Sedimentary fluvial processes at the margin of an eolian sand sea: the example of the Botucatu Formation, Paraná Basin

Abstract

The Upper Jurassic to Lower Cretaceous Botucatu Formation of the Paraná Basin consists primarily of trough cross-bedded eolian sandstone deposited in an intracontinental desert. In southeastern Brazil, an unusual occurrence of conglomerate and coarse-grained sandstone unconformably overlies the Neoproterozoic basement and is overlapped by volcanic rocks of the Serra Geral Formation. This represents one of the few documented occurrences of fluvial facies described for the Botucatu Formation in the northern part of the basin. To determine its depositional system and source area, we conducted facies and paleocurrent analysis, detrital grain modal counting, macroscopic provenance, and conventional heavy mineral analysis. The integrated analysis suggests that the studied succession was deposited by an ephemeral fluvial system at the margins of the extensive dune fields of the Botucatu paleodesert. Paleocurrent data and detrital composition indicate sediment dispersal toward the SE, with a possible source area associated with a recycled orogen to the NW (Brasília Belt). Although ephemeral rivers acted as important sediment suppliers for this continental-scale desert, recycling of eolian sediments by fluvial systems is not ruled out. Finally, using a multi-proxy approach, we present a paleogeographic scenario for the northern margin of the Botucatu paleodesert during the last period of Gondwana, prior to its initial rifting.

KEYWORDS:
Paleodesert; fluvial system; sediment provenance; Gondwanan rifting; Botucatu Formation

1 INTRODUCTION

The Upper Permian and Mesozoic periods were characterized by significant climatic changes driven by complete paleogeographic rearrangement of the Gondwana supercontinent (^{Parrish, 1993}; ^{Scherer & Goldberg, 2007}; ^{Scherer et al., 2023}). During the Cretaceous, the fragmentation of continental blocks combined with rising atmospheric carbon dioxide levels due to tectonic and magmatic activity intensified the greenhouse effect (^{Veevers, 1990}). This led to a significant temperature increase, reaching the highest levels of the Mesozoic era (^{Fluteau et al., 2007}; ^{Scherer & Goldberg, 2007}; ^{Scotese et al., 2021}).

In the late Jurassic, the presence of a pre-Andean belt at the SW Gondwana margin acted as a topographic barrier, restricting winds from transporting humidity into the continent’s interior, resulting in an extensive arid zone in the inner continental areas (^{Fernandes et al., 2014}). These conditions, combined with the continentalization and confinement of basins within the newly separated continental masses, provided favorable conditions for the development of large deserts such as the Botucatu paleoerg (^{Scherer & Goldberg, 2007}). Despite the rarity of fluvial deposits associated with desert environments, eolian systems may be laterally associated with ephemeral or perennial fluvial streams developed in alluvial fans and braided river systems along the dune field margins (^{Manes et al., 2021}; ^{Miyazaki & Basilici, 2015}). Punctual occurrences of fluvial successions are documented in the Rio Grande do Sul, Paraná, and eastern and northeastern regions of São Paulo and west of Minas Gerais states in Brazil (see ^{Soares, 1975}). These successions consist of medium- to coarse-grained sandstone and conglomerate, interpreted as deposited by torrential and ephemeral flows (^{Almeida & Melo, 1981}; ^{Bigarella & Salamuni, 1961}; ^{Scherer & Goldberg, 2007}; ^{Soares, 1975}). Between Araguari and Uberlândia, in the western Minas Gerais state, breccia, conglomerate, conglomeratic sandstone, and siltstone covered by the basalts of the Serra Geral Formation are attributed to the Botucatu Formation (e.g., ^{Moraes & Seer, 2018}).

The fluvial occurrences in the Botucatu Formation remain poorly studied, with limited sedimentological data. Detailed information regarding depositional processes, flow patterns, and the potential source areas of these fluvial systems is still lacking. This work aims to provide a detailed sedimentological description, including the determination of sediment dispersal patterns and analysis of the heavy mineral assemblages associated with sediment input in the desert margin. Considering that fluvial deposits in the Botucatu Formation are extremely rare, this study aims to offer insights into the interplay between sediment input and recycling processes in sedimentary basins. Additionally, by analyzing facies associations and variations in heavy mineral indices, this study seeks to understand how these sandstone and conglomerate deposits formed along the margins of the Botucatu paleodesert differ from the eolian sandstone found within the desert interior. To address these questions, this contribution employs a multi-proxy approach, comprising facies and paleocurrent analysis, diagenetic history, and provenance study of these uncommon fluvial deposits. By integrating newly acquired data, we propose a depositional model for these occurrences, as well as a paleogeographic reconstruction for the fluvial systems of the last Gondwanan deserts.

2 GEOLOGICAL SETTING

2.1 The Cretaceous successions of the northern Paraná Basin

The Paraná Basin is an intracontinental basin that extends across different geotectonic domains (^{Scherer et al., 2023}), including Archean to Paleoproterozoic cratonic terrains and Neoproterozoic Brasiliano/Pan-African mobile belts associated with the Gondwana assembly (^{Holz et al., 2006}; ^{Scherer et al., 2023}; ^{Zalán et al., 1990}; Fig. 1). The basin is divided into six supersequences separated by regional unconformities (^{Milani et al., 2007}; ^{Zalán et al., 1990}). The oldest three supersequences, including the Rio Ivaí (Ordovician-Silurian), Paraná (Devonian), and Gondwana I (Carboniferous-Triassic), record complete transgressive-regressive cycles related to oscillations of the Panthalassa relative sea level. The youngest three supersequences, Gondwana II (Triassic), Gondwana III (Jurassic-Cretaceous), and Bauru (Cretaceous), comprise successions that were exclusively deposited in continental settings.

Figure 1
Paleogeography of Gondwana.

The basement of the northern part of the Paraná Basin is associated with Archean-Paleoproterozoic cratonic terrains, as well as the Brasiliano/Pan-African orogenic belts (^{Bahlburg et al., 2009}; ^{Bertolini et al., 2021b}; ^{Zalán et al., 1990}), consisting of metasedimentary successions, magmatic arcs, granitoids, and deformed crustal metamorphic rocks, resulting in a quite broad provenance signature (^{Bertolini et al., 2021b}; ^{Campanha et al., 2019}; ^{McGee et al., 2015}; ^{McGee et al., 2018}; ^{Piuzana et al., 2003}). The basin is bordered on the northeastern margin by the Ribeira and Brasília belts, on the northwestern margin by the Paraguay belt and the Amazonian and Rio Apa cratons, and to the southwest by the Rio de la Plata craton (^{Bahlburg et al., 2009}; ^{Bertolini et al., 2021b}; Fig. 1). Due to the irregular distribution of the Paraná Basin units, which occur in response to different depocenters and the NE-SW and NW-SE fault systems that significantly affect the basin, the sedimentary record is spatially heterogeneous (^{Scherer et al., 2023}). Thus, the stratigraphic stacking pattern does not follow the typical model observed in intracratonic basins, leading to conflicting and/or incomplete stratigraphic sections (^{Scherer et al., 2023}). Therefore, recent studies propose that the definition of stacking should be determined through multiple stratigraphic transects (e.g., ^{Scherer et al., 2023}; Fig. 2).

Figure 2
(A) Stratigraphic transections used in the construction of the charts. (B) E-W stratigraphic chart of the southern Paraná Basin. The contact between the Botucatu Formation and the Guará and Pirambóia formations is marked by a hiatus. (C) NE-SW stratigraphic chart. Note the contact between the Botucatu, Caturrita, and Santa Maria formations and Pedreira Sandstone. (D) E-W stratigraphic chart in the northern part of the Paraná Basin.

In the northern part of the Paraná Basin, the Cretaceous successions consist of sedimentary and volcanic rocks of the Botucatu and Serra Geral formations, respectively (^{Milani et al., 2007}). In the Triângulo Mineiro and Alto Paranaíba regions, the deposits of the Paraná Basin overlie the Neoproterozoic metasedimentary and metaigneous basement of the Brasília Belt (^{Moraes & Seer, 2018}). During the early Cretaceous to early Paleozoic, the Brasília Belt acted as an important orographic barrier, forming a NW/SE mountain range that acted as the northeastern boundary of the Paraná Basin (^{Moraes & Seer, 2018}). This mountain range is considered the main source of sediments to the Botucatu Formation and served as an obstacle to the advance of the dunes and lavas of the Botucatu and Serra Geral formations, respectively (^{Campos & Dardenne, 1997}).

2.2 The Botucatu Formation

The eolian Botucatu Formation, representing the lower to intermediate part of the Gondwana III Supersequence, Serra Geral Formatio underlies the basaltic lavas of the Lower Cretaceousn (^{Milani et al., 2007}). This unit covers an extensive area of about 1,300,000 km² (Fig. 3A), outcropping across the states of Rio Grande do Sul, Santa Catarina, Paraná, São Paulo, Minas Gerais, Mato Grosso do Sul, and Goiás (^{Assine et al., 2004}; ^{Scherer & Lavina, 2006}). Correlated units are also described in western Uruguay (Tacuarembó and Buena Vista formations; ^{Assine et al., 2004}), eastern Paraguay (Misiones Formation; ^{Assine et al., 2004}), northeastern Argentina (Tacuarembó Formation; ^{Assine et al., 2004}), and correlated deposits in southern Africa (Twyfelfontein Formation in Namibia; ^{Grove, 2017}; ^{Scherer & Goldberg, 2007}; ^{Stanistreet & Stollhofen, 1999}). They consist mainly of large-scale (meters to tens of meters thick) planar to trough cross-bedded eolian quartz sandstone with high textural and mineralogical maturity (e.g., ^{Almeida, 1953}; ^{Assine et al., 2004}; ^{Bertolini et al., 2021a}; ^{Milani et al., 2007}; ^{Scherer & Goldberg, 2007}; ^{Scherer & Lavina, 2006}; ^{Scherer et al., 2023}).

Figure 3
(A) Location of the Paraná Basin in the South American continent highlighting the outcrop area of the Botucatu Formation (modified after ^{Scherer, 2000}, and ^{Bertolini et al., 2020}). (B) Geographic location of the western part of Minas Gerais State and the studied outcrop at the northeastern limit of the Paraná Basin. (C) Location of the studied outcrop in a railroad cut at the Centro Atlântica Railway – FCA, Araguari and Uberlândia cities (UTM coordinates: 790638E/7924922S). The image was obtained from Google Satellite.

The textural, mineralogical, and packing features of these sandstones are interpreted as deposited in an extensive desert or a dry eolian system (^{Almeida, 1953}; ^{Scherer & Goldberg, 2007}). Based on geometrical features of dune stratifications, it is possible to infer that the Botucatu paleoerg was formed by barchanoid and star dunes as well as complex linear draas (^{Scherer & Goldberg, 2007}; ^{Scherer et al., 2023}). In the lower succession, fluvial deposits with reworked ventifacts are locally described beneath the eolian dune deposits (^{Almeida & Melo, 1981}). Sporadic occurrences of fluvial conglomerate and conglomeratic sandstone at the base of the Botucatu Formation are very rare and are mostly found in topographic depressions with erosive surfaces (^{Milani et al., 1998}).

In addition to eolian deposits, sets and cosets of cross-bedded fine- to coarse-grained sandstone comprise the primary components of the Botucatu Formation (^{Scherer 2000}, ²⁰⁰²). These facies associations described in Rio Grande do Sul state also include large-scale trough cross-bedded sandstone, interpreted as residual deposits from migrating three-dimensional (3D) crescentic bedforms (^{Scherer, 2000}). Architectural elements comprising conglomeratic sandstone and horizontally stratified, fine- to coarse-grained sandstone are also interpreted as laterally unconfined stream flow deposits (^{Scherer, 2002}). Basal conglomerate deposits identified in the São Paulo and Paraná states were similarly interpreted as products of episodic flash floods along the margins of the Botucatu paleodesert (^{Soares, 1975}). It is noteworthy that part of these deposits were later redefined as belonging to the Guará Formation (^{Reis et al., 2019}). Therefore, the occurrence of fluvial deposits surely associated with the Botucatu Formation is rare and restricted.

In the northern exposures of the Botucatu Formation near Araguari town, Minas Gerais State, punctual occurrences of fluvial deposits are already known. These fluvial facies are characterized by basal conglomerate deposited within topographic depressions of the metamorphic basement, indicating proximal source areas and infilling of the basement lowlands with poorly sorted fluvial sediments (^{Milani et al., 1998}; ^{Moraes & Seer, 2018}). Alternatively, these rare deposits are interpreted as products of episodic flows within interconnected ephemeral fluvial systems related to alluvial fans and fluvio-lacustrine settings (^{Seer & Moraes, 2017}; ^{Famelli et al., 2021a}; ^2021b). The presence of silicified conifer logs in the same region suggests the occurrence of wetter periods during the arid/desertic climate that prevailed during the deposition of the eolian deposits of the Botucatu Formation (^{Pires et al., 2011}; ^{Suguio & Coimbra, 1972}). Ichnological studies of the Botucatu Formation, based on trace fossils, suggest the occurrence of episodic wet events (^{Manes et al., 2021}; ^{Peixoto et al., 2025}).

Given the extensive geographic distribution of the Botucatu Formation across diverse geological terrains, its sedimentary provenance signature is typically linked to multiple potential sediment source areas and transport pathways (^{Bertolini et al., 2020}). The sand provenance of the Botucatu paleoerg is predominantly interpreted as products of intense recycling from the underlying strata of the Paraná Basin (^{Bertolini et al., 2020}, ^2021b). In the southern Paraná Basin, it exhibits a west-to-east provenance shift (^{Scherer et al., 2023}), reflecting lateral variations in the older units beneath it (^{Scherer et al., 2023}). In Uruguay and western Rio Grande do Sul state, the Botucatu Formation unconformably overlies the Guará Formation (^{Scherer et al., 2023}; Figs. 2A and 2B). In the central region of Rio Grande do Sul, it rests unconformably on sandstone and mudstone of the Triassic Caturrita and Santa Maria formations (^{Almeida, 1953}; ^{Assine et al., 2004}; ^{Scherer et al., 2023}; Fig. 2C), while it overlaps the Rio do Rasto Formation and Pedreira Sandstone to the east (^{Scherer et al., 2023}; Fig. 2C). In the São Paulo and Mato Grosso do Sul states, the basal contact of the Botucatu Formation is predominantly with the Pirambóia Formation (^{Scherer et al., 2023}; Fig. 2D). The contact between the Botucatu Formation and the overlying Serra Geral Formation (Fig. 2D) is concordant, marked by interbedded sandstone within the basal lava flows (^{Assine et al., 2004}).

3 MATERIALS AND METHODS

The studied outcrop (UTM:790638 mE, 7924922 mS) comprises a railway cut located between the town of Araguari and the city of Uberlândia in the Triângulo Mineiro—Alto Paranaíba Mesoregion (Fig. 3B), western Minas Gerais State, Brazil. The primary access to the site is via the Viveiro-Linha de Ferro unpaved road, connected to the BR-050 highway between Araguari and Uberlândia (Fig. 3C).

Sedimentary facies were characterized based on physical and mineralogical grain properties, sedimentary structures, bed geometry, and contacts, following the protocol of ^{Miall (1996)}, and were represented in a 1:30-scale columnar section. Siliciclastic lithotypes were classified macroscopically using Folk’s nomenclature (1968). Depositional environment was interpreted through lateral and vertical facies association and comparisons with analogous modern and ancient fluvial systems (^{Walker & Middleton, 1977}).

Paleocurrent measurements were measured in three different stations, with ≥ 25 measurements to ensure statistical reliability in determining the average sediment transport vector (^{Miall, 1974}; ^{Selley, 1982}). Data obtained from cross-bedded sandstone and imbricated clasts was plotted in rose diagrams using Stereonet 11 free software.

Clast counting was performed by randomly analyzing 300 clasts in conglomerate and conglomeratic sandstone facies, based on their lithology and size (major and minor axes). Only clasts > 1 cm were evaluated to encompass the broadest grain size range and to minimize bias (^{Haughton et al., 1991}; ^{Howard, 1993}). Modal counting of detrital framework grains was conducted via petrographic analysis of three thin sections from coarse-grained sandstone, following the Gazzi-Dickinson method (e.g., ^{Ingersoll et al., 1984}; ^{Zuffa, 1980}). This analysis was carried out at the Center for Applied Natural Sciences (UNESPetro) of the Institute of Geosciences and Exact Sciences (UNESP, Rio Claro, São Paulo) using an Axio Imager.A2 Zeiss petrographic microscope. For each thin section, 300 points were counted along a transverse at a pre-defined spacing of 0.5 mm to identify primary constituents (^{Dickinson, 1970}; ^{Folk, 1968}; ^{Gazzi, 1966}).

During fieldwork, ~5 kg samples were collected from trough cross-bedded conglomeratic sandstone facies (Csc; Bot-01), planar cross-bedded sandy conglomerate (Scc; Bot-02), and massive sandstone (Sm; Bot-03) for heavy mineral analysis. The sample was prepared in six steps:

• crushing and grinding with a jaw crusher;

• sieving;

• panning;

• magnetic mineral separation using a hand-held magnet;

• heavy mineral separation via dense liquid (bromoform);

• mounting petrographic slides with concentrated heavy minerals.

Heavy minerals were identified using the same Axio Imager.A2 Zeiss petrographic microscope, with 300 translucent grains counted from the total mineralogical assemblage (^{Mange & Maurer, 2012}; ^{Remus et al., 2008}). This procedure was conducted at the Center for Applied Natural Sciences (UNESPetro) of the Institute of Geosciences and Exact Sciences (UNESP, Rio Claro, São Paulo). Data were compiled and processed in Excel spreadsheets, and the following ratios were calculated to characterize heavy mineral indexes:

• GZi (100 × Garnet/Garnet + Zircon);

• RZi (100 × Rutile/Rutile + Zircon);

• MZi (100 × Monazite/Monazite + Zircon);

• ZTR (100 × Zircon + Tourmaline + Rutile/Total translucent grains).

Compositional data from the Gazzi-Dickinson modal counting method (^{Dickinson, 1970}; ^{Gazzi, 1966}) and conventional heavy mineral analysis (^{Hubert, 1962}; ^{Morton & Hallsworth, 1994}; ^{Remus et al., 2008}) were organized in Excel spreadsheets. Statistical analysis was performed using the “provenance” package in R (^{Vermeesch et al., 2016}) with the “chi-square distance” configuration for principal component analysis (PCA). To enable multidimensional statistical clustering and conduct a comparative analysis of the sedimentary provenance, the data obtained from the fluvial sandstones of the Botucatu Formation were incorporated into the PCA along with petrographic and heavy minerals data obtained from eolian sandstones of the Botucatu Formation (data from ^{Bertolini et al., 2020}, ^2021b for the southern and northern parts of the Paraná Basin, respectively).

4 RESULTS

4.1 Sedimentary facies analysis

The studied section comprises a 60-meter-long and 4-meter-high NE-SW cut in the Centro-Atlântica Railway (Fig. 4A). The rocks comprise fine- to coarse-grained sandstone and conglomerate of the Botucatu Formation. The lower boundary is characterized by an unconformity with the metamorphic rocks of the Neoproterozoic basement (Fig. 4B), and the upper boundary is marked by sandstone conformably covered by a thick pile of basaltic lava from the Serra Geral Formation (Fig. 4C).

Figure 4
(A) Outcrop of the fluvial deposits of the Botucatu Formation in a railroad cut between Araguari town and Uberlândia city (UTM coordinates: 790638E/7924922S). (B) Detail of unconformity (red dashed line) between coarse-grained sandstone of the Botucatu Formation and the augen gneiss of the Neoproterozoic Maratá Complex. (C) Outcrop view showing the concordant contact between the fluvial deposits of the Botucatu Formation and the volcanic rocks of the Serra Geral Formation (red dashed line).

Due to the small outcropping area, only four sedimentary facies were identified (Figs. 5 and 6):

• massive sandstone (Sm);

• trough cross-bedded conglomeratic sandstone (Csc);

• horizontally stratified conglomeratic sandstone (Csp);

• planar cross-bedded sandy conglomerate (Scc).

Figure 5
Lithofacies of the Botucatu Formation in the studied section. All images are oriented to NE-SW. (A) Fluvial facies from the studied section. From base to top of planar cross-bedded sandy conglomerate (Scc), trough cross-bedded conglomeratic sandstone (Csc), and massive sandstone (Sm) facies. Some facies (Scc) occur as discontinuous lenses and channels. (B) Detail of erosive contact between planar cross-bedded sandy conglomerate (Scc) and trough cross-bedded conglomeratic sandstone (Csc) facies. Yellow lines indicate cross-stratification. (C) Contact between planar cross-bedded sandy conglomerate (Csc) and horizontally-stratified conglomeratic sandstone (Csp) at the upper part of the studied section. Yellow lines indicate cross-stratification. The scale is positioned at the contact between sandstone of the Botucatu Formation and weathered basalts of the Serra Geral Formation. (D) Upper part of the succession showing, from bottom to top, decimeter-scale beds of trough cross-bedded conglomeratic sandstone (Scc) and sandy conglomerate with planar cross-stratification (Csc). At the top of the section, detail for the horizontally-stratified conglomeratic sandstone (Csp). Yellow lines indicate cross-stratification. (E) Detail of decimeter-thick bed of sandy conglomerate with incipient planar cross-stratification (Scc). (F) Imbricated lamellar lithoclasts of schists (white arrows) in planar cross-bedded sandy conglomerate (Csc). The geologic hammer in B and F is ~33 cm in length. The scale in C and E is ~8 cm long. The scale in D is ~17 cm long.

Figure 6
Photomicrographs of fluvial facies of Botucatu Formation. (A) Framework of the conglomeratic sandstone (sample Bot-01; crossed polarizers) showing the predominance of monocrystalline quartz grains. Some detrital quartz grains exhibit discontinuous syntaxial overgrowth of silica (green arrows). Note the occurrence of secondary porosity of the intragranular type (pink arrows). Note the presence of plagioclase grain (white arrow). (B) Same as A but with natural light/uncrossed polarizers. Grain-coating iron oxide surrounds the grains and occupies the porosity. Note the presence of pseudomatrix in the lower-left corner. Note the occurrence of secondary moldic porosity (brown arrows). (C) Detail of a muscovite-schist lithoclast (sample Bot-02; crossed polarizers; red arrows). (D) Same as A but with natural light/uncrossed polarizers. Note the significant presence of iron oxide coatings around the quartz grains and the primary porosity. (E) Detail of framework (sample Bot-02), showing grains with low to intermediate roundness and sphericity. Note the elongated schist lithoclast in the center of the image (yellow arrow). (F) Same as E but with natural light/uncrossed polarizers.

Descriptions and the respective depositional processes of each facies are summarized and compiled in Table 1. The lower part of the columnar section consists of gneiss, which is overlaid by ~4 m of sandstone and conglomerate, forming a fining upward succession. As observed in the columnar section (Fig. 5), the sandstone and conglomerate facies are mainly disposed in the small-sized (m-scale) channels with erosive base (Figs. 4C and 5A). Laterally, the individual channelized beds are amalgamated, configuring a slightly tabular appearance to the deposit (Fig. 4). The conglomeratic facies are in contact with basaltic rocks of the Serra Geral Formation (Fig. 4C), which are weathered.

Table 1.
Sedimentary facies of the Botucatu Formation in the studied area.

Code	Facies	Description	Depositional process
Scc	Planar cross-bedded sandy conglomerate	Sandy conglomerate with planar cross-bedding disposed in decimeter- to meter-scale lenticular beds with erosive base. Imbricated (mean vector of 116º) schist clasts ranging from 0.3 to 18 cm in diameter occur concentrated in the foresets of the planar stratifications. Sparse quartz clasts have mostly subspherical and rarely prolate shape. Presence of grains with low sphericity varying between sub-angular and sub-rounded.	Migration of straight-crested transverse bedforms and 2D subaqueous dunes under unidirectional flow
Csp	Horizontally-stratified conglomeratic sandstone	Horizontally-stratified conglomeratic sandstone disposed in decimeter- to meter-scale lenticular beds with erosive base. Framework grains with low sphericity varying between sub-angular and rounded. Schist clasts are concentrated at the base of the stratifications, with diameters ranging from 0.3 to 4 cm. Quartz clasts present subspherical geometry. These facies are covered by the basaltic rocks of the Serra Geral Formation.	Deposition by planar bed in upper flow regime in the upper part of transverse or longitudinal fluvial bedforms
Csc	Trough cross-bedded conglomeratic sandstone	Conglomeratic sandstone with trough cross-bedding disposed in lenticular to tabular decimeter- to meter-scale beds with erosive base. Presence of imbricated (mean vector 127º) schist and quartz clasts with diameters ranging between 0.4 and 6 cm. As observed in the facies Csp and Scc, the clasts occur concentrated in the foreset of the stratifications. The sand grains show low sphericity, varying between sub-angular and sub-rounded.	Migration of subaqueous sinuous-crested (3D) or linguoid dunes
Sm	Massive sandstone	Well sorted, very fine-to-medium-grained massive sandstone with an erosive base and low sphericity grains varying between angular and sub-rounded.	Sediment gravity-flow deposits. Alternatively, the massive aspect may be related to liquefaction that has obliterated the original sedimentary structures.

4.1.1 Massive sandstone (Sm)

This facies consists of massive fine- to medium-grained sandstone with low sphericity and rounded to subrounded grains. Massive sandstone forms 40-cm-thick tabular beds without lithoclasts. It is interpreted as associated with sediment gravity-flow deposits (Table 1). However, its massive aspect may result from liquefaction, bioturbation, and/or absence of granulometry contrast, which could have obliterated the original sedimentary structure.

4.1.2 Trough cross-bedded conglomeratic sandstone (Csc)

This facies consists of trough cross-bedded, fine- to coarse-grained conglomeratic sandstone, disposed in lenticular to tabular decimeter- to meter-scale beds with erosive base. The grains are predominantly angular to sub-rounded with low sphericity. The foresets are tangential at the base and commonly present imbricated lithoclasts of schists and quartz vein, with diameters ranging from 0.4 to 6 cm. In general, the lithoclasts have a prolate shape and roundness ranging from poorly rounded to subrounded. It is interpreted as residual deposits left by subaqueous sinuous-crested 3D or subaqueous linguoid dune migration.

4.1.3 Horizontally stratified conglomeratic sandstone (Csp)

This facies consists of horizontally bedded fine- to coarse-grained conglomeratic sandstone, disposed in decimeter- to meter-scale lenticular to tabular beds. The quartz grains display low sphericity and are sub-angular to rounded. The poorly rounded to subrounded lithoclasts (0.3–4 cm in diameter) are concentrated in the horizontal bedding and are compositionally similar to those observed in the facies Csc. Interestingly, most schist lithoclasts present in the Csc and Csp are found in the dune foresets, while vein quartz lithoclasts typically occur dispersed within the sandy matrix. This facies is interpreted as under upper flow acting in the upper parts of transverse or longitudinal fluvial bedforms.

4.1.4 Planar cross-bedded sandy conglomerate (Scc)

This facies consists of fine- to coarse-grained sandy conglomerate exhibiting planar cross-bedding, organized into decimeter- to meter-scale lenticular and tabular beds with erosive base. The planar cross-bedded sandy conglomerate (Scc) is associated with packages of trough cross-bedded conglomeratic sandstones (Csc), forming discontinuous lenticular layers with centimeter-scale thickness at the base of the section. In other occurrences, this conglomerate facies occurs as tabular beds with an erosive base (Fig. 5). The sub-angular to sub-rounded quartz grains display low sphericity. As in the Csp facies, the schist lithoclasts (0.3–18 cm in diameter) are concentrated in the planar stratifications, while quartz veins are floating in the sandy matrix. Quartz lithoclasts are predominantly subspherical in shape, while schist lithoclasts are mainly prolate. This facies is interpreted to have formed through the migration of straight-crested transverse bedforms and two-dimensional (2D) subaqueous dunes under unidirectional flow.

4.2 Paleocurrents and macroscopic provenance analysis

A total of 108 paleocurrent measurements were collected from imbricated clasts (n = 90) and cross-stratifications in trough cross-bedded conglomeratic sandstone (Csc; n = 18), which showed a consistent sediment transport direction toward the SE (Fig. 5). To achieve greater statistical coverage, data from both imbricated clasts and cross-stratifications were combined and plotted in rose diagrams.

Macroscopic analysis was conducted on planar cross-bedded sandy conglomerate (Scc; Fig. 5E) with a total clast count of 300. The lithoclasts generally exhibit angular to sub-angular shapes, with both prolate and oblate forms. The analysis identified 242 lithoclasts as schist fragments of varying diameters (0.3–18 cm), while the remaining 58 clasts consisted of subangular to sub-spherical quartz vein fragments.

4.3 Microscopic petrography and Gazzi-Dickinson point count method

Petrographic analyses were conducted on three samples:

• Bot-01; Csc facies;

• Bot-02; Scc facies;

• Bot-03; Sm facies (Fig. 6).

The petrographic analyses reveal that the framework is predominantly composed of poorly sorted grains (Figs. 6 and 7). Essentially, the main constituents of the samples are similar, with notable variations in lithic fragment content. The primary grain components include monocrystalline quartz (as a monomineralic detrital grain; Figs. 6A and 6B), polycrystalline quartz (as a fragment of metamorphic rock), and lithic fragments of muscovite schist and quartz-muscovite schist (Figs. 6C and 6D). Additionally, there are rare occurrences of detrital feldspar and muscovite, both as monomineralic detrital grains and contained in fragments of metamorphic rocks. The samples (e.g., lower part in Bot-01, intermediate part in Bot-02, and upper part in Bot-03) exhibit a marked increase in monocrystalline quartz grain content toward the top of the columnar section, accompanied by a decrease in the total content of lithic fragments (Figs. 6C and 6D). Although iron oxide cementation is present in all samples, its occurrence tends to increase toward the top of the columnar section, particularly in Bot-03.

The textural evolution of the grains reveals notable differences among the samples collected from the base, intermediate, and top of the described succession. For instance, while grains in samples Bot-01 and Bot-02 exhibit low to intermediate roundness and sphericity (Figs. 6E and 6F), those in sample Bot-03 are predominantly well-rounded and highly spherical, with diameters significantly larger than those observed in the other samples.

The presence of diagenetic minerals is primarily represented by continuous and discontinuous syntaxial overgrowths of quartz (Figs. 6A and 6B) and discontinuous syntaxial overgrowths of feldspar (Figs. 7A and 7B). Clays occurring as coatings around the grains (Figs. 7C and 7D). Discontinuous syntaxial overgrowths of quartz and feldspar around detrital quartz and feldspar grains fill the primary porosity in parts not occupied by iron oxide cement. A significant presence of iron oxide cementation was also observed (Figs. 7E and 7F), appearing as large amorphous masses between the grains and significantly reducing the primary porosity of the samples.

Figure 7
Photomicrographs of fluvial facies of Botucatu Formation. (A) Syntaxial overgrowth of feldspar (yellow arrows) in sample Bot-01. Note the occurrence of secondary moldic porosity (purple arrows). The presence of lithoclasts and deformed micas (white arrows) is a common feature in the samples analyzed. (B) Same as A but with natural light/uncrossed polarizers. (C) Note the occurrence of well-rounded and highly spherical grains. Mechanically infiltrated clays occurring as coatings around the grains (red arrows) in sample Bot-03. Note the occurrence of concavo-convex contact between grains (black arrows). (D) Same as C but with natural light/uncrossed polarizers. (E) Cementation by large amorphous masses of iron oxide (green arrows) between quartz grains in sample Bot-03. (F) Same as E but with natural light/uncrossed polarizers.

Additionally, amorphous aggregates of oxidized material were observed. Secondary porosity of the moldic type is also present (Figs. 7A and 7B). A detailed analysis of the rock framework reveals that the grains commonly exhibit punctual contacts; however, some concavo-convex contacts were also noted (Figs. 7C and 7D), along with occurrences of deformed micas (Figs. 6C and 6D). In parts with a higher degree of iron oxide cementation, the grains are embedded within the cement (Figs. 7E and 7F). The less cemented portions tend to display point and concave-convex contacts, while the more cemented parts are characterized by floating grains (Fig. 7F).

The mineral assemblage identified is summarized in Table 2. According to Folk’s classification (1968), the samples are classified as sublitharenite (Fig. 8A). The data obtained using the Gazzi-Dickinson point count method were plotted on the Qt-F-L tectonic provenance diagram of ^{Dickinson (1985}; Fig. 8B).

Table 2.
Concentration and composition of primary and accessory detrital grains of sandstone and conglomerate of the Botucatu Formation in the studied area.

Constituent	Bot-01		Bot-02		Bot-03
	n	%	n	%	n	%
Detrital monocrystalline quartz presenting slightly undulating extinction (MG)	71	23.7	55	18.3	60	20
Detrital monocrystalline quartz presenting abrupt extinction (MG)	69	23	75	25	84	28
Detrital monocrystalline quartz presenting strong undulating extinction (MG)	63	21	65	21.7	72	24
Detrital polycrystalline quartz (MRF)	32	10.7	16	5.3	23	7.7
Iron oxide cement	19	6.3	10	3.3	16	5.3
Diagenetic silica—Intergranular, discontinuous grain overgrowth	10	3.3	-	-	5	1.7
Detrital microcline (MG)	7	2.3	12	4	9	3
Detrital muscovite (MG)	7	2.3	13	4.3	1	0.3
Schist fragment (MRF)	7	2.3	29	9.7	2	0.7
Moldic pore—Secondary interstitial porosity	7	2.3	-	-	3	1
Detrital garnet (MG)	3	1	-	-	-	-
Diagenetic Feldspar - Interparticle as a continuous grain covering	2	0.7	-	-	2	0.7
Detrital plagioclase (MG)	1	0.3	4	1.3	2	0.7
Detrital apatite—Weathered (MG)	1	0.3	-	-	-	-
Detrital clay—Intergranular	1	0.3	-	-	-	-
Detrital biotite (MG)	-	-	9	3	-	-
Intergranular pore—Primary interstitial porosity	-	-	7	2.3	-	-
Detrital garnet (MG)	-	-	5	1.7	4	1.3
Undifferentiated pore—Secondary porosity	-	-	-	-	13	4.3
Detrital zircon (MG)	-	-	-	-	2	0.7
Sandstone fragment (SRF)	-	-	-	-	2	0.7
Total	300	100	300	100	300	100

• MG: monomineralic grain; MRF: metamorphic rock fragment.

Figure 8
(A) Analyzed samples (Bot-01, Bot-02, and Bot-03) plotted on tectonic provenance diagram by ^{Dickinson (1985)}; (B) Same samples of A plotted in the terrigenous rock classification diagram (^{Folk, 1968}). (C) Pie-chart diagram of the heavy mineral assemblages obtained from samples Bot-01, Bot-02, and Bot-03. (D) Principal Component Analysis (PCA) plot showing the main composition and assemblage for fluvial (this work) and eolian deposits of the Botucatu Formation (data by ^{Bertolini et al., 2020}; ^2021b). (E) PCA plot showing the main heavy mineral assemblages of the fluvial deposits from the Botucatu Formation. Minerals acronyms: Actinolite (Actn); Andalusite (Andl); Anatase (Ants); Apatite (Aptt); Augite (Augt); Biotite (Biot); Clay (Clay); Corundum (Crnd); Diagenetic Feldspar (Dfel); Diagenetic Silica (Dsil); Epidote (Epdt); Fluorite (Fluo); Garnet (Grnt); Granite (Gran); Hornblende (Horn); Iron Cement (Ircm); Kaolinitic Cement (Kacm); Kyanite (Kyan); Lithics (Lith); Microcline (Mcrc); Monazite (Mnzt); Muscovite (Mscv); Opal (Opal); Plagioclase (Pgcl); Pore (Pore); Pyrolusite (Pyrl); Quartz (Qrtz); Quartzite (Qtzt); Rutile (Rutl); Sandstone (Snds); Sillimanite (Sill); Smectite Cement (Smcm); Spinel (Spin); Staurolite (Strl); Sulfides (Sulf); Titanite (Ttnt); Topaz (Topz); Tremolite (Trem); Turmaline (Trml); Xenotime (Xent); Zircon (Zrcn).

4.4 Heavy minerals analysis

The heavy mineral assemblage (Table 3) consists of zircon, tourmaline, rutile, staurolite, kyanite, sillimanite, garnet, and monazite, with epidote, andalusite, anatase, titanite, and hornblende being less abundant (Fig. 8C). The translucent heavy minerals are primarily composed of ultra-stable minerals such as zircon, rutile, and tourmaline. Zircon grains (~30-65% of the total) are generally rounded to well-rounded, exhibiting high to low sphericity, with diameters ranging from 0.05 to 0.2 mm. Rutile grains (~10-12% of the total) are rounded to sub-rounded, with diameters between 0.1 and 0.2 mm. Tourmaline grains (~17-29% of the total) are rounded to subrounded, have low sphericity, and range in diameter from 0.1 to 0.25 mm. The degree of roundness and sphericity of the heavy minerals, particularly the ultra-stable heavy minerals (ZTR; e.g., ^{Hubert, 1962}), is consistent with observations from the petrographic analysis, increasing toward the top of the described columnar section. Therefore, there is a clear increase in the roundness and sphericity of both heavy minerals and quartz grains from facies deposited by lower flow regime (Scc; Csc; Sm) to upper flow regime (Csp).

Table 3.
Table showing respective percentages of the heavy minerals’ assemblages obtained from samples Bot-01, Bot-02, and Bot-03.

Mineral	Samples
	Bot-01	Bot-02	Bot-03
Zircon	34.67%	30.00%	64.67%
Tourmaline	28.67%	25.00%	17.00%
Rutile	11.33%	11.00%	10.67%
Silimanite	8.00%	9.33%	1.00%
Garnet	6.00%	8.33%	2.67%
Staurolite	4.67%	1.00%	2.00%
Monazite	3.67%	4.00%	1.00%
Kyanite	2.00%	8.67%	0.33%
Hornblende	0.33%	0.33%	-
Anatase	0.33%	-	-
Titanite	0.33%	0.67%	0.33%
Andalusite	-	1.00%	0.33%
Epidote	-	0.67%	-
Total	100%	100%	100%

The heavy mineral indexes calculated from the analyzed samples, including GZi, RZi, MZi, and ZTR (^{Hubert, 1962}; ^{Remus et al., 2008}), are shown in Table 4. Metamorphic minerals are prevalent in all samples and include garnet, rutile, monazite, kyanite, sillimanite, and staurolite. The concentration of ultra-stable mineral grains such as zircon, tourmaline, and rutile increases significantly from the base to the top of the columnar section. In contrast, the predominance of metamorphic minerals, including kyanite, sillimanite, andalusite, and staurolite, decreases upward in the section (Fig. 8C).

Table 4.
Heavy mineral index values of sandstone and conglomerate of the Botucatu Formation (samples Bot-01, Bot-02, and Bot-03).

Sample	GZi (%)	RZi (%)	MZi (%)	ZTR (%)
Bot-03	3,96	14,16	1,52	92,33
Bot-02	21,74	26,83	11,76	66,00
Bot-01	14,75	24,64	9,57	74,67

The PCA (^{Vermeesch et al., 2016}) reveals notable disparities in the framework composition of the samples (Fig. 8D). Certain components, even in low concentrations, contribute to the differentiation of samples within the plot. For example, eolian samples PB-02, PB-03, and PB-04 are more closely associated with quartz, whereas fluvial samples Bot-01, Bot-02, and Bot-03 exhibit a distinct composition linked to lithic and diagenetic components. This variation suggests significant compositional differences between these sample groups, reflecting the influence of different sedimentary facies. This small disparity is marked by the occurrence of minerals such as biotite, apatite, and plagioclase, occurring in trace amounts. This disparity was also observed in the heavy mineral assemblage (Fig. 8E). It is observed that eolian samples PB-02, PB-03, and PB-04, for example, are more closely associated with ultra-stable minerals such as zircon, tourmaline, and rutile, whereas fluvial samples Bot-01, Bot-02, and Bot-03, contain minerals like epidote, anatase, and andalusite, which are exclusively present in these facies.

A comparison of fluvial and eolian samples in the PCA reveals a clear distinction in both primary mineral composition and heavy mineral assemblage. To ensure a robust comparative analysis, data from the northern (^{Bertolini et al., 2021b}) and southern (^{Bertolini et al., 2020}) sectors of the Paraná Basin were included. The results indicate that, regardless of the location within the basin, eolian samples consistently differ from fluvial samples. While the eolian samples cluster in one section of the plot, the fluvial samples occupy an opposing region (Fig. 8D). This compositional difference is primarily attributed to the absence of kaolinitic and smectite cement and plagioclase and by the predominance of ultra-stable minerals (zircon, garnet, and apatite) and lithic and diagenetic minerals in the fluvial samples. The PCA analysis of heavy mineral assemblages (Fig. 8E) further confirms a different trend, where metamorphic components are more prevalent in fluvial samples, while eolian samples exhibit a strong affinity for ultra-stable heavy minerals, such as zircon, tourmaline, and rutile. An exception to this trend is the fluvial sample Bot-03, which, according to the PCA results, shares compositional characteristics very similar to the eolian samples.

5 DISCUSSION

5.1 Depositional System

Studies in the bibliography indicate the occurrence of these fluvial deposits (^{Milani et al., 2007}; ^{Scherer et al., 2023}; ^{Soares, 1975}); however, they provide only simplistic descriptions or fail to address aspects of the depositional system responsible for the deposition of these facies. The data presented and discussed here offer a more detailed perspective, albeit punctual, on the behavior of these fluvial systems that operated at the margin of the Botucatu paleodesert. Furthermore, this case study enhances our understanding of how torrential flow processes supplied sediment to a desert-margin area, which consequently provided material for the eolian reworking of large Cretaceous paleoergs.

The coarse-grained and conglomeratic sandstone facies described here share several similarities, making their differentiation a difficult task. The joint analysis of facies associations and textural features observed in thin sections provided insights into depositional processes at outcrop and microscopic scales.

The conglomerate and coarse-grained sandstone facies of the Botucatu Formation are interpreted as deposits of a proximal fluvial system developed directly over a regional Neoproterozoic basement. The lateral and vertical relationship between cross-bedded sandstone, conglomeratic sandstone, and conglomerate facies suggests that these deposits resulted from the amalgamation of longitudinal sandy bars in a fluvial system influenced by pronounced seasonality and orographic factors (^{Fernandes et al., 2014}; ^{Fielding et al., 2018}; ^{Fielding et al., 2025}). Additionally, the sedimentary succession represented by the conglomeratic facies indicates a dynamic system associated with periods of increased humidity that progressively dried up. This pattern reflects ephemeral water discharges originating from topographic highs bordering the Paraná Basin (^{Almeida & Melo, 1981}; ^{Bigarella & Salamuni, 1961}; ^{Soares, 1975}). Textural characteristics, such as the rounding of clasts and the presence of poorly sorted frameworks, reinforce the interpretation of a depositional environment shaped by ephemeral fluvial reworking processes. The size of lithoclasts and the characteristics of the coarse-grained sandstone facies suggest that this fluvial system had significant transport competence, leading to the formation of conglomeratic facies under unidirectional flow. However, the absence of sandstone with structures formed under the upper flow regime (e.g., antidunes, cyclic steps, chute, and pool) suggests that these facies are more compatible with fluvial systems with low to moderate discharge variability (^{Fielding et al., 2018}; ^{Fielding et al., 2025}). ^{Fielding et al. (2018)} point out that facies variability is common in arid and semi-arid environments, which are driven by variation in precipitation. Locally, channel beds appear either amalgamated or filling substrate irregularities, suggesting that sandy and gravel bars, along with residual deposits from high- and low-competence flows, filled an ancient paleorelief with better-established fluvial channels. In this context, the conglomerate facies indicate the migration of gravel bars and thalweg deposits (i.e., basal channel-fill deposits; ^{Miall, 1977}; ¹⁹⁸⁵; ¹⁹⁹⁶; ²⁰¹⁰; ²⁰¹³; ²⁰¹⁴).

The integration of sedimentary facies with petrographic features suggests that, as described by ^{Fielding et al. (2018)}, the depositional system is characterized by irregular water flow discharges at the basin margins. This irregularity, evidenced by the coexistence of sedimentary structures from both lower and upper flow regimes, is also reflected in the textural characteristics of the grains observed in thin sections. This relationship supports the hypothesis that the upper flow regime, responsible for generating horizontally bedded facies at the top of the section, also contributed to the increased grain roundness and sphericity.

5.1.1 Diagenetic signatures of fluvial deposits under arid climate

Petrographic analysis reveals the presence of reliable indicators of early diagenesis (^{Tucker, 1991}), such as:

• syntaxial overgrowths of quartz and feldspar;

• oxidation and compaction of micas;

• iron oxide cementation;

• mechanical infiltration of clays.

The growth of authigenic feldspar results from the transport of fluids rich in Na⁺, K⁺, Al³⁺, and Si⁴⁺, generally associated with the dissolution of less stable grains, possible mica-rich lithoclasts (^{De Ros et al., 1994}).

The infiltration of clays likely occurred in association with aqueous flows carrying large amounts of fine-grained sediments (clay) as suspended load (^{Moraes & De Ros, 1990}; ^{Walker et al., 1978}). In response to fluctuations in the water table under dry conditions, clay-rich flows may have infiltrated the vadose zone, depositing clay as a coating around sand grains (^{Moraes & De Ros, 1990}; ^{Walker et al., 1978}).

As mentioned, intense iron oxide cementation is also a common early diagenetic feature (^{Tucker, 1991}). Therefore, diagenetic processes were chronologically interpreted within an evolutionary framework (Table 5), leading to the conclusion that these processes are predominantly associated with eodiagenesis. In the sandstone and conglomerate of the Botucatu Formation, this process can be explained as a product of mineral oxidation or dehydration of detrital hydroxides deposited alongside mechanically infiltrated clay (^{Walker et al., 1978}). The described eodiagenetic processes may have occurred in abandoned channels following river avulsion or during floods associated with ephemeral streams (^{De Ros & Scherer, 2013}). In other words, it is plausible that the fluvial sediments possibly underwent combined early diagenetic processes under desertic or arid climate (^{Kessler, 1978}). In this context, moderate clay infiltration, the dissolution of unstable clasts, and intense iron oxide cementation substantially altered the porosity and permeability of the rock.

Table 5.
Diagenetic stages identified in the Botucatu Formation rocks in the region between Araguari and Uberlândia.

Diagenetic Stages	Eodiagenesis	Mesodiagenesis	Telodiagenesis
Clay infiltration	xx
Iron oxide cementation	xxx
Mechanical compaction	x
Quartz and feldspar ovegrowth	x
Secondary porosity		x
Compaction			x

• Relative importance of diagenetic stages: x = weak; xx = medium; xxx = strong.

5.2 Sedimentary Provenance

The paleocurrent data indicate sediment dispersion from the NW and W toward the SE and E. Macroscopic provenance analysis using the gravel composition revealed a significant contribution of metamorphic lithotypes (mainly muscovite schists) and quartz. The shape analysis of the clasts showed a wide range of diameters (0.3–18 cm) and forms varying from angular to sub-angular prolates and oblates, indicating deposition close to the source area.

Petrographic modal analysis of the detrital composition indicates that all samples can be classified as sublitharenite (Fig. 8A), reinforcing their provenance from recycled orogenic terranes (Fig. 8B). The heavy mineral assemblage reveals the predominance of metamorphic minerals, including garnet, staurolite, sillimanite, kyanite, and andalusite, indicating an important contribution from a medium to high-grade metamorphic source. However, aluminum polymorphs (i.e., sillimanite, kyanite, and andalusite) and staurolite decrease in abundance toward the top of the succession. Conversely, the increasing abundance of zircon and tourmaline grains toward the top results in a significant increase in the ZTR index and, consequently, in mineralogical sediment maturity. Alternatively, this pattern may reflect a gradual replacement from metamorphic to granitic sources and/or an increase in sediment recycling within the basin. The hypothesis of recycling and gradual modification of the primary source areas is related to:

• the presence of angular to sub-angular grains at the base transitioning to sub-rounded to rounded grains toward the top;

• the increasing of ultra-stable minerals toward the top;

• the rise of the ZTR index, indicating increasing maturity toward the top.

The presence of rounded grains in the upper part of the succession suggests that the ephemeral fluvial system progressively interacted more with arid/desert areas, where sediments are subjected to constant reworking and transport to the more interior sectors of the basin.

A PCA analysis of the compositional data for fluvial and eolian sandstone of the Botucatu Formation highlights clear differences between these deposits. While eolian samples exhibit relatively uniform composition (Fig. 8D), with the exception of samples PB-19, PB-20, and PB-29A, which show an anomalous concentration of opal grains, the fluvial samples contain a significantly higher volume of lithic fragments and trace minerals such as muscovite, garnet, and biotite. The heavy mineral assemblages PCA (Fig. 8E) indicates a strong metamorphic influence in the fluvial samples, evidenced by the significant presence of metamorphic minerals, along with a notable influence of recycling, as indicated by the presence of ultra-stable minerals. In contrast, the eolian samples display a much higher degree of recycling, as reflected in the greater concentration of minerals such as zircon, tourmaline, and rutile. An exception to this trend is fluvial sample Bot-03, plotting alongside the eolian samples. This suggests a greater influence of recycled sediments and a reduced contribution of sediments from the Brasília Belt toward the top.

The Brasília Belt configured the northeastern boundary of the Paraná Basin during the Jurassic and Triassic, where paleotopography favored the influx of sediments from marginal elevated areas (^{Campos & Dardenne, 1997}). Furthermore, the intense erosion during the Lower Mesozoic reworked and recycled the sedimentary units overlying the Neoproterozoic basement (^{Campos & Dardenne, 1997}). In this context, both granitic (Maratá Complex) and schist and quartzite of the Araxá Group (Unit B) contributed to the sediments of the proximal fluvial systems of the Botucatu Formation (Fig. 9). The sedimentary input from uplifted Neoproterozoic metamorphic and granitic terrains was also described for deposits of the Botucatu Formation in São Paulo, Paraná, Santa Catarina, and Rio Grande do Sul. (^{Bertolini et al., 2020}; ^{Bertolini et al., 2021b}; ^{Campos & Dardenne, 1997}; ^{Scherer & Goldberg, 2007}). However, in these regions, the sediment influx originates from other Neoproterozoic mobile belts beyond the Brasília Belt, like the Ribeira, Dom Feliciano, and Paraguay belts, and the Rio Apa Craton (^{Bertolini et al., 2020}; ^{Bertolini et al., 2021b}; ^{Philipp et al., 2023}).

Figure 9
Paleogeographic model of the Botucatu Formation on the northeastern part of the Paraná Basin.

5.3 Paleogeographic reconstruction

The unusual and rare sedimentary deposits described here are characterized by deposition in ephemeral rivers. The presence of eolian deposits to the SE and E of the study area suggests that these fluvial systems developed in the marginal areas of the extensive dune field of the Botucatu Formation. The proximal fluvial system was also situated near uplifted terrains to the west and probably influenced by drainages originating in these highlands and flowing eastward (Fig. 10). Thus, much of the fluvial deposits of the Botucatu Formation has been eroded or overlain by volcanic rocks of the Serra Geral Formation (^{Seer & Moraes, 2017}).

Figure 10
(A) Simplified geological map of the study area. (B) Detailed view of the area shown in “A.”

The basaltic flows, which produced pillow lavas in areas very close to the study area (approximately 2 km), were responsible for mixing the wet fluvial sediments with volcanic rocks, forming true peperitic breccia (^{Famelli et al., 2021a}; ^2021b). Notably, the presence of unconsolidated fluvial and lacustrine deposits (^{Famelli et al., 2021a}) interbedded with lava flows indicates periods of volcanic quiescence. The alternation between fluvial sedimentation and volcanism suggests a continuous influx of sediments through ephemeral flows during the magmatic pulses that formed the Serra Geral lavas. The presence of pillow lavas and clay sediments within the thick lava package further suggests that ephemeral flows responsible for depositing the conglomeratic facies may have accumulated in nearby areas, leading to the formation of small lakes (^{Famelli et al., 2021a}). Despite the predominantly dry climate that characterized much of Gondwana’s interior during the Cretaceous, the occurrence of fine-grained facies interbedded with pillow lavas indicates localized humid conditions (^{Famelli et al., 2021a}). Similarly, the presence of pillow lavas and peperites in different parts of the Paraná-Etendeka Igneous Province reinforces the punctual presence of more humid areas in the margins of the Botucatu paleodesert (^{Famelli et al., 2021a}; ^2021b). Indeed, dune field margins host fluvial systems, with humid areas developing adjacent to fluvial channels and floodplains (^{Bookfiled & Silvestro, 2010}) and the incidence of rivers with high-competent flows coming from the topographic highs of the Brasília Belt. Ultimately, these rivers were a response to delivering sediments by migration of longitudinal sandy bars, gravel bars, and residual deposits to the desertic area (marginal sand sheets and dune fields). During drier periods, these ephemeral rivers could completely dry out, leaving sediments exposed to wind reworking. The observed increase in grain sphericity and rounding, along with higher ZTR index values from the base to the top, further supports sediment reworking. Although the incorporation of eolian grains into the fluvial system remains a possibility, the available data are insufficient to confirm this process definitively. However, the evident changes in textural and mineralogical characteristics, particularly the increase in ultrastable mineral content toward the top, suggest sedimentary reworking. These patterns of intense sediment recycling in the Botucatu Formation are explored and discussed in ^{Bertolini et al. (2020}, ^2021b). Finally, the role of aqueous flows in sediment transport and influx in the Paraná Basin highlights the sedimentary dynamics history of the region. Even after the onset of volcanic eruptions of the Serra Geral Formation, ephemeral aqueous flows continued to influence the sedimentation, as evidenced by the presence of fluvial and lacustrine deposits interbedded within basaltic sequences (^{Moraes & Seer, 2018}).

6 CONCLUSIONS

In the northeastern margin of the Paraná Basin, the Botucatu Formation consists of trough cross-bedded coarse-grained sandstone and conglomeratic deposits, formed by the amalgamation of fluvial longitudinal sandy bars in an ephemerous fluvial system probably subjected to pronounced climate seasonal variations. This fluvial system, located in the margins of the Botucatu paleodesert, flowed southeastward over the Neoproterozoic basement, representing localized, temporary humid areas within an otherwise arid to desert climate. Paleoflow direction, gravel composition, and heavy mineral analysis indicate a significant contribution of muscovite-quartz schist and quartz, likely sourced by Unit B from the Araxá Group from the Brasília Belt. The upper part of the section exhibits a notable decrease in metamorphic components and an increase in zircon and tourmaline concentrations. This shift, combined with the increased grain sphericity and rounding, may suggest the reworking of fluvial deposits during dry seasons when ephemeral channels dried out or a clastic supply derived from older siliciclastic successions in the source area, reflecting multicycle detritus. Compositional clustering reveals distinct sedimentary domains within the paleodesert: intensely reworked, ZTR-rich portions typically associated with the central part of the paleoerg and marginal domains receiving sedimentary input from both recycling processes and elevated terrains bordering the paleodesert. Although the interpretations and discussions presented here, the studied outcrop represents a rare occurrence of fluvial facies within the Botucatu Formation. This rarity is primarily due to the small size and/or poor preservation of most outcrops. Nevertheless, our comprehensive dataset provides a novel insight into these rare deposits. Thus, the findings presented here not only enhance our understanding of this unit but also offer a valuable and novel contribution to the characterization of fluvial deposits along erg margins.

ACKNOWLEDGMENTS

The authors thank the Institute of Geosciences and Exact Sciences (IGCE), UNESP, Rio Claro, for providing the necessary infrastructure. We also extend our thanks to the Center for Geosciences Applied to Petroleum Geology (UNESPetro), IGCE/UNESP/Rio Claro, for supporting the development of the study. L.G.S. Albino is a fellow of a master’s scholarship from the Training of Human Resources in Petroleum Geology Program—Unesp—PRH 40.1-ANP, funded by financial investment of oil companies qualified under the PD&I Clause of ANP (Resolution No. 50/2015). Rodrigo I. Cerri is a research fellow of the São Paulo Research Foundation (FAPESP). Lucas V. Warren is a research fellow of the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq).

ARTICLE INFORMATION

• Manuscript ID: 20240066. Received on: 13 NOV. 2024. Approved on: 31 MAR. 2025.
  How to cite: Albino, L. G. S., Alessandretti, L., Warren, L., Bertolini, G., & Cerri, R. I. (2025). Sedimentary fluvial processes at the margin of an eolian sand sea: the example of the Botucatu Formation, Paraná Basin. Brazilian Journal of Geology, 55, e20240066. https://doi.org/10.1590/2317-488920240066

REFERENCES

• ^1. Almeida, F. D. (1953). Botucatu, a triassic desert of South America. In XIX International Geological Congress, Algiers 1952 (Vol. 7, pp. 9-24).
• ^2. Almeida, F. D., & Melo, M. D. (1981). A Bacia do Paraná e o vulcanismo mesozóico. Mapa Geológico do Estado de São Paulo (pp. 46-81). Escala: 1:500.000.
• ^3. Assine, M. L., Piranha, J. M., & Carneiro, C. D. R. (2004). Os paleodesertos Pirambóia e Botucatu. In C. R. Grohmann (Ed.), Geologia do continente sul-americano: Evolução da obra de Fernando Flávio Marques de Almeida (pp. 71-88). Beca.
• ^4. Bahlburg, H., Vervoort, J. D., Du Frane, S. A., Bock, B., Augustsson, C., & Reimann, C. (2009). Timing of crust formation and recycling in accretionary orogens: Insights learned from the western margin of South America. Earth-Science Reviews, 97(1-4), 215-241. https://doi.org/10.1016/j.earscirev.2009.10.006
  » https://doi.org/10.1016/j.earscirev.2009.10.006
• ^5. Bigarella, J. J., & Salamuni, R. (1961). Early Mesozoic wind patterns as suggested by dune bedding in the Botucatu Sandstone of Brazil and Uruguay. Geological Society of America Bulletin, 72(7), 1089-1105. https://doi.org/10.1130/0016-7606(1961)72[1089:EMWPAS]2.0.CO;2
  » https://doi.org/10.1130/0016-7606(1961)72[1089:EMWPAS]2.0.CO;2
• ^6. Bertolini, G., Hartley, A. J., Marques, J. C., Healy, D., & Frantz, J. C. (2021). The effects of basaltic lava flows on the petrophysical properties and diagenesis of interbedded aeolian sandstones: an example from the Cretaceous Paraná Basin, Brazil. Petroleum Geoscience, 27(2), petgeo2020-036. https://doi.org/10.1144/petgeo2020-036
  » https://doi.org/10.1144/petgeo2020-036
• ^7. Bertolini, G., Marques, J. C., Hartley, A. J., Basei, M. A. S., Frantz, J. C., & Santos, P. R. D. (2021). Determining sediment provenance history in a Gondwanan erg: Botucatu formation, Northern Paraná Basin, Brazil. Sedimentary Geology, 417, 105883. https://doi.org/10.1016/j.sedgeo.2021.105883
  » https://doi.org/10.1016/j.sedgeo.2021.105883
• ^8. Bertolini, G., Marques, J. C., Hartley, A. J., Da-Rosa, A. A., Scherer, C. M., Basei, M. A., & Frantz, J. C. (2020). Controls on Early Cretaceous desert sediment provenance in south-west Gondwana, Botucatu Formation (Brazil and Uruguay). Sedimentology, 67(5), 2672-2690. https://doi.org/10.1111/sed.12715
  » https://doi.org/10.1111/sed.12715
• ^9. Bookfiled, M. E., & Silvestro, S. (2010). Eolian systems. In: Dalrymple & James (eds.). Facies Models 4 (pp. 139-166). Geological Association of Canada.
• ^10. Campanha, G. A. D. C., Faleiros, F. M., Cawood, P. A., Cabrita, D. I. G., Ribeiro, B. V., & Basei, M. A. S. (2019). The Tonian Embu complex in the Ribeira belt (Brazil): revision, depositional age and setting in Rodinia and West Gondwana. Precambrian Research, 320, 31-45. https://doi.org/10.1016/j.precamres.2018.10.010
  » https://doi.org/10.1016/j.precamres.2018.10.010
• ^11. Campos, J. E. G., & Dardenne, M. A. (1997). Origem e evolução tectônica da Bacia Sanfranciscana. Revista Brasileira de Geociências, 27(3), 283-294.
• ^12. Cichowolski, M., Rustán, J. J., & Uriz, N. J. (2018). Middle Palaeozoic cephalopods from southwestern Gondwana and their palaeobiogeographic meaning. In X Internacional Symposium: Cephalopods “Present and Past” (Vol. 21).
• ^13. De Ros, L. F., & Scherer, C. M. (2013). Stratigraphic controls on the distribution of diagenetic processes, quality and heterogeneity of fluvial-aeolian reservoirs from the Recôncavo Basin, Brazil. Linking Diagenesis to Sequence Stratigraphy, 105-132. https://doi.org/10.1002/9781118485347.ch5
  » https://doi.org/10.1002/9781118485347.ch5
• ^14. De Ros, L. F., Sgarbi, G. N. C., & Morad, S. (1994). Multiple authigenesis of K-feldspar in sandstones; evidence from the Cretaceous Areado Formation, Sao Francisco Basin, central Brazil. Journal of Sedimentary Research, 64(4a), 778-787. https://doi.org/10.1306/D4267EBF-2B26-11D7-8648000102C1865D
  » https://doi.org/10.1306/D4267EBF-2B26-11D7-8648000102C1865D
• ^15. Dickinson, W. R. (1970). Interpreting detrital modes of graywacke and arkose. Journal of Sedimentary Research, 40(2), 695-707. https://doi.org/10.1306/74D72018-2B21-11D7-8648000102C1865D
  » https://doi.org/10.1306/74D72018-2B21-11D7-8648000102C1865D
• ^16. Dickinson, W. R. (1985). Interpreting provenance relations from detrital modes of sandstones. In Provenance of arenites (pp. 333-361). Springer Netherlands. https://doi.org/10.1007/978-94-017-2809-6_15
  » https://doi.org/10.1007/978-94-017-2809-6_15
• ^17. Famelli, N., Lima, E. F., & Carmo, I. D. O. (2021a). Lithostratigraphy of the Serra Geral Formation in the northern portion of the Paraná-Etendeka Igneous Province: A tool for tracking Early Cretaceous paleoenvironmental changes. Journal of Volcanology and Geothermal Research, 410, 107152. https://doi.org/10.1016/j.jvolgeores.2020.107152
  » https://doi.org/10.1016/j.jvolgeores.2020.107152
• ^18. Famelli, N., Millett, J. M., Hole, M. J., Lima, E. F., Carmo, I. D. O., Jerram, D. A., Jolley, D.W., Pugsley, J. H., & Howell, J. A. (2021b). Characterizing the nature and importance of lava-sediment interactions with the aid of field outcrop analogues. Journal of South American Earth Sciences, 108, 103108. https://doi.org/10.1016/j.jsames.2020.103108
  » https://doi.org/10.1016/j.jsames.2020.103108
• ^19. Fernandes, M. A., Ghilardi, A. M., & de Souza Carvalho, I. (2014). Paleodeserto Botucatu: Inferências ambientais e climáticas com base na ocorrência de icnofósseis. Paleontologia: Cenários de vida (Vol. 5, pp. 71-80). Interciência.
• ^20. Fielding, C. R., Alexander, J., & Allen, J. P. (2018). The role of discharge variability in the formation and preservation of alluvial sediment bodies. Sedimentary Geology, 365, 1-20. https://doi.org/10.1016/j.sedgeo.2017.12.022
  » https://doi.org/10.1016/j.sedgeo.2017.12.022
• ^21. Fielding, C. R., Alexander, J., & Argiro, A. (2025). Towards a new generation of fluvial facies models for the interpretation of ancient deposits, based on inter-annual peak discharge variance of modern rivers. Sedimentology, 72(1), 275-329. https://doi.org/10.1111/sed.13234
  » https://doi.org/10.1111/sed.13234
• ^22. Fluteau, F., Ramstein, G., Besse, J., Guiraud, R., & Masse, J. P. (2007). Impacts of palaeogeography and sea level changes on Mid-Cretaceous climate. Palaeogeography, Palaeoclimatology, Palaeoecology, 247(3-4), 357-381. https://doi.org/10.1016/j.palaeo.2006.11.016
  » https://doi.org/10.1016/j.palaeo.2006.11.016
• ^23. Folk, R. L. (1968). Petrology of sedimentary rocks (170 p.). Hemphill’s Publishing Company.
• ^24. Gazzi, P. (1966). Le arenarie del flysch sopracretaceo dell’Appennino modenese; correlazioni con il flysch di Monghidoro. Mineralogica et Petrographica Acta, 12, 69-97.
• ^25. Grove, C., Jerram, D. A., Gluyas, J. G., & Brown, R. J. (2017). Sandstone diagenesis in sediment–lava sequences: exceptional examples of volcanically driven diagenetic compartmentalization in Dune Valley, Huab Outliers, NW Namibia. Journal of Sedimentary Research, 87(12), 1314-1335. https://doi.org/10.2110/jsr.2017.75
  » https://doi.org/10.2110/jsr.2017.75
• ^26. Haughton, P. D., Todd, S. P., & Morton, A. C. (1991). Sedimentary provenance studies. Geological Society, London, Special Publications, 57(1), 1-11. https://doi.org/10.1144/GSL.SP.1991.057.01.01
  » https://doi.org/10.1144/GSL.SP.1991.057.01.01
• ^27. Henrique-Pinto, R., Basei, M. A. S., Santos, P. R. D., Saad, A. R., Milani, E. J., Cingolani, C. A., & Frugis, G. L. (2021). Paleozoic Paraná Basin transition from collisional retro-foreland to pericratonic syneclise: Implications on the geodynamic model of Gondwana proto-Andean margin. Journal of South American Earth Sciences, 111, 103511. https://doi.org/10.1016/j.jsames.2021.103511
  » https://doi.org/10.1016/j.jsames.2021.103511
• ^28. Holz, M., Küchle, J., Philipp, R. P., Bischoff, A. P., & Arima, N. (2006). Hierarchy of tectonic control on stratigraphic signatures: base-level changes during the Early Permian in the Paraná Basin, southernmost Brazil. Journal of South American Earth Sciences, 22(3-4), 185-204. https://doi.org/10.1016/j.jsames.2006.09.007
  » https://doi.org/10.1016/j.jsames.2006.09.007
• ^29. Howard, J. L. (1993). The statistics of counting clasts in rudites: a review, with examples from the upper Palaeogene of southern California, USA. Sedimentology, 40(2), 157-174. https://doi.org/10.1111/j.1365-3091.1993.tb01759.x
  » https://doi.org/10.1111/j.1365-3091.1993.tb01759.x
• ^30. Hubert, J. F. (1962). A zircon-tourmaline-rutile maturity index and the interdependence of the composition of heavy mineral assemblages with the gross composition and texture of sandstones. Journal of Sedimentary Research, 32(3), 440-450. https://doi.org/10.1306/74D70CE5-2B21-11D7-8648000102C1865D
  » https://doi.org/10.1306/74D70CE5-2B21-11D7-8648000102C1865D
• ^31. Ingersoll, R. V., Bullard, T. F., Ford, R. L., Grimm, J. P., Pickle, J. D., & Sares, S. W. (1984). The effect of grain size on detrital modes: a test of the Gazzi-Dickinson point-counting method. Journal of Sedimentary Research, 54(1), 103-116. https://doi.org/10.1306/212F83B9-2B24-11D7-8648000102C1865D
  » https://doi.org/10.1306/212F83B9-2B24-11D7-8648000102C1865D
• ^32. Kessler, L. G. (1978). Diagenetic sequence in ancient sandstones deposited under desert climatic conditions. Journal of the Geological Society, 135(1), 41-49. https://doi.org/10.1144/gsjgs.135.1.0041
  » https://doi.org/10.1144/gsjgs.135.1.0041
• ^33. Manes, M. I. L., da Silva, R. C., & Scheffler, S. M. (2021). Dinosaurs and rivers on the edge of a desert: A first recognition of fluvial deposits associated to the Botucatu Formation (Jurassic/Cretaceous), Brazil. Journal of South American Earth Sciences, 110, 103339. https://doi.org/10.1016/j.jsames.2021.103339
  » https://doi.org/10.1016/j.jsames.2021.103339
• ^34. Mange, M. A., & Maurer, H. (2012). Heavy minerals in colour Springer Science & Business Media.
• ^35. McGee, B., Babinski, M., Trindade, R., & Collins, A. S. (2018). Tracing final Gondwana assembly: Age and provenance of key stratigraphic units in the southern Paraguay Belt, Brazil. Precambrian Research, 307, 1-33. https://doi.org/10.1016/j.precamres.2017.12.030
  » https://doi.org/10.1016/j.precamres.2017.12.030
• ^36. McGee, B., Collins, A. S., Trindade, R. I., & Payne, J. (2015). Age and provenance of the Cryogenian to Cambrian passive margin to foreland basin sequence of the northern Paraguay Belt, Brazil. Geological Society of America Bulletin, 127(1-2), 76-86. https://doi.org/10.1130/B30842.1
  » https://doi.org/10.1130/B30842.1
• ^37. Miall, A. D. (1974). Paleocurrent analysis of alluvial sediments: a discussion of directional variance and vector magnitude. Journal of Sedimentary Research, 44(4), 1174-1185. https://doi.org/10.1306/212F6C6C-2B24-11D7-8648000102C1865D
  » https://doi.org/10.1306/212F6C6C-2B24-11D7-8648000102C1865D
• ^38. Miall, A. D. (1977). Fluvial sedimentology: An historical review. In A. D. Miall (Ed.), Fluvial sedimentology (pp. 1-22). Canadian Society of Petroleum Geologists.
• ^39. Miall, A. D. (1985). Architectural-element analysis: a new method of facies analysis applied to fluvial deposits. Earth-Science Reviews, 22(4), 261-308. https://doi.org/10.1016/0012-8252(85)90001-7
  » https://doi.org/10.1016/0012-8252(85)90001-7
• ^40. Miall, A. D. (1996). The geology of fluvial deposits: Sedimentary facies, basin analysis, and petroleum geology Springer-Verlag.
• ^41. Miall, A. D. (2010). Alluvial deposits. In James & Dalrymple (eds). Facies Models 4 (pp. 105-138). Geological Association of Canada.
• ^42. Miall, A. D. (2013). The geology of fluvial deposits: Sedimentary facies, basin analysis, and petroleum geology Springer.
• ^43. Miall, A. D. (2014). Fluvial depositional systems (Vol. 14, p. 316). Springer International Publishing.
• ^44. Milani, E. J., Araujo, L. M., Cupertino, J. A., Faccini, U. F., & Scherer, C. M. (1998). Sequences and stratigraphic hierarchy of the Paraná Basin (Ordovician-Cretaceous), Southern Brazil. Boletim IG-USP. Série Científica, 29, 1-36. https://doi.org/10.11606/issn.2316-8986.v29i0p125-173
  » https://doi.org/10.11606/issn.2316-8986.v29i0p125-173
• ^45. Milani, E. J., Melo, J. H. G., Souza, P. A., Fernandes, L. A., & França, A. B. (2007). Bacia do Paraná. Boletim de Geociências da Petrobras, 15(2), 265-287.
• ^46. Miyazaki, M. R., & Basilici, G. (2015). Interação eólica-fluvial na Formação Pirambóia (Permiano-Triássico, oeste de Rio Claro, SP) (Doctoral dissertation).
• ^47. Moraes, L. C., Seer, H. J. (2018). Pillow lavas and fluvio-lacustrine deposits in the northeast of Paraná Continental Magmatic Province, Brazil. Journal of Volcanology and Geothermal Research, 355, 78-86. https://doi.org/10.1016/j.jvolgeores.2017.03.024
  » https://doi.org/10.1016/j.jvolgeores.2017.03.024
• ^48. Moraes, M. A., & De Ros, L. F. (1990). Infiltrated clays in fluvial Jurassic sandstones of Recôncavo Basin, northeastern Brazil. Journal of Sedimentary Research, 60(6), 809-819. https://doi.org/10.1306/212F928C-2B24-11D7-8648000102C1865D
  » https://doi.org/10.1306/212F928C-2B24-11D7-8648000102C1865D
• ^49. Morton, A. C., & Hallsworth, C. (1994). Identifying provenance-specific features of detrital heavy mineral assemblages in sandstones. Sedimentary Geology, 90(3-4), 241-256. https://doi.org/10.1016/0037-0738(94)90041-8
  » https://doi.org/10.1016/0037-0738(94)90041-8
• ^50. Parrish, J. T. (1993). Climate of the supercontinent Pangea. Journal of Geology, 101(2), 215-233. https://doi.org/10.1086/648217
  » https://doi.org/10.1086/648217
• ^51. Peixoto, B., Sedorko, D., de Barros, G. E., Francischini, H., Ghilardi, R. P., & Fernandes, M. A. (2025). Pulses of life: Wet events in Botucatu Paleodesert evidenced by trace fossils analysis (earliest Cretaceous, Paraná Basin, Brazil). Palaeogeography, Palaeoclimatology, Palaeoecology, 658, 112608. https://doi.org/10.1016/j.palaeo.2024.112608
  » https://doi.org/10.1016/j.palaeo.2024.112608
• ^52. Philipp, R. P., Faccini, U. F., Schultz, C. L., Zvirtes, G., Bruckmann, M. P., Lavina, E., Cagliari, J., Ritter Jelinek, A., Guimarães Netto, R., Roessler Viana, A., & Basei, M. A. S. (2023). U-Pb zircon geochronology of detrital and ash fall deposits of the southern Paraná Basin: A contribution for provenance, tectonic evolution, and the paleogeography of the SW Gondwana. Geosciences, 13(8), 225. https://doi.org/10.3390/geosciences13080225
  » https://doi.org/10.3390/geosciences13080225
• ^53. Pires, E. F., Guerra-Sommer, M., dos Santos Scherer, C. M., dos Santos, A. R., & Cardoso, E. (2011). Early Cretaceous coniferous woods from a paleoerg (Paraná Basin, Brazil). Journal of South American Earth Sciences, 32(1), 96-109. https://doi.org/10.1016/j.jsames.2011.04.001
  » https://doi.org/10.1016/j.jsames.2011.04.001
• ^54. Piuzana, D., Pimentel, M. M., Fuck, R. A., & Armstrong, R. (2003). SHRIMP U–Pb and Sm–Nd data for the Araxá Group and associated magmatic rocks: constraints for the age of sedimentation and geodynamic context of the southern Brasılia Belt, central Brazil. Precambrian Research, 125(1-2), 139-160. https://doi.org/10.1016/S0301-9268(03)00107-4
  » https://doi.org/10.1016/S0301-9268(03)00107-4
• ^55. Reis, A. D., dos Santos Scherer, C. M., do Amarante, F. B., Rossetti, M. D. M. M., Kifumbi, C., de Souza, E. G., Ferronatto, J. P. F., & Owen, A. (2019). Sedimentology of the proximal portion of a large-scale, Upper Jurassic fluvial-aeolian system in Paraná Basin, southwestern Gondwana. Journal of South American Earth Sciences, 95, 102248. https://doi.org/10.1016/j.jsames.2019.102248
  » https://doi.org/10.1016/j.jsames.2019.102248
• ^56. Remus, M. V. D., Souza, R. S., Cupertino, J. A., De Ros, L. F., Dani, N., & Vignol-Lelarge, M. L. (2008). Proveniência sedimentar: métodos e técnicas analíticas aplicadas. Brazilian Journal of Geology, 38(2), 166-185. https://doi.org/10.25249/0375-7536.2008382S166185
  » https://doi.org/10.25249/0375-7536.2008382S166185
• ^57. Scherer, C. M. (2000). Eolian dunes of the Botucatu Formation (Cretaceous) in southernmost Brazil: morphology and origin. Sedimentary Geology, 137(1-2), 63-84. https://doi.org/10.1016/S0037-0738(00)00135-4
  » https://doi.org/10.1016/S0037-0738(00)00135-4
• ^58. Scherer, C. M. (2002). Preservation of aeolian genetic units by lava flows in the Lower Cretaceous of the Paraná Basin, southern Brazil. Sedimentology, 49(1), 97-116. https://doi.org/10.1046/j.1365-3091.2002.00434.x
  » https://doi.org/10.1046/j.1365-3091.2002.00434.x
• ^59. Scherer, C. M., & Goldberg, K. (2007). Palaeowind patterns during the latest Jurassic–earliest Cretaceous in Gondwana: evidence from aeolian cross-strata of the Botucatu Formation, Brazil. Palaeogeography, Palaeoclimatology, Palaeoecology, 250(1-4), 89-100. https://doi.org/10.1016/j.palaeo.2007.02.018
  » https://doi.org/10.1016/j.palaeo.2007.02.018
• ^60. Scherer, C. M., Lavina, E. L. (2006). Stratigraphic evolution of a fluvial–eolian succession: the example of the Upper Jurassic—Lower Cretaceous Guará and Botucatu formations, Paraná Basin, Southernmost Brazil. Gondwana Research, 9(4), 475-484. https://doi.org/10.1016/j.gr.2005.12.002
  » https://doi.org/10.1016/j.gr.2005.12.002
• ^61. Scherer, C. M., Reis, A. D., Horn, B. L., Bertolini, G., Lavina, E. L., Kifumbi, C., & Aguilar, C. G. (2023). The stratigraphic puzzle of the permo-mesozoic southwestern Gondwana: The Paraná Basin record in geotectonic and palaeoclimatic context. Earth-Science Reviews, 240, 104397. https://doi.org/10.1016/j.earscirev.2023.104397
  » https://doi.org/10.1016/j.earscirev.2023.104397
• ^62. Scotese, C. R., Song, H., Mills, B. J., & van der Meer, D. G. (2021). Phanerozoic paleotemperatures: The earth’s changing climate during the last 540 million years. Earth-Science Reviews, 215, 103503. https://doi.org/10.1016/j.earscirev.2021.103503
  » https://doi.org/10.1016/j.earscirev.2021.103503
• ^63. Seer, H. J., & Moraes, L. C. (2017). Geologia regional do Triângulo Mineiro CODEMIG/UFMG.
• ^64. Selley, R. C. (1982). An introduction to sedimentology (449 p.). Academic Press.
• ^65. Soares, P. C. (1975). Divisão estratigráfica do Mesozóico no estado de São Paulo. Revista Brasileira de Geociências, 5(4), 229-251.
• ^66. Stanistreet, I. G., & Stollhofen, H. (1999). Onshore equivalents of the main Kudu gas reservoir in Namibia. Geological Society, London, Special Publications, 153(1), 345-365. https://doi.org/10.1144/GSL.SP.1999.153.01.21
  » https://doi.org/10.1144/GSL.SP.1999.153.01.21
• ^67. Suguio, K., & Coimbra, A. M. (1972). Madeira fóssil silicificada na Formação Botucatu. Ciência e Cultura, 24(11), 1049-1055.
• ^68. Tucker, M. E. (1991). An introduction to the origin of sedimentary rocks (2nd ed.). John Wiley & Sons.
• ^69. Veevers, J. J. (1990). Tectonic-climatic supercycle in the billion-year plate-tectonic eon: Permian Pangean icehouse alternates with Cretaceous dispersed-continents greenhouse. Sedimentary Geology, 68(1-2), 1-16. https://doi.org/10.1016/0037-0738(90)90116-B
  » https://doi.org/10.1016/0037-0738(90)90116-B
• ^70. Vermeesch, P., Resentini, A., & Garzanti, E. (2016). An R package for statistical provenance analysis. Sedimentary Geology, 336, 14-25. https://doi.org/10.1016/j.sedgeo.2016.01.009
  » https://doi.org/10.1016/j.sedgeo.2016.01.009
• ^71. Walker, R. G., & Middleton, G. V. (1977). Facies models 9: Eolian sands. Geoscience Canada, 4(4), 55-63.
• ^72. Walker, T. R., Waugh, B., & Grone, A. J. (1978). Diagenesis in first-cycle desert alluvium of Cenozoic age, southwestern United States and northwestern Mexico. Geological Society of America Bulletin, 89(1), 19-32. https://doi.org/10.1130/0016-7606(1978)89≤19:DIFDAO≥2.0.CO;2
  » https://doi.org/10.1130/0016-7606(1978)89≤19:DIFDAO≥2.0.CO;2
• ^73. Zalán, P. V., Wolff, S. J. C. J., Conceição, J. D. J., Marques, A., Astolfi, M. A. M., Vieira, I. S., & Zanotto, O. A. (1990). Bacia do Paraná. Origem e evolução das bacias sedimentares (pp. 135-168). Universidade de São Paulo.
• ^74. Zuffa, G. G. (1980). Hybrid arenites; their composition and classification. Journal of Sedimentary Research, 50(1), 21-29. https://doi.org/10.1306/212F7950-2B24-11D7-8648000102C1865D
  » https://doi.org/10.1306/212F7950-2B24-11D7-8648000102C1865D

Edited by

• SCIENTIFIC EDITOR:
  Carlos Grohmann http://orcid.org/0000-0001-5073-5572

• ASSOCIATE EDITOR:
  Andres Folguera http://orcid.org/0000-0001-8965-8543

Publication Dates

• Publication in this collection
  02 June 2025
• Date of issue
  2025

History

• Received
  13 Nov 2024
• Accepted
  31 Mar 2025