1), situé plus haut sur la pente, présentant la MOS la plus récente (5919-5664 ans cal BP).

Les résultats des analyses des phytolithes montrent que, d’une manière générale, les stocks sont plus importants en P1 qu’en P2. Le stock de phytolithes du profil P1 suit une tendance normale en diminuant avec la profondeur, tandis que sur le profil 2, on observe une très forte augmentation du stock de phytolithes dans l'horizon organique supérieur. Sur les deux profils, les phytolithes sont très altérés et fragmentés. Ainsi, sur tous les horizons, à l’exception de l’horizon organique du profil 2, les phytolithes indéterminés (unclassified) sont en nombre plus important que les phytolithes identifiés (classified). Les résultats indiquent également une prédominance des morphotypes de phytolithes les plus grands (blocky, bulliform parallelepiped, elongate), car ils sont plus résistants à l’altération, et, à l'opposé, les phytolithes plus petits (short cells), moins résistants, sont bien moins abondants, voire absents. Cela peut indiquer une mobilisation et/ou une altération intense du matériau dans les horizons formés au-dessus de la quartzite altérée. Si l’on prend en compte les observations de terrain, les analyses granulométriques et isotopiques, on peut supposer que le ruissellement superficiel soit responsable de la mobilisation et du transport des phytolithes, ce qui a provoqué pour partie leur destruction, surtout en ce qui concerne les morphotypes les plus fragiles. On peut également observer que les phytolithes de plus petite taille sont plus abondant sur le profil P2 que sur le profil P1, en particulier dans l'horizon organique. Cela s'explique probablement parce que dans P2 les processus d'accumulation de la matière organique sont plus intenses, comme on a pu le constater lors de l'analyse des données de stocks de carbone organique dans le sol. En P1, en raison de sa proximité avec l'affleurement quartzitique et d'une couverture végétale moins dense, le transport par ruissellement superficiel s'avère plus efficace, ce qui rend l'accumulation de matière organique difficile sur cette partie de la pente, de sorte qu'elle tend à s’étaler et à s'accumuler dans les petites dépressions se développant le long des surfaces faiblement inclinées, comme c'est le cas du profil P2.

En ce qui concerne les indices phytolithiques, les mêmes tendances ont été observées dans les deux profils. L’indice D/P (densité de couverture arborée) présente des valeurs typiques des formations de savane (Alexandre et al., 1997, Barboni et al., 1999), avec une prédominance de graminées. Les valeurs sont plus élevées en P1 qu’en P2, ce que l'on peut probablement expliquer par la présence de quelques arbres, bien que rares, sur la partie supérieure de la pente. Cela est certainement du à une plus grande humidité contenue dans le sol, en raison de la proximité de l'affleurement de quartzite. Bien que la végétation qui recouvre le profil soit composée exclusivement de graminées, les phytolithes de plantes ligneuses présents dans les assemblages de P1 peuvent également provenir des parties plus hautes de la pente pour être transportés jusqu’au profil sans générer de dépôts organiques, pour les raisons signalées précédemment. Sur le profil P2, qui est situé 200 m en contrebas dans une petite dépression, le transport de phytolithes de plus grande taille est moins important, et les assemblages de phytolithes reflète la végétation sus-jacente. L'indice Bi (stress hydrique) est élevé et similaire dans les deux profils. Sur le profil P1, cet indice atteint sa plus faible valeur dans le niveau caillouteux, ce qui peut indiquer un environnement plus humide que dans l'horizon sus-jacent. La présence de conditions plus humides peut être le résultat d’une augmentation des précipitations dans le passé, ou alors d’une meilleure capacité d'infiltration du niveau caillouteux.

Les résultats isotopiques indiquent aussi la prédominance d'un couvert végétal ouvert sur les deux profils, avec une tendance à l'appauvrissement en δ13C dans les couches plus profondes des deux profils. Ceci suggère la présence d'un environnement plus humide, avec une végétation moins ouverte et la prédominance de plantes à métabolisme photosynthétique du type Calvin-Bresson (C3). Dans ces couches plus profondes et aussi les plus anciennes, le signal phytolithique est très faible à cause de la dégradation des phytolithes. Ces processus d'érosion très prononcé sont peut-être dus à des niveaux d'énergie élevés, résultant de la présence d’une disponibilité plus grande d'eau courante dans l'environnement.

Les résultats de toutes les analyses suggèrent ainsi que les processus d'érosion dans cette partie de la Serra do Engenho ont été très intenses, et que les couches plus profondes ont peut être déjà été en surface, lorsque le matériau s’est déposé. En outre, la grande quantité de phytolithes inclassables, en raison de leur altération et de leur fragmentation, ainsi que la prédominance des types les plus robustes, peuvent être interprétées comme la preuve d'une érosion intense et du transport du matériau par le ruissellement superficiel concentré et dispersé, suggérant que ce matériau est colluvial et que son origine est l’affleurement de quartzite situé à proximité. Après son dépôt le long du versant légèrement inclinée, qualifiée de « rampe colluviale », le matériau a été relativement préservé, ce qui a permis l'occurrence des processus pédogénétiques, et le développement de néosols quartzo-sableux. La rampe elle-même semble être le résultat d'une très faible altération et remobilisation du matériau, avec un fort contrôle des macro-structures rocheuses sous-jacentes. Cependant, les résultats ne sont pas complètement concluants, car les caractéristiques du matériau, ainsi que celles des phytolithes, peuvent également être liées à l'altération chimique in situ.

En raison du pourcentage élevé de phytolithes inclassables, que l'on peut éventuellement associer à l'érosion, la reconstitution de la végétation régionale par la seule caractérisation des phytolithes a parfois montré, comme c'est le cas pour les horizons les plus profonds, les limites de l'interprétation. Cependant, même lorsque le signal phytolithique s'est avéré moins manifeste, la recherche a été complétée par des analyses isotopiques. L'altération physico-chimique et la remobilisation du matériau semblent avoir été très lentes, l'épaisseur du matériau déposé au-dessus de la couche la plus ancienne (datation 14C de la MOS -MRT, profil P2), qui est âgée d’au moins 10200 ans BP, n’atteignant pas 1 mètre. On arrive à la même conclusion à partir des résultats apportés par le 10Be, qui montrent l'occurrence d'un taux très faible de dénudation dans la chaîne de l' Espinhaço (Barretos et al., 2013).

Les datations radiocarbone, combinées aux analyses des isotopes du carbone et des phytholithes, suggèrent qu'il n'y a pas eu de modifications très significatives de la pluviosité de la région, suffisantes en tout cas pour répondre des changements dans le modèle de dépôt du matériau ou même dans les types d'altération ou des processus d'érosion pendant la durée de la formation du matériau dans les rampes colluviales. Il apparait néanmoins que la température vers la fin du Pléistocène/début de l'Holocène aurait été plus fraiche, et responsable de la décomposition complète de la matière organique. Ces observations confortent d'autres résultats (Augustin, 1995; Barros et al., 2011) indiquant que la relative stabilité tectonique et de l'environnement a favorisé l’occurrence de processus pédogénétiques favorisant le développement des néosols quartzo-sableux.

On peut conclure que les analyses combinées des phytolithes et des isotopes du carbone indiquent que le matériau rencontré dans les zones plates de la Serra do Engenho, dans la chaine de montagnes de l'Espinhaço, est en fait le résultat de l'érosion, du transport et du dépôt de sédiments en provenance des affleurements de quartzite proches désagrégés par l'altération physique et chimique. Cependant, les résultats n'excluent pas la possibilité que le développement de la rampe elle-même ait été, dans le passé, associée à la dissolution chimique de la silice (Willems et al., 2004; Fabri et al., 2013).

Cette recherche est l'une des premières tentatives d'utilisation des phytolithes, combinés à l'étude des isotopes du carbone, à des fins d'analyse géomorphologique et, bien qu'il y ait encore des questions non résolues, les résultats s'avèrent très intéressants et souligne le potentiel de ce type d'approche pour les études géomorphologiques. Ces analyses s'avèrent encourageantes pour la compréhension de l'évolution du paysage et des changements environnementaux, surtout si elles sont associées à d'autres indicateurs (analyse multiproxy).

Haut de page

Table des illustrations

Titre Fig. 1 – Location of the study area. It occurs in the Espinhaço massif in southeast Brazil, characterized by the dominance of quartzite from the Espinhaço Supergroup.Fig. 1Localisation de la zone d'étude. Massif d'Espinhaço, dans le sud-est du Brésil, caractérisé par la prédominance de quartzite du Supergroupe Espinhaço.
URL http://geomorphologie.revues.org/docannexe/image/10794/img-1.png
Fichier image/png, 2,1M
Titre Fig. 2 – Location of P1 and P2 on the low inclined surface near the quartzite outcrops of the Serra do Engenho syncline, in the Espinhaço massif. Fig. 2L'emplacement de P1 et P2 sur la surface basse inclinée près des affleurements de quartzite du synclinal de la “Serra do Engenho”, dans le massif Espinhaço.
Légende The inclination is gentle and ends at a valley wall. Grassland is predominant and riparian cover follows a creek before it cuts the portion of the ramp near the main scarp of the syncline. L'inclination est douce et se termine sur un versant de vallée. La prairie domine et une végétation riveraine borde un ruisseau avant qu'il ne coupe la partie de la rampe située près de l'escarpement principal du synclinal.
URL http://geomorphologie.revues.org/docannexe/image/10794/img-2.png
Fichier image/png, 972k
Titre Fig. 3 – Soil profiles from P1 (A) and P2 (B). Fig. 3Les profils de sol étudiés : P1 (A) et P2 (B).
Légende Both present a sandy organic superficial horizon with signs of laterization, and sequences of angular quartz pebbles below. In P2 the stony layer follows a sandy horizon rich in OM. In both profiles the lower horizons are formed by weathered quartzite with different levels of laterization. Les deux profils présentent un horizon superficiel organique sablonneux avec des signes de latérisation et des séquences de galets de quartz angulaires au-dessous. Sur le profil P2, la couche de pierres suit un horizon sablonneux riche en matière organique. Sur les deux profils, les horizons inférieurs sont formés de quartzite altérée avec différents niveaux de latérisation.
URL http://geomorphologie.revues.org/docannexe/image/10794/img-3.png
Fichier image/png, 1,4M
Titre Tab. 2 – Phytolith, carbon and isotopic results. Tab. 2Résultats des analyses de phytolithes et isotopiques pour les profils étudiés.
URL http://geomorphologie.revues.org/docannexe/image/10794/img-4.png
Fichier image/png, 82k
Titre Fig. 4 – Phytolith and isotopic analyses of Profile 1. Fig. 4Analyses phytolithiques et isotopiques du profil 1.
URL http://geomorphologie.revues.org/docannexe/image/10794/img-5.png
Fichier image/png, 340k
Titre Fig. 5 – Phytolith and isotopic analyses of Profile 2. Fig. 5Analyses phytolithiques et isotopiques du profil 2.
URL http://geomorphologie.revues.org/docannexe/image/10794/img-6.png
Fichier image/png, 345k
Titre Fig. 6 – Photos of main types of observed phytoliths in P1 and P2 horizons. Fig. 6 – Photos des principaux types de phytolithes observés dans les échantillons des profils 1 et 2.
Légende a: bulliform parallelepiped; b: bulliform cuneiform; c: elongate; d: polylobate; e: bilobate; f: cross; g, h: rondel; i: trapeziform; j, k, l: acicular; m: globular granulate; n, o: globular echinate. a : bulliforme parallélépipédique ; b : bulliforme cunéiforme ; c : allongé ; d : polylobé ; e : bilobé ; f : en croix ; g, h : rondelle ; i : forme en trapèze ; j, k, l : aciculaire ; m : sphérique granuleux ; n, o : échinulé sphérique.
URL http://geomorphologie.revues.org/docannexe/image/10794/img-7.png
Fichier image/png, 2,1M
Haut de page

Pour citer cet article

Référence papier

Cristina Helena Ribeiro Rocha Augustin, Heloisa Helena Gomes Coe, Karina Ferreira Chueng et Jenifer Garcia Gomes, « Analysis of geomorphic dynamics in ancient quartzite landscape using phytolith and carbon isotopes, Espinhaço Mountain Range, Minas Gerais, Brazil », Géomorphologie : relief, processus, environnement, vol. 20 - n° 4 | 2014, 355-376.

Référence électronique

Cristina Helena Ribeiro Rocha Augustin, Heloisa Helena Gomes Coe, Karina Ferreira Chueng et Jenifer Garcia Gomes, « Analysis of geomorphic dynamics in ancient quartzite landscape using phytolith and carbon isotopes, Espinhaço Mountain Range, Minas Gerais, Brazil », Géomorphologie : relief, processus, environnement [En ligne], vol. 20 - n° 4 | 2014, mis en ligne le 01 janvier 2016, consulté le 26 mars 2017. URL : http://geomorphologie.revues.org/10794 ; DOI : 10.4000/geomorphologie.10794

Haut de page

Auteurs

Cristina Helena Ribeiro Rocha Augustin

Instituto de Geociências – Universidade Federal de Minas Gerais – Av. Antônio Carlos, 6.627 – CEP: 31270-901 – Belo Horizonte, MG – Brazil (chaugustin@hotmail.com).

Heloisa Helena Gomes Coe

Departamento de Geografia – Universidade do Estado do Rio de Janeiro – R. Dr. Francisco Portela, 1470 – CEP: 24435-005 – São Gonçalo, RJ – Brazil (heloisacoe@yahoo.com).

Karina Ferreira Chueng

Departamento de Geologia – Universidade Federal Fluminense – Av. Gen. Milton Tavares de Souza s/nº – CEP: 24210-346 – Niterói, RJ – Brazil (karinachueng@yahoo.com.br).

Jenifer Garcia Gomes

Departamento de Geologia – Universidade Federal Fluminense – Av. Gen. Milton Tavares de Souza s/nº – CEP: 24210-346 – Niterói, RJ – Brazil (jenifergg16@hotmail.com).

Haut de page

Droits d’auteur

© Groupe français de géomorphologie

Haut de page
  • Logo CNRS - Institut des sciences humaines et sociales
  • Logo Groupe français de géomorphologie
  • Revues.org