Re-investigating Miocene age control and paleoenvironmental reconstructions in western Amazonia (northwestern Solimões Basin, Brazil)

Journal Pre-proof Re-investigating Miocene age control and paleoenvironmental reconstructions in western Amazonia (northwestern Solimões Basin, Brazil)

Andrea K. Kern, Martin Gross, Cristiano P. Galeazzi, Fabiano N. Pupim, André O. Sawakuchi, Renato P. Almeida, Werner E. Piller, Gabriel G. Kuhlmann, Miguel A.S. Basei PII:

S0031-0182(20)30097-3

DOI:

https://doi.org/10.1016/j.palaeo.2020.109652

Reference:

PALAEO 109652

To appear in:

Palaeogeography, Palaeoclimatology, Palaeoecology

Received date:

17 May 2019

Revised date:

27 January 2020

Accepted date:

7 February 2020

Please cite this article as: A.K. Kern, M. Gross, C.P. Galeazzi, et al., Re-investigating Miocene age control and paleoenvironmental reconstructions in western Amazonia (northwestern Solimões Basin, Brazil), Palaeogeography, Palaeoclimatology, Palaeoecology (2020), https://doi.org/10.1016/j.palaeo.2020.109652

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

© 2020 Published by Elsevier.

Journal Pre-proof Re-investigating Miocene age control and paleoenvironmental reconstructions in western Amazonia (northwestern Solimões Basin, Brazil) Andrea K. Kern*1, Martin Gross2, Cristiano P. Galeazzi1, Fabiano N. Pupim1,3, André O. Sawakuchi1, Renato P. Almeida1, Werner E. Piller4, Gabriel G. Kuhlmann1, Miguel A.S. Basei1 Instituto de Geociências, Universidade de São Paulo, São Paulo, SP, Brazil

2

Department for Geology and Palaeontology, Universalmuseum Joanneum, Graz, Austria.

3

Departamento de Ciências Ambientais, Universidade Federal de São Paulo, Diadema, SP, Brazil

of

1

4

ro

Institute of Earth Science, NAWI Graz Geocenter, University of Graz, Graz, Austria

re

-p

*[email protected]

lP

Abstract

The modern Amazonian rainforest has a great fascination and global significance, still its past biodiversity and paleoecology includes various controversies based on inaccurate and sparse data but

na

our knowledge of past landscape changes is still limited due to sparse data and the lack of radiometric age constrains. Precise dating in records older than the late Pleistocene are difficult to obtain and often

ur

regionally confined, therefore biostratigraphic correlations are used to estimate depositional ages across

Jo

sedimentary basins. The lack of age constrains in records older than the late Pleistocene hinders the integration of paleoenvironmental data across different sedimentary basins limiting large-scale reconstructions limited to biostratigraphic correlations. Aiming for a better understanding of the Neogene environmental changes in western Brazil Amazonia, we studied palynology and ostracod assemblages from the Solimões Formation, a key site for the paleoenvironmental models to discuss the biotic changes in Amazonia the Amazonian biodiversity. Further, we compare these biostratigraphic information of palynological and ostracod data both groups with radiometric maximum deposition ages constrained by U/Pb measurements on detrital zircon grains. For the topmost section of the Solimões Formation d Down to the depth of ~95 m, zircon populations show a maximum deposition age of 11.42 ± 0.66 Ma documenting Tortonian (late Miocene ) or younger ages for the top of the Solimões Formation (Tortonian/late Miocene). Palynological biostratigraphy 1

Journal Pre-proof show the Younger ages are indicated by the palynological Psilatricolporites caribbiensis zone (latest Miocene–Pliocene) reaching down to a depth of 181.8 m, where the Grimsdalea magnaclavata zone suggests a late middle Miocene to early late Miocene age. In comparison, ostracod biostratigraphy indicates the Cyprideis cyrtoma zone (late Miocene) down to 176.0 m and the Cyprideis minipunctata (late middle to late Miocene) for the bottom of the section. Thus, both biostratigraphic concepts suggest ages close to the maximum ages indicated by the detrital zircons. This comparison of the biozonations with U/Pb maximum deposition ages represents the first independent calibration of these widely used methods for age determination in the Solimões Basin.

of

Palynological and ostracod data illustrate dynamic environments varying from fluvial settings with

ro

wetland forests and swamps to lakes and/or abandoned channels with temporary slightly saline conditions. Instead of a slow evolution of Neogene Amazonian environments, our results show

-p

migrating rivers and waterbodies surrounded by rainforest during the period recorded by the base and top of the studied core material, separated by a phase of more extensive lakes and swamps (core depth

re

90–120 m). Indicators for slightly elevated salinity are found until the topmost sample and thus

lP

persisted during the late Miocene.

This highlights the importance of new age control to improve the regional biostratigraphic timescale and

ur

na

reconstructed environmental changes.

Jo

Keywords: palynology; ostracods; detrital zircon ages; biostratigraphy; Neogene; Amazon rainforest.

1. Introduction

Amazonia is of global importance due to its exceptional biodiversity (Hoorn et al., 2010; Antonelli and Sanmartin, 2011, Rull, 2011) and its importance for global climate (Malhi et al., 2008). However, the documentation of evolution of its species and environments is still incomplete through time and space. During the Neogene, large-scale geological events influenced the Amazonian landscapes including the Andean Uplift (Horton et al., 2010) and major changes in river drainage patterns (Hoorn et al., 1995; Figueiredo et al., 2009; Mora et al., 2010; Albert et al., 2018). Phases of rivers systems in western Amazonia were interrupted during the middle to late Miocene by the development of the “Pebas megawetland”, an extensive wetland with many short-lived lakes (e.g. Wesselingh et al., 2002; Hoorn et al., 2

Journal Pre-proof 2010). During this time, marine incursions from the Caribbean reached the interior of western Brazil (Hoorn, 1993; Rebata et al., 2006a, 2006b; Hovikoski et al., 2007; Jaramillo et al., 2017; Bernal et al., 2019), although questions on the on the timing and the levels of salinity of inland waters related are ongoing (Latrubesse et al., 2010; Gross et al., 2016). The major focus of most Neogene studies documents the long-term development following established biostatigraphic zones while analyses discussing small-scale paleoenvironmental changes such as alternations of freshwater and saline waterbodies are rare. Considering the extension of Amazonia, there is a relative lack of documentation concerning Neogene

of

deposits across the total area and drilled material for scientific purpose is hardly available. This gap of

ro

knowledge is further complicated by the scarcity of precise age control on sediments in the Amazonian lowlands. Currently, in the entire western Amazonia, Neogene radiometric ages are limited to two Ar/Ar

-p

dates from ash layers in the southeastern Peruvian Madre de Dios Basin (9.01 ± 0.28 Ma and 3.12 ± 0.02 Ma; Campbell et al., 2001), as well as maximum depositional ages of 8.5 ± 0.5 Ma and 10.89 ± 0.13 Ma

re

based on U/Pb measurements of detrital zircon grains from the Solimões Formation in the Brazilian Acre

lP

Basin (Bissaro et al., 2019) (Fig. 1A). The majority of age estimates are currently based on mammal biostratigraphy (e.g. Cozzuol, 2006; Latrubesse et al., 2010; Souza-Filho and Guilherme, 2015), palynostratigraphy (e.g. Lorente, 1986; Muller et al., 1987; Hoorn, 1993, 1994a, 1994b; Jaramillo et al.,

na

2010, 2011), and biostratigraphic schemes based on mollusks (e.g. Wesselingh et al., 2002, 2006a) and ostracods (Muñoz-Torres et al., 2006; Linhares et al., 2019). However, instead calibrating these

ur

biostratigraphic concepts with independent age contains, these methods are correlated to each other

Jo

despite missing western Amazonian verification. Particularly, palynostratigraphy, as the most widespread of all biostratigraphic concepts, is based on a correlation with a few tie point sections in Venezuela and Colombia (e.g. Lorente, 1986; Jaramillo et al., 2011), suggesting coeval vegetation changes across possible barriers and great distances. This highlights the need for an independently established calibration for biostratigraphy in the Solimões Basin. An interesting approach to calibrate biostratigraphy is to target U/Pb age measurements on detrital zircon grains. These report oldest possible ages for a sample as the maximum deposition age and can successfully constrain the lowest boundary for biozones (e.g. Bissaro et al., 2019). Therefore, we studied one drill core from the Solimões Basin (1AS-14-AM; AM14 herein; Fig. 1A-C) of the Alto de Solimões Coal Project (Maia et al., 1977). The same sediment beds were studied for palynological and ostracod content to compare shifts in the extension of local terra firme forest and wetland vegetation and water 3

Journal Pre-proof conditions. Age estimates of palynological and ostracod biostratigraphy were compared to maximum depositional ages based on detrital zircons to discuss persisting controversies in the paleoenvironmental reconstruction of Neogene Amazonia and their age control. Amazonia has global importance due to its exceptional biodiversity (Antonelli et al., 2018) and its importance for global climate (Malhi et al., 2008); however, uncertainties exist on its environmental history. Although a tropical rainforest dominated by ferns and gymnosperms already existed in Amazonia since the Cretaceous (Dino et al., 1999; Jaramillo et al., 2010), little is known about the rise, migration and extinction of most species through time. A key period for the assembly of Amazonian

of

biota is the Neogene (Rull, 2011; Ribas et al., 2012; Smith et al., 2014), when landscape changed, which

ro

possibly strongly contributed to diversification (Hoorn et al., 2010). Large-scale geological events reshaped the environments across northern South America, including Andean Uplift (Horton et al.,

-p

2010), major changes in river drainage patterns (Hoorn et al., 1995; Figueiredo et al., 2009; Mora et al., 2010; Albert et al., 2018) and the occurrence of marine incursions (Hoorn, 1993; Rebata et al., 2006a,

re

2006b; Hovikoski et al., 2007; Latrubesse et al., 2010; Jaramillo et al., 2017; Bacon and Tuomisto, 2019).

lP

These events are center to the majority of discussions on the history of Amazonia but conclusive data on how changes in the physical landscape drive Amazonian diversification are often missing. Due to the hardly accessible outcrops of Neogene sediments across the Amazonian lowlands and the scarcity of

na

available Neogene drilled material for scientific purpose, investigations commonly are limited to largescale comparisons and short time windows of observation. As a result, our understanding of Neogene

ur

paleotopography and its paleoenvironments is still incomplete and further exacerbated by the lack of a

Jo

precise age control on sediment deposition in the Amazonian lowlands. Currently from the entire western Amazonia, Neogene radiometric ages are limited to two Ar/Ar dates from ash layers in the Peruvian Madre de Dios Formation (Cocama ash, 9.01 ± 0.28 Ma; Piedras ash, 3.12 ± 0.02 Ma; Campbell et al., 2001), as well as maximum depositional ages of 8.5 ± 0.5 Ma and 10.89 ± 0.13 Ma based on U/Pb measurements of detrital zircon grains from the Solimões Formation in the Brazilian Acre Basin (Bissaro et al., 2019). These maximum deposition ages are the first independent data in western Amazonia, where previous age estimates are based on mammal biostratigraphy (Cozzuol, 2006; Latrubesse et al., 2010; Souza-Filho and Guilherme, 2015). In northwestern and western Amazonia, biostratigraphic data are predominately based on palynostratigraphy, which was established across Colombia, Venezuela and the Caribbean and further extrapolated towards the lowland sedimentary basins of Peru, Colombia and Brazil (e.g. Lorente, 1986; Muller et al., 1987; Hoorn, 1993, 1994a, 1994b; Jaramillo et al., 2010, 2011). Based on these palynological zones, regional biostratigraphic schemes based on mollusks (e.g. 4

Journal Pre-proof Wesselingh et al., 2002, 2006a) and ostracods (Muñoz-Torres et al., 2006; Linhares et al., 2019) were established. The Neogene Pebas and Solimões formations cover millions of km² in western Amazonia and are famous for their richness in fossils, mainly reptiles, fishes, mollusks, ostracods, leaves and pollen (Wesselingh et al., 2001, Latrubesse et al., 2010; Hoorn et al., 2010). Formerly reconstructed as “Lake Pebas”, researchers nowadays describe this environment as an extensive wetland with many short-lived lakes and rivers (“mega-wetland Pebas”; Hoorn et al., 2010). Despite the distance to the coast, several invertebrate taxa present in these formations are related to marine species from the Caribbean Sea, which promoted a hypothesis of one or multiple marine incursions via the Llanos Basin to western

of

Amazonia (e.g. Hoorn, 1993; Hoorn et al., 2010; Jaramillo et al., 2017). Recent studies, however, raised questions on levels of salinity of inland waters related to western Amazonian deposits (Latrubesse et al.,

ro

2010; Gross et al., 2016). Further, the regional comparison and temporal extent of the proposed marine

-p

settings remain imprecise as biozones estimate suggest large time intervals and radiometric ages are not available.

re

Therefore, a need of new analyses focusing on regional variations in paleoenvironments in connection

lP

with a better age control of these deposits is evident. Recently, optically stimulated luminescence (OSL) dating of sediments overlaying the Solimões Formation in lowlands of western Amazonia indicate late Quaternary deposition ages (250–45 ka) for the high-level fluvial terraces (Pupim et al., 2019), but this

na

technique is not applicable for Neogene deposits. Another approach is to target U/Pb age measurements on detrital zircon grains. Although these report a maximum age of deposition, detrital

ur

zircon ages yield important information to constrain biozones (e.g. Bissaro et al., 2019). To test this

Jo

approach, we studied one drill core from the Solimões Basin (1AS-14-AM; AM14 herein; Fig. 1) of the Alto de Solimões Coal Project (Maia et al., 1977) in relative proximity to previously studied (e.g. Hoorn, 1993; Latrubesse et al., 2010; Jaramillo et al., 2017; Linhares et al., 2019; Leandro et al., 2019) (Fig. 1b; 1c). The same sediment beds were studied for palynological and ostracod content to compare the vegetation and water conditions of the Pebas mega-wetland. Age estimates of both biostratigraphic approaches will be compared to maximum depositional ages based on detrital zircons. Based on this data, we discuss persisting controversies in the paleoenvironmental reconstruction of Neogene Amazonia and their age control.

2. Geological setting

5

Journal Pre-proof The Neogene history of the Amazonian lowlands is recorded in the sedimentary basins along the Andean foothills, which formed as a result of Andean uplift (Garzione et al., 2008). Initial stages of orogeny are difficult to reconstruct as dates and rates for single blocks and mountain ridges are sparse, imprecise and occasionally contradicting (e.g. Horton et al., 2010). Reliable data show multiple major uplift pulses during the Miocene (e.g. Ghosh et al., 2006; Garzione et al., 2006; 2014; Saylor and Horten, 2014; Perez and Horten, 2014), controlling landscapes and strongly effecting regional climate patterns (e.g. Mulch et al., 2010). During the Miocene and Pliocene, uplift extended into the lowlands and created regional topographical highs such as the Vaupes Swell (Mora et al., 2010) and the Fitzcarrald

of

Arch (Räsänen et al., 1987; Espurt et al., 2007) (Fig. 1A).

ro

In the Brazilian western Amazonia, Neogene deposits are grouped in the Solimões and Içá formations (Rego, 1930; Caputo et al., 1971, Maia et al., 1977), which cover most of the Solimões and Acre

-p

sedimentary basins (Fig. 1A). The Solimões Formation is cited to either overlie Paleozoic and Precambrian basement, deposits of the (Paleogene?) Ramon Formation (Maia et al., 1977; Hoorn, 1993)

re

or the (Cretaceous?) Alter do Chão Formation (Latrubesse et al., 2010). The formation thickness is

lP

described to reach a maximum of about 1,000 m (Caputo et al., 1971; Latrubesse et al., 2010) and is covered by fluvial sediments forming several terrace levels attributed to the Içá Formation, originally considered Pliocene–Pleistocene in age (Maia et al., 1977). However, diachronous features in both

ur

et al., 2019).

na

formations complicate a clear stratigraphic boundary (Rossetti et al., 2005; Nogueira et al., 2013; Pupim

The Solimões Formation can further be subdivided due to regional variations; in Acre, late Miocene

Jo

deposits are dominated by medium to fine sandstone, siltstone, mudstone and subordinate by intraformational conglomerates, typically indicating low energy fluvial environments with layers of vertebrate fossils (Cozzuol, 2006; Latrubesse et al., 2010, Bissaro et al., 2019). During the Pliocene, these sediments were uplifted as a result of the development of the Fitzcarrald Arch (Espurt et al., 2007, Fig. 1A). In the western Brazilian lowlands, the Solimões Formation is characterized by fossil rich claystone sequences and intercalations of grey sandstones and lignite (Wesselingh et al., 2010; Latrubesse et al., 2010). Based on sedimentological information from the Solimões Formation and the similar Pebas Formation (Räsänen et al., 1998; Wesselingh et al., 2002), environmental reconstructions varied from shallow, low-energy marine influenced settings with tide- and wave-influenced bay-margins and swamps (e.g. Gingras et al., 2002; Hovikoski et al., 2005, 2007, 2010; Rebata et al., 2006a, 2006b; Wesselingh et al., 2006b; Hoorn et al., 2010) to a mix of channel-dominated and floodplain assemblages (e.g. 6

Journal Pre-proof Latrubesse et al., 2010). Generally, most fossils of the Pebas mega-wetland predominately indicate a freshwater wetland system of multiple lakes and swamps. Lakes did not exceed 10 to 30(?) meters water depth (Wesselingh et al., 2002) and endemism in all invertebrate groups was high, commonly reaching over 85% to even 100% of the assemblages (Purper, 1979; Muñoz-Torres et al., 1998; Wesselingh et al., 2002). Several fossils indicate relations to marine conditions (foraminifers, dinoflagellate cysts, ostracods, mollusks), which are explained by short episodes of connection with the Caribbean Sea (Wesselingh et al., 2006a; Boonstra et al., 2015; Gross et al., 2016). Additional indicators for marine conditions are trace fossils (Gingras et al., 2002; Hovikoski et al., 2007) as well as strontium,

of

oxygen and carbon isotope ratios from mollusk (Vonhof et al., 1998, 2003). These marine phases, however, ended before 13.7 Ma (middle Miocene) due to the uplift of the Vaupes Swell (Fig. 1A), which

ro

separated the Solimões Basin from the north (Mora et al., 2010; Jaramillo et al., 2017). Although the

-p

Solimões Formation is primarily characterized by pelitic deposits and a high abundance of fossils, also many sandstone layers are present (Maia et al., 1977) and paleosols were reported in the area around

re

Eirunepé (Gross et al., 2011, 2013; Fig. 1A) and from core 1AS-105-AM close to Leticia (Jaramillo et al.,

na

3. Material and methods

lP

2017; Fig. 1B).

We studied the core 1AS-14-AM (AM14), drilled along the Solimões River close to the town of

ur

Sururuá (4°21’S/69°36’W, 70 m asl, drilling depth 300.6 m) (Fig. 1B-C) and stored at the Companhia de

Jo

Pesquisa de Recursos Minerais (CPRM) in Manaus. The sedimentary succession comprises alternations of claystone, siltstone and sandstone (Maia et al., 1977). Thicker sequences packages of sandstone are found in the upper 100 m of the core, where single sandstone packages reach up to 10 m (depth 84.5– 94.4m) and 25 m (depth 36–61 m) (Fig. 1D). Downcore from approx. 240 m to the bottom of 300.6 m, sandstone layers are thinner (~1 to 3 m thickness). Several claystone horizons contain coquinas, which are limited to thin layers (~5–15 cm). Lignite is present throughout the core (Fig. 1D) with 27 layers in total varying between 7 cm (depth 70 m) and 1 m (depth 71–72 m); most are found in few meter distance to each other, e.g. seven lignite horizons between 173–189 m core depth and 6 lignite horizons between 98–122 m core depth. From fine grained sediments, 66 samples were taken for palynology and calcareous microfossil analyses (Fig. 1D). Three integrated sandstone samples were taken for dating of detrital zircon grains between 7

Journal Pre-proof the depths of 36.5–45.5; 46.5–53.0 and 84.5–94.5 m (Fig. 1D). No sandstone layers in greater depth could be sampled, as the lower sandstone layers were not preserved in the core storage facility. Likewise, all lignite layers were removed for coal petrography prior to our sampling. 3.1. Detrital zircon U/Pb dating Zircon preparation was done at the University of São Paulo, Institute of Geoscience. Each sample ranging from fine to coarse sandstones was first crushed, disaggregated and sieved. Heavy minerals were separated using a Wilfley table before ferromagnetic and paramagnetic minerals were picked on a

of

Frantz Isodynamic Separator. Afterwards, gravity separation with bromoform (CHBr3) and diiodomethane (CH2I2 ) was performed followed by nitrite nitric acid (HNO3) washes. Paramagnetic

ro

zircon grains were selected removed in high amperages on the by 1.4 A on the Frantz Isodynamic

-p

Separator at 1.4 A. In the grain size fraction above below 0.149 mm, zircon grains were selected randomly for all zircon populations and non-randomly for small and well-formed zircon grains with

re

presumed low rates of reworking and targeting youngest ages. These zircon grains were mounted and covered in epoxy resin, then photographed using a light microscope, a scanning electron microscope

lP

and a cathodoluminescence detector as well as a FEI Quanta 250 Scanning Electron Microscope and XMAX CL detector (Oxford Instruments). Uranium-lead (238U/206Pb) isotopic age measurements were

na

made with an Inductively Coupled Plasma Mass Spectrometer with a Laser Ablation device (LA-ICPMS), using Thermo-Scientific Neptune equipment with Excimer 193 nm laser (Sato et al., 2010). Detailed

ur

analyses using Sensitive High-Resolution Ion Micro-Probe (SHRIMP) following Sato et al. (2014) had been carried out on the pre-selected presumed younger grains only. Statistical comparison between the age

Jo

spectra of all three samples were done using DZmix (Sundell and Saylor, 2017) and MDS analysis following Vermeesh et al. (2013). 3.2. Palynological samples Palynological preparation followed the steps of Faegri and Inversen (1975) using hydrochloric acid (HCl conc.) and hydrofluoric acid (HF conc.) on ~10 g dry sediment to remove all carbonate and silicate matter. Afterwards, sample residue was sieved with a 6 µm nylon mesh sieve to remove small organic matter before slides were prepared in using Canada balsam. No centrifuge was used to ensure fragile dinoflagellates, foraminifera linings or/and large pollen remain intact. If a significant number of pollen and spores were present (supplementary Table 2), at least 200 pollen grains were identified using morphological taxonomy following e.g. Germeraad et al. (1968), Lorente (1986), Hoorn (1993; 8

Journal Pre-proof 1994a,1994b), Jaramillo and Dilcher (2001) and Silva-Caminha et al. (2010) and compared to a fossil pollen collection at the Smithsonian Tropical Research Institute. Palynological zonations followed the same literature as well as Muller et al. (1987) and Jaramillo et al. (2011). Ecological classification and botanical affinities are shown in Table 1. 3.3. Calcareous microfossil samples For analysis of calcareous microfossils, 250 g of dried sediment (40°C for 24 hours) per sample was disintegrated by using diluted hydrogen peroxide (H2O2:H2O=1:5) and washed through standard

of

sieves (63/125/250/500 µm). Wet sieve residues were washed with ethanol (99%) before drying (40°C for 24 hours). Ostracods and foraminifers of the residues ≥250 µm were picked out completely,

ro

identified and counted (except four samples were only splits were investigated: 34.6 m: 1/14, 36.4 m:

-p

1/4, 77.6 m: 1/3, 106.3 m: 1/4). Of the ≥125 µm sieve residue, 0.2 g per sample was qualitatively inspected for its microfossil content (supplementary Table 3). For species identification the publications

4. Results 4.1. Detrital zircon ages

na

lP

and Boonstra et al. (2015) were consulted.

re

of Purper (1979), Muñoz-Torres et al. (1998), Whatley et al. (1998, 2000), Gross et al. (2013, 2014, 2016)

ur

Laser ablation results LA-ICPMS age measurements of the randomly selected detrital zircon grains show

Jo

the age spectra of each sample of show constant measurements for all for provenance and deposition ages, which are interpreted in terms of provenance and maximum deposition ages. The age spectra and contain a mixture of Miocene, Eocene, Jurassic, Triassic and few Paleozoic and Neoproterozoic to Paleoproterozoic detrital zircon grains (Fig. 2A-D); supplementary Table 1). All The detrital zircons show features of low abrasion suggesting low grade of sedimentary reworking (supplementary Figure 1). Strong peaks are between 100 and 400 Ma with the highest around 200 Ma and additional less intense modes signals range around 500 Ma. All older peaks are low and not consistent among samples, showing small peaks intensities around 800 Ma, 1,000 Ma, 1,200 Ma or 2,000 Ma. No peaks components older than 2,000 Ma are significant in AM14. The average of the youngest populations of all three samples in the laser ablation spans around 10.8 ±1.3 Ma (late Miocene; Tortonian) mixed with a high number of older zircons of 15.4 ± 1.1 Ma, 20.14 ± 0.7 Ma as well as many Eocene zircons in the interval of ~35–45 Ma (Fig. 2E). Additional and more precise SHRIMP measurements improved the estimation 9

Journal Pre-proof ages of maximum depositional ages to 11.31 ± 0.39 Ma (depth of 36.5–45.5 m) and 11.42 ± 0.66 Ma (depth of 84.5–94.54 m; Fig. 2F-G, supplementary Figure 1). Therefore, all sediments above the depths of 94.5 m are definitely of late Miocene age or younger. Statistical comparison for provenance using DZMix (Sundell and Saylor, 2017) compares the three age spectra with an inverse Monte Carlo model (10,000 trials). The results were compared by applying Kolmogorov-Smirnov (KS) test D statistic and Kuiper test V statistic, calculated with cumulative distribution plots and cross-correlation coefficient (R2) in probability density plots (PDP) and kernel density estimates (KDE) (supplementary Table 1, supplementary Figure 2).The Monte Carlo model is able

of

to reproduce the mixing of input courses, where the Cross-correlation illustrates the relative measure of

ro

similarity between mixture distributions with a relatively high coefficient (0.86 +/- 0.001 in PDP and 0.93 +/- 0 in KDE) with a strong contribution of sample 36.5-45.5 m (supplementary Table 1, supplementary

-p

Figure 2). Both K-S test D values and Kuiper test V values result in 0.1 +/- 0 and 0.058 +/- 0.001, respectively, with a very low contribution of sample 84.5-94.6 m. Identical results are obtained by

re

running an optimisation (Iterative optimization unmixing model; R2 = 0.86; v test 0.1 and D test 0.058;

lP

supplementary Table 1, supplementary Figure 2). Additional multidimensional scale illustration following the software Vermeesch (2013) resulted in a higher similarity of sample 36.5–45.5 m and 84.5–94.4 m

supplementary Figure 2).

na

with the central sample of 46.5-53.0 m in both KS and Kupier tests with very low stress levels (stress = 0;

ur

4.2. Palynological assemblages

In AM14, a total number of 244 different pollen morpho-taxa and 88 spore morpho-taxa were

Jo

identified (total 332). Out of the 66 samples prepared, only 22 contain enough material to count at least 200 pollen grains. Counts ranged from 212 (sample depth 7.5 m) to 442 (sample depth 21.1 m) identified pollen grains per sample. The number of identified pollen morpho-taxa show low diversity of 27 in sample depth of 18.4 m, 36 at 108.4 m, 40 at 163.5 m and 44 at 108.4 m, all others range from 49 to 80 taxa. Shannon H index supports diversity trends resulting in 2.3 (sample depth 18.4), 2.2 (sample depth 21.1 m), 2.8 (102.7 m) and 2.7 (104.7 m), while high values of Dominance are caused by high counts of Grimsdalea magnaclavata and Mauritiidites franciscoi (Fig. 3). Pollen make up to 29.9% of the whole palynological assemblages at a sample depth of 78.6 m, while 55.6% at a depth of 23.3 m. Nonpollen-palynomorphs show low values with the exception of high abundance of Azolla, which reach 24.0% in the depth 102.7 m, 8.9% at 108.4 m and 7.5% and 7.1% at depths of 96.4 m and 104.9 m, respectively. Foraminifera linings are present at certain horizons and reach a maximum of 1.8% in a 10

Journal Pre-proof sample depth of 102.7 m as well as 1.5% at 11.9 m, 1.5% at 118.1 m (Fig. 3; supplementary Table 2). The most abundant fern spores Deltoidospora adriennsis and Laevigatosporites tibuensis commonly make up more than 10% of the whole pollen and spore assemblage, followed by Magnastiatites grandiosus, Polypodiisporites usmensis and various other Polypodiisporites spp. and Psilatriletes spp. Strongly varying is the number of the palynomorpho-taxa Grimsdalea magnaclavata and Monoporopollenites annulatus (Fig. 3), which occur in every sample, but their number ranges from less than 0.3% up to 19.6% (G. magnaclavata) and 0.3% to 22.9% (M. annulatus). Mauritiidites franciscoi appears in almost every sample yet never tops 3.2% (Fig. 3). Echiperiporites stelae (0–1.3%), E. akanthos

of

(0–1.6%) and E. jutaiensis (0–2.2%, with higher values towards the top) also show a low but regular occurrence, similar to the Bombacacidites morpho-species B. nacimientoenis (0–9.5%), B. brevis (0–

ro

2.6%) and B. fossulatus (0–1.9%). Furthermore, different palynomorpho-species associated with

-p

Arecaceae are very diverse and abundant, such as Arecipites regio, Psilamonocolpites medius, Psilamonocolpites rinconii and Retimonocolpites maximus (supplementary Table 2).

re

4.3. Palynolocial biozones

lP

The entire 300.6 m of the core contain the extinct palm Grimsdalea magnaclavata and frequent occurrences of Deltoidospora adriennsis, Crassoretitriletes vanraadshovenii and Mauritiidites franciscoi.

na

The numbers of Monoporopollenites annulatus are increasing towards the top of the core (Fig. 4) and Echitricolporites maristellae and Crototricolpites annemariae are present but rare in most samples.

ur

Several Asteraceae are recorded, e.g. Echitricolporites spinosus appears in sample depth of 34.5 m and Fenestrites spinosus occurs regularly already in a depth below a depth of 100 m (Fig. 4). Despite the

Jo

presence of these Asteraceae species, Grimsdalea magnaclavata zone (late middle to late Miocene) according to Lorente (1986) and Hoorn (1993) appears most likely in the bottommost section of AM14 considering all detected markers mentioned above. The lowest occurrence of Ladakhipollenites? caribbiensis (previously Psilatricolporites caribbiensis) in a depth of 181.1 m assigns all sediments above as Psilatricolporites caribbiensis zone (late Miocene to early Pliocene) according to Lorente (1986) (see also Latrubesse et al., 2010). Due to the presence of Fenestrites spinosus, Corsinipollenites oculusnoctis, Perisyncolporites pokornyi, Monoporopollenites annulatus and Magnastriatites grandiosus, a lower position within the P. caribbiensis zone is assumed. Since no pollen grains of Echitricolporites mcneillyi or Alnipollenites verus were found, an age older than Pleistocene is presumable. 4.4. Calcareous microfossil assemblages

11

Journal Pre-proof In total, 23 ostracod and two foraminifera species have been identified, showing a high dominance of the genus Cyprideis (16 species) in abundance and diversity in all samples (supplementary Table 3). Except some poorly preserved juvenile Cyprideis valves and Ammonia tests (sample at 264.6 m depth), the basal core interval up to 181.8 m was barren, followed by samples from 179.2–176.0 m depth showing low diversity (up to six Cyprideis species). Aside one Cyprideis sulcosigmoidalis valve (at 138.7 m depth), preservation remains poor between 172.4–130.3 m depth recording mainly indeterminable ostracods at 130.3 m. The samples from 125.7–108.4 m depth contained five species of Cyprideis in low to moderate quantity and very rare tests of Ammonia (125.7, 121.8, 111.7 m). At 106.3

of

m depth, the most diverse ostracod fauna of AM14 was recovered with 11 Cyprideis species along with Cytheridella danielopoli, Alicenula olivencae, Perissocytheridea acuminata, P. ornellasae,

ro

Rhadinocytherura amazonensis and Skopaeocythere tetrakanthos. In contrast, at 105.4 m depth

-p

microfossils are rare, while samples at 104.9–99.5 m depth have low to moderately diverse Cyprideisdominated assemblages (up to 11 Cyprideis species) associated with P. acuminata and P. ornellasae. At

re

sample depth 102.7 m, this spectrum is supplemented by the occurrence of A. olivencae and very rare Ammonia. The samples from 96.4–78.6 m depth comprise rare and poorly preserved Cyprideis valves,

lP

accompanied at 88.6 m depth by A. olivencae and very rare Ammonia tests. At 77.6 m, abundant but very poorly preserved ostracods (four Cyprideis species and C. danielopoli) have been found. The core

na

interval between 75.3–39.0 m depths contained low diverse Cyprideis associations or was barren (64.2– 39.0 m). Samples from 36.4 and 34.6 m depth yielded very abundant but almost monospecific Cyprideis

ur

amazonica ostracod faunas, complemented by Cyprideis olivencai and A. olivencae (both samples), Cyprideis munoztorresi (at 36.4 m) and Cypria sp. (at 34.6 m). Much less and only poorly preserved C.

Jo

amazonica valves were recovered at 34.9 m sample depth. Aside few Cyprideis valves (at 32.0 and 11.9 m depth) and very rare Elphidium tests (11.9 m depth), the samples from 32.0–7.5 m are barren of ostracods and foraminifers. 4.5. Ostracod biozones Muñoz-Torres et al. (2006) proposed five ostracod biozones for the Pebas and Solimões formations. This biozonation is based on the recovered ostracod assemblages and the stratigraphic position (height in the section) of the studied samples. Recently, this biozones have been modified by Linhares et al. (2019) in accordance with earlier taxonomic works (Gross et al. 2013, 2014, 2016). Nevertheless, Linhares et al. (2019) used the same appearances/disappearances of a few Cyprideis index species as Muñoz-Torres et al. (2006) to characterise ostracod zones (note: the “first and last appearance datum” of index species as used in Linhares et al., 2019 are in fact first/lowest and 12

Journal Pre-proof last/highest occurrences). In addition, one new ostracod zone (Cyprideis paralela zone) has been introduced and one (Cyprideis obliquosulcata zone) abandoned by Linhares et al. (2019). The ostracod biozonation of Medeiros et al. (2019) (2017), established on core AM33 (Fig. 1B-C), seems only of local importance and is not applicable for regional purposes (Linhares et al., 2019). Among the ostracods in AM14, Cyprideis amazonica, C. machadoi, C. multiradiata, C. olivencai, C. sulcosigmoidalis, Cytheridella danielopoli, Alicenula olivencae, Perissocytheridea acuminata and Rhadinocytherura amazonensis range through all ostracod zones (Linhares et al., 2019). The index species C. minipunctata has its highest occurrence in AM14 at 176.0 m depth (Fig. 4) suggesting C.

of

minipunctata zone for the bottom of the core. Associated with the lowest occurrence of C. minipunctata

ro

are C. machadoi, C. multiradiata, C. sulcosigmoidalis and C. paralela. Except the latter one, which seems to appear in the C. caraionae zone, the other three species range down to the C. sulcosigmoidalis zone

-p

(Linhares et al., 2019). Therefore the lower limit of the C. minipunctata zone cannot be defined in AM14. The index species C. cyrtoma has its lowest occurrence at 125.7 m depth. According to Muñoz-Torres et

re

al. (2006), C. cyrtoma appears in the C. obliquosulcata zone, a zone which was not recovered by

lP

subsequent works (e.g. Gross et al., 2014; Linhares et al., 2019) and dropped in Linhares et al. (2019). The lower boundary of the C. cyrtoma zone is defined by the last appearance of C. minipunctata (Linhares et al., 2019), which has its last occurrence at 176 m depth in AM14 where the limit between

na

both zones is tentatively set here. C. cyrtoma has its highest occurrence at 77.6 m depth. The last appearance of C. cyrtoma defines the top of C. cyrtoma zone and the last appearance of C. paralela the

ur

top of the C. paralela zone (Muñoz-Torres et al., 2006; Linhares et al., 2019). Due to the last occurrence

defined in AM14.

Jo

of C. paralela together with C. cyrtoma at 77.6 m, the upper limit of the C. cyrtoma zone cannot be

5. Discussion 5.1. Provenance and river systems The age spectra high similarity of the three investigated zircon samples populations found in from AM14 core suggests a constant source area of sediments deposited in the Solimões Basin show similar components based on visual comparison (Fig. 2A-C; supplementary Figure 2), which might indicate a common source area during the sedimentation of the top ~95 m. Similarity statistics support this assumption with a high cross-correlation coefficient (Fig. 2H) between mixture distribution among samples and with inverse Monte Carlo simulations, as well as low values of the Kolmogorov-Smirnov test 13

Journal Pre-proof D statistic and Kuiper test V statistic in comparison for cumulative distribution functions. Further, the low stress levels of the MDS analysis shows an excellent fit among samples (supplementary Table 1; supplementary Figure 2). However, the statistical analysis should be interpreted with caution due to the low sample number and small sample size (Sundell and Saylor, 2017), which therefore does not allow the discussion of shifts in the sediment source area through time. Ages on Detrital ages on detrital zircon grains of Neogene deposits in the Brazilian Amazonia are currently limited to a few analyses on outcrops along the Solimões River and its tributaries (Mapes, 2009; Horbe et al., 2013; Bissaro et al., 2019). Modern river samples from the Solimões-Amazon River

of

outline general trends despite possible hydrodynamic fractionation fractioning (Mapes, 2009; Lawrence

ro

et al., 2011) and show that with increasing distance to the Andes, the higher sediment input from rivers of the cratonic mountains of the Brazilian and Guiana Shield (Fig. 1A) results in higher abundance of

-p

zircon grains older than 1,500 Ma (Mesoproterozoic and older) (Mapes, 2009). In the samples of AM14 core, such older zircon grains are rare or absent (Fig.2 A-D), hence Phanerozoic zircon grains dominate

re

the assemblages of the studied core. The results of AM14 are also similar to the top of the assemblage

lP

of the modern Solimões River around Leticia (western Solimões Basin, Fig. 1B) and differ from measurements taken further down the Solimões River in the central and eastern Solimões Basin between Tefé and Manaus (Mapes, 2009; Horbe et al., 2013; Fig. 1A). Considering the observed zircon

na

age spectra distribution of all studied samples (Fig. 2A-D), an Andean source area for the top ~95 m of the drill site AM14 is interpreted, with the Northern Volcanic Zone and Central Volcanic Zone as

ur

potential original sources. Both areas record a strong Miocene volcanic activity (Cordani et al., 2016).

Jo

Despite pronounced Andean volcanism during the late Miocene, Pliocene and Quaternary (Tibaldi et al., 2017), younger zircon grains are extremely rare in the modern sediments of the Brazilian lowlands (Mapes, 2009; Mason et al., 2019) and were not found in the top section of AM14. This Andean-sourced river drainage is further supported by the strontium isotopic analysis of well-preserved mollusk shells of the Solimões Formation down to a depth of 140 m (AM4A; Vonhof et al., 2003; Fig. 1B-C) and in outcrops of the Pebas Formation in Peru (Roddaz et al., 2006). Considering the provenance of the detrital zircons in the Neogene drilled deposits, a past river draining the Andes appears as the most likely source to transport the Phanerozoic zircons to the Solimões Basin. and the deposition of detrital zircon grains in other Neogene depositsa drainage configuration comparable to today is most likely for this time interval in lowland Amazonia. 5.2. Biostratigraphy and maximum ages of the Solimões Formation 14

Journal Pre-proof U/Pb maximum depositional ages on detrital zircon grains in AM14 display Tortonian (11.42 ± 0.66 Ma) or possibly younger ages constraints down to a depth of ~95 m. These are similar to previous age estimates in western Amazonia such as that derived from outcrop samples of the Solimões Formation (Mapes, 2009; Bissaro et al., 2019) and the Cocama ash from the Peruvian Madre de Dios Formation (9.01 ± 0.28 Ma; Campbell et al., 2001, Fig. 1A). Ages for a maximum deposition in AM14 agree with the rough age estimates of ostracod and palynological stratigraphy (Fig. 4), showing results closer to the oldest possible ages according to the U/Pb measurements. Considering the palynological assemblages, the determination of the Psilatricolporites caribbiensis zone according to Lorente (1986)

of

and consequently a late Miocene to Pliocene age is determined for the top 181.8 m of AM14. Below this depth, indicators of Asteraceae and/or Grimsdalea magnaclavata zone point towards deposition ages

ro

varying between late middle and late Miocene (Lorente, 1986; Hoorn, 1993). Ostracod biozonations

-p

results in similar uncertainties; from the uppermost samples down to 176.0 m, Cyprideis cyrtoma zone ranges in the late Miocene (Linhares et al., 2019) but could stretch into the late middle Miocene

re

(Muñoz-Torres et al., 2006) while ostracod age estimates of the Cyprideis minipunctata zone below suggest a deposition between late middle to late Miocene (Muñoz-Torres et al., 2006; Linhares et al.,

lP

2019). This overall match in the biostratigraphies might derive from the primary correlation of ostracod

control in the Solimões Basin.

na

zones to the previously defined palynological zones, however, which both lack an independent age

The ostracod and palynological biostratigraphic age estimates are widespread in the Solimões Basin (e.g.

ur

Hoorn, 1993; Silva-Caminha et al., 2010; Leite et al., 2017; Jaramillo et al., 2017; Linhares et al., 2017,

Jo

2019; Leandro et al., 2019; Fig. 1B-C). The succession of biozones remains conclusive within each core, yet show ambiguous age estimates, particularly for the deeper sequences of the studied records. These can range in similar depth from early to late Miocene (e.g. Leite et al., 2017; Linhares et al., 2019) or completely disagree (Medeiros et al., 2019). For the top of most records, a late Miocene to Pliocene age is suggested, which support are in agreement with the reconstructed maximum deposition ages in this study (Fig. 2, Fig. 4). Discrepancies among previous studies, however, occur for age estimates of important events in the evolution of Amazonia such as marine incursions (Hoorn, 1993; Jaramillo et al., 2017; Linhares et al., 2017, 2019) and the origin of the Amazon River (Hoorn, 1993; Leite et al., 2017). With the new age data from detrital zircons of this core, we can support that rivers draining the Andes reached the Solimões Basin at and possibly since the last 11.4 Ma. While no single or multiple event(s) of marine transgression is evident in the lithological, palynological and ostracodological data of AM14, it records saline water bodies during younger sediments than previously documented (Hoorn, 1993; 15

Journal Pre-proof Jaramillo et al., 2017). Therefore, the addition of the maximum age to the younger biozones in the Solimões Basin represents a strong improvement in the local biostratigraphy. Age ranges of each palynological and ostracod zone still remain large, and thus require further verification by independent dating particularly in deeper layers of the Solimões Formation. 5.3. Paleoenvironmental indications and contradictions Throughout the core AM14, the palynological record shows diverse vegetation in resemblance to the modern Amazonian rainforest. Despite the low level of botanical identifications of Neogene

of

palynomorpho-taxa of approximately 10% on genus level and up to 15% on family level (Jaramillo et al., 2010), the vegetation comprises elements of the recent terra firme forest (upland forest not exposed to

ro

seasonal flooding) next to a zonal vegetation of wetlands including flooded forests (várzea and igapó)

-p

and swamps (Table 1). Along the core, these two dominating vegetation types alternate; terra firme rainforest elements have a stronger signal at the bottom and the top of the record, but decline between

re

approx. 120 m and 70 m where the swamp/floodplain palms Mauritia flexuosa (Table 1; Fig. 3) and Grimsdalea magnaclavata dominate (e.g. Hoorn, 1993; Rull, 1998; Pocknall and Jarzen, 2012) next to a

lP

high abundance of the aquatic fern Azolla (Fig. 3). Azolla commonly grows today in small ponds, ditches or other still water environments (Lumpkin and Plucknett, 1980) and is reported in other cores of the

na

Solimões Basin (Hoorn, 1993). Although Azolla is normally associated with freshwater conditions, in several samples foraminifera linings co-occur (sample depth 11.9, 14.5, 96.4, 102.7, 104.9, 118.1, 118.1,

ur

121.8, 205.0 m; Fig. 3). Laboratory experiments have shown a maximum salinity tolerance of Azolla up to be 13 psu (Lumpkin and Plucknett, 1980) and thus brackish conditions with potential foraminifer

Jo

coexistence could be possible. Additional indicators such as mangrove pollen (Zonocosites ramonae) or dinoflagellate cysts are ambiguous in the AM14 record as the identified palynomorphs are badly preserved. Within the foraminifera record of AM14, Elphidium (11.9 m) as well as Ammonia are present (86.8, 102.7, 111.7, 121.8, 125.7, 264.6 m), which are known to colonize also oligohaline habitats like lagoons or mangroves (e.g. Boonstra et al., 2015; Bright et al., 2018). The formation of calcified foraminiferan tests requires at least low salinity regimes (e.g. review of Iglikowska and Pawłowska, 2015). Such phases of raised salinity in the Pebas system were repeatedly explained by marine incursions (e.g. Hoorn, 1993; Boonstra et al., 2015; Jaramillo et al., 2017) although evaporation processes have been inferred as well (Whatley et al., 1998). Cyprideis strongly dominates the ostracod faunas; this genus is today common in shallow, oligo- to mesohaline waters (e.g. lagoons, estuaries, marshes, lakes) and tolerates fluctuating salinity (freshwater to hyperhaline) and oxygenation (Gross et 16

Journal Pre-proof al., 2013). Another euryhaline, brackish water taxon (e.g. lagoons, lakes) is Perissocytheridea (Nogueira and Ramos, 2016), which co-occurs with Rhadinocytherura and Skopaeocythere in the middle part of AM14 core (106.3–99.5 m depth). The latter are supposedly brackish or marine genera (Sheppard and Bate, 1980; Whatley et al., 2000) but are exclusively known from western Amazonia (Gross et al., 2013, 2016; Gross and Piller, 2018) with no modern analog for ecological references. Cytheridella, Alicenula and Cypria are freshwater ostracods and have been found in the middle (106.3–66.8 m) and upper (36.4–34.6 m) part of this core (Fig. 4). Cytheridella dwells shallow, perennial, freshwater to slightly saline (<3 psu) settings (e.g. abandoned channels, lakes, marshes, lagoons; Gross et al., 2013; Wrozyna

of

et al., 2018). Recent Alicenula and Cypria regularly settle freshwater environments like ponds, lakes and slow flowing rivers but also occur in oligo- to mesohaline coastal swamps (e.g. Keyser, 1976; Higuti et al.,

ro

2009; Karanovic, 2012). Thus, we interpret Cyprideis-dominated faunas related to stressful, short-lived

-p

possibly ephemeral waterbodies, strongly affected by seasonally fluctuating river dynamics (e.g. flooding, desiccation; Kaandorp et al., 2006; Gross et al., 2013). Shallow unsteady waterbodies (lakes,

re

abandoned channels, marshes) were present and conditions varied between freshwater to slightly elevated salinity. In contrast, the diverse and abundant ostracod fauna at 106.3 m depth with Cyprideis,

lP

Cytheridella, Alicenula, Perissocytheridea, Rhadinocytherura and Skopaeocythere (without foraminifers) can be related to a stable freshwater lake. Thus, between 120 and 80 m depth, diverse freshwater

na

environments alternate with slightly saline ponds co-existed in close proximity. The contradicting presence of marine and freshwater taxa in identical samples (Fig. 3) could otherwise be explained by

ur

reworking of older Neogene sediments. However, the thin walled foraminifera in the microfossil assemblages and foraminifera linings in palynological samples are still well-preserved and comparable to

Jo

other older records of marine incursions (e.g. Jaramillo et al., 2017). Questions on the existence of marine connections and their ages within the Pebas mega-wetland were previously raised (e.g. Campbell et al., 2006; Latrubesse et al., 2010; Boonstra et al., 2015). Considering the existence of the Pebas megawetland through time, Wesselingh et al. (2002) described aquatic systems dominated by freshwater conditions, interrupted possibly only by short phases of marine incursions, which created marginal marine conditions during the middle to early late Miocene (Wesselingh and Ramos, 2010). Salinity was raised to 3–5 psu (Wesselingh et al., 2002; 2006), although others support lower values of only ~1 psu (Vonhof et al., 2003). Both phases are indicated by Wesselingh et al. (2002) to range within the Crassoretitriletes vanraadshoovenii (Jaramillo et al., 2011) or Grimsdalea magnaclavata zones (Lorente, 1986; Hoorn, 1993), approximately from 11.5–10.2 Ma (Wesselingh et al., 2006). This is in agreement with maximum deposition ages down to a depth of 94.5 m of approx. 11.4 Ma, above which 17

Journal Pre-proof foraminifera are still present (Fig. 4). Yet, this contradicts the most recent palynological-based ages for marine transgressions to the Solimões Basin of 18.0–17.8 Ma and 14.1–13.7 Ma (Jaramillo et al., 2017, drilling 1AS-105-AM in Fig. 1B-C). Despite the presence of foraminifera indicated elevated salinities within the Solimões Basin during the Tortonian or younger, a connection to the Caribbean appears unlikely due to the previous formation uplift of the Vaupes Swell as a barrier (Mora et al., 2010; Jaramillo et al., 2017, Fig. 1A). This, however, does not exclude the existence and possibility of persisting slightly saline waterbodies within the basin. Considering the mixture of marine and freshwater taxa in one sample and the alternating environmental conditions between samples, a landscape with diverse

of

habitats co-occurring in close proximity is indicated. The driving mechanism of changes in the landscape presumably are related to changes in the river patterns of avulsive rivers (e.g. Latrubesse et al., 2010;

ro

Gross et al., 2011) during a tropical climate with high seasonality (Kaandrop et al., 2005). The occurrence

-p

of terra firme forest elements accounts for the presence of elevated non-flooded soils, showing a potential similarity with river dynamics described by Pupim et al. (2019) for Quaternary times prior to

lP

necessary to confirm such landscape changes.

re

phases of strong river incision. However, higher resolution studies and a more precise age model are

na

6. Conclusions

Herein, we present the first U/Pb ages of detrital zircon grains from a core drilled in the western

ur

Brazilian Amazon showing that the topmost ~95 m sediments of the Solimões Formation have a maximum deposition age of 11.42 ± 0.66 Ma and are deposited during Tortonian (late Miocene) or

Jo

younger times in the Solimões Basin. Palynological and ostracod biostratigraphic zones reveal age estimates close to the maximum deposition age with less precise determinations age estimates of the Psilatricolporites caribbiensis zone (late Miocene to Pliocene) and the Cyprideis cyrtoma zone (late Miocene), which both continue towards depths of 181.8 m and 176.0 m, respectively. For the bottom portion of the core, no zircons sample could be obtained, resulting in, a broad age range from the late middle Miocene to late Miocene is as suggested by both biozonations (Grimsdalea magnaclavata and Cyprideis minipunctata zones). In comparison with previous records, a late Miocene age for the top of the Solimões Formation is favored, but disagreement uncertainties among both biostratigraphic concepts are is evident in greater depth. A possible diachronous character of the Solimões Formation in greater depth and thus a more complex depositional history of this extensive formation could explain to

18

Journal Pre-proof the age variation; however, further age calibration of the established biostratigraphic zonations in a greater depth is needed. In addition, Phanerozoic ages of detrital zircon grains from form the top ~95 m confirm indicate an Andean source for the Solimões Formation sediments accumulated since the late Miocene in this area, which suggest a possible drainage pattern comparable to the modern Amazon River drainage. Fluvialdominated landscapes are further indicated by the palynological and ostracod assemblages; while terra firme rainforest and rivers dominated the landscape associated with short-lived lakes (possibly seasonal; e.g. flooding, desiccation) during the bottom phase of the core. The upper section is characterized by

of

newly formed aquatic habitats such as abandoned channels indicated by mass-occurrences of pioneer

ro

species (~40–30 m, Cyprideis amazonica). Following the palynological data, in the middle part of the core (between ~120–80 m depths) ostracods show calm waters like a shallow lake interchanging with

-p

unstable slightly saline ponds. Foraminifers (calcareous tests and inner linings) reach up to the topmost sample of 7.6 m depth, thus range within the Tortonian or younger as indicated by the ages measured

re

on zircon grains. These are the first independent age constraints for the palynological and ostracod

lP

biozones applied throughout the Solimões Basin, which verify an the occurrence of marine-derived environments characterized by slightly elevated salinity within the Solimões Basin in the late Miocene, thus extending age limits proposed by Jaramillo et al. (2017) to younger times. and therefore question

na

previous age estimates throughout the basin. Further, the The existence of marine indicators distributed along the core suggests a diverse landscape with possibly coexistence of freshwater and slightly saline

ur

habitats, instead of strong inland marine transgressions during this phase. Such waterbodies which

Jo

were often short-lived expressing unstable environments. Instead of a gradual landscape evolution throughout the Neogene late Miocene marked by events, our data suggest high levels of adaption to changing conditions and therefore a more dynamic landscape with shifting between rivers, wetlands and small freshwater or slightly saline lakes. possibly similar to the Quaternary.

Acknowledgements This research had the main financial support of the project “Dimensions US-BIOTA-Sao Paulo: Assembly and evolution of the Amazonian biota and its environment: an integrated approach”, a collaborative Dimensions of Biodiversity BIOTA grant supported by grant 2012/50260-6, Sao Paulo Research Foundation (FAPESP, Brazil), National Science Foundation (NSF, United States) and NASA (United States). 19

Journal Pre-proof Additional FAPESP grands supporting this research were grants 2014/05582-0, 2014/23334-4, 2017/06874-3, the Brazilian Foundation for Support and Evaluation of Graduate Education (CAPES) 88887.370034/2019-00 and the National Council for Scientific and Technological Development (CNPq, Brazil) grants #302411/2018-6, #3009223/2014-8 respectively. Additional funding was provided by the Austrian Science Fund project P21748-N21. Special thanks goes to cooperating the DNPM/Manaus, (especially Gert Woeltje) and CPRM/Manaus, (especially Marco Antônio de Oliveira) for sampling permits. This work initial help in pollen identification by Carlos Jaramillo and Ingrid Romero at STRI in Panama City. We further thank Cody C. Mason and an anonymous reviewer for their great comments

ro

of

and corrections.

-p

References

Albert, J.S., Val, P., Hoorn, C., 2018. The changing course of the Amazon River in the Neogene: center

re

stage for Neotropical diversification. Neotrop. Ichthyol. 16, e180033. DOI: 10.1590/1982-022420180033

lP

Antonelli, A., Zizka, A., Carvalho, F.A., Scharn, R., Bacon, C.D., Silvestro, D., Condamine, F.L., 2018. Amazonia is the primary source of Neotropical biodiversity. Proc. Natl. Acad. Sci. USA, 115, 6034–

na

6039. DOI: 10.1073/pnas.1713819115

Antonelli, A., Sanmartin, I., 2011. Why are there so many plant species in the Neotropics? Taxon 60,

ur

403–414. DOI: 10.1002/tax.602010

Bernal, R., Bacon, C.D., Balslev, H.,Hoorn, C., Bourlat,S.J., Tuomisto, H., Salamanca, S., van Manen, m.T.,

Jo

Romero, I., Sepulchre, P., Antonelli, A., 2019. Could coastal plants in western Amazonia be relicts of past marine incursions? J. Biogeogr. 46, 1749– 1759. DOI:10.1111/jbi.13560 Bissaro Jr., M.C., Kerber, L., Crowley J.L., Ribeiro, A.M., Ghilardi, R.P., Guilherme, E., Negri, F.R., SouzaFilho, J.P., Hsiou, A.S., 2019. Detrital zircon U–Pb geochronology constrains the age of Brazilian Neogene deposits from Western Amazonia. Palaeogeogr. Palaeoclimatol. Palaeoecol. 516, 64–70. DOI: 10.1016/j.palaeo.2018.11.032 Boonstra, M., Ramos, M. I. F., Lammertsma, E. I., Antoine, P.-O., Hoorn, C., 2015. Marine connections of Amazonia: Evidence from foraminifera and dinoflagellate cysts (early to middle Miocene, Colombia/Peru). Palaeogeogr. Palaeoclimatol. Palaeoecol. 417, 176–194. DOI: 10.1016/j.palaeo.2014.10.032

20

Journal Pre-proof Bright, J., Cohen, A.S., Starratt, S. W., 2018. Distinguishing brackish lacustrine from brackish marine deposits in the stratigraphic record: A case study from the late Miocene and early Pliocene Bouse Formation, Arizona and California, USA. Earth Sci. Rev., 185, 974–1003. DOI: 10.1016/j.earscirev.2018.08.011 Campbell Jr., K.E., Frailey, C.D., Romero-Pittman, L., 2006. The Pan-Amazonian Ucayali Peneplain, late Neogene sedimentation in Amazonia, and the birth of the modern Amazon River system. Palaeogeogr. Palaeoclimatol. Palaeoecol. 239, 166–219. DOI: 10.1016/j.palaeo.2006.01.020 Campbell, K.E., Heizler, M., Frailey, C.D., Romero-P, L., Prothero, D.R., 2001. Upper Cenozoic

of

chronostratigraphy of the southwestern Amazon Basin. Geology 29, 595–598. DOI: 10.1130/0091-

ro

7613(2001)029<0595:UCCOTS>2.0.CO;2

-p

Caputo, M.V., Rodrigues, R., Vasconcelos, D.N.N., 1971. Litoestratigrafi a da Bacia do Amazonas. Relatório Interno Técnico, 641-A. Belém: Petrobras.

re

Cordani, U.G., Ramos, V.A., Fraga, L.M., Cegarra, M., Delago, I., Souza, K.G. de, Gomes, F.E.M.,

lP

Schobbenhaus, C., 2016. Tectonic map of South America: 1:5.000.000 scale. CGMW-CPRMSEGEMAR, Second Edition.

na

Cozzoul, M., 2006. The Acre vertebrate fauna: diversity, geography and time. J. South Am. Earth Sci. 21, 185–203. DOI: 10.1016/j.jsames.2006.03.005

ur

Dino, R., Silva, O.B., Abrahão, D., 1999. Caracterização palinológica e estratigráfica de estratos cretáceos

Jo

da Formação Alter do Chão, Bacia do Amazonas. In: Boletim do 5º Simpósio sobre o Cretáceo do Brasil. Campus de Rio Claro, UNESP, pp. 557–565. Espurt, N., Baby, P., Brusset, S., Roddaz, M., Hermoza, W., Regard, V., Antoine, P.-O., Salas Gismondi, R., Bolños, R., 2007. How does the Nazca Ridge subduction influence the modern Amazonian foreland basin? Geology 35, 515–518. DOI: 10.1130/G23237A.1 Faegri, K., Iversen, J., 1975. Textbook of Pollen Analysis. New York, Hafner Press, 295 p. Figueiredo, J., Hoorn, C., van der Ven, P., Soares, E., 2009. Late Miocene onset of the Amazon River and the Amazon deep-sea fan: Evidence from the Foz do Amazonas Basin. Geology 37, 619–622. DOI: 10.1130/G25567A.1

21

Journal Pre-proof Garzione, C.N., Molnar, P., Libarkin, J.C., MacFadden, B.J., 2006.Rapid late Miocene rise of the Bolivian Altiplano: Evidence for removal of mantle lithosphere. Earth Planet. Sci. Lett. 241, 543–556. DOI: 10.1016/j.epsl.2005.11.026 Garzione, C.N., Hoke, G.D., Libarkin, J.C., Withers, S., MacFadden, B., Eiler, J., Ghosh, R., Mulch, A., 2008. Rise of the Andes. Science 320, 1304–1307. DOI: 10.1126/science.1148615 Garzione, C.N., Auerbach, D.J., Smith, J.J., Rosario, J.J., Passey, B.H., Jordan, T.E., Eiler, J.M., 2014. Clumped isotope evidence for diachronous surface cooling of the Altiplano and pulsed surface uplift of the Central Andes. Earth Planet. Sci. Lett. 393, 173–181. DOI: 10.1016/j.epsl.2014.02.029

of

Germeraad, J.H., Hopping, C.A., Muller, J., 1968. Palynology of Tertiary sediments from tropical areas.

ro

Rev. Palaeobot. Palyno. 6, 189–348. DOI: 10.1016/0034-6667(68)90051-1

-p

Gingras, M.K., Räsänen, M.E., Ranzi, A., 2002. The significance of bioturbated inclined heterolithic stratification in the southern part of the Miocene Solimões Formation, Rio Acre: Amazonia Brazil.

re

Palaios 17, 591–601. DOI: 10.1669/0883-1351(2002)017<0591:TSOBIH>2.0.CO2

lP

Ghosh, P., Garzione, C.N., Eiler, J.M., 2006. Rapid uplift of the Altiplano revealed through 13C-18O bonds in paleosol carbonates. Science 311, 511–515. DOI: 10.1126/science.1119365

na

Gross, M., Piller, W.E., Ramos, M.I., Paz, J.D.S., 2011. Late Miocene sedimentary environments in southwestern Amazonia (Solimões formation; Brazil). J. South Am. Earth Sci. 32, 169–181. DOI:

ur

doi:10.1016/j.jsames.2011.05.004

Gross, M., Ramos, M.I., Caporaletti, M., Piller, W.E., 2013. Ostracods (Crustacea) and their

Jo

palaeoenvironmental implication for the Solimões Formation (Late Miocene; Western Amazonia/Brazil). J. South Am. Earth Sci. 42, 216–241. DOI: 10.1016/j.jsames.2012.10.002 Gross, M., Ramos, M.I.F., Piller, W.E., 2014. On the Miocene Cyprideis species flock (Ostracoda; Crustacea) of Western Amazonia (Solimões Formation): Refining taxonomy on species level. Zootaxa 3899, 1–69. DOI: 10.11646/zootaxa.3899.1.1 Gross, M., Ramos, M.I.F., Piller, W.E., 2016. A minute ostracod (Crustacea: Cytheromatidae) from the Miocene Solimões Formation (western Amazonia, Brazil): evidence for marine incursions? J. Syst. Palaeontol. 14, 581–602. DOI: 10.1080/14772019.2015.1078850 Gross, M., Piller, W.E., 2018. Fossil glochidia (Bivalvia: Unionida: Hyriidae) from the middle Miocene of western Amazonia (Peru). J. Syst. Palaeontol. DOI: 10.1080/14772019.2018.1498034. DOI: 10.1590/S1519-69842009000200004 22

Journal Pre-proof Higuti, J., Lansac-Tôha, F.A., Velho, L.F.M., Pinto, R.L., Vieira, L.C.G., Martens, K., 2009. Composition and distribution of Darwinulidae (Crustacea, Ostracoda) in the alluvial valley of the upper Paraná River, Brazil. Brazilian Journal of Biology 69, 253–262. DOI: 10.1590/S1519-69842009000200004 Hoorn, C., 1993. Marine incursions and the influence of Andean tectonics on the Miocene depositional history of northwestern Amazonia: results of a palynostratigraphic study. Palaeogeogr. Palaeoclimatol. Palaeoecol. 105, 267–309. DOI: 10.1016/0031-0182(93)90087-Y Hoorn, C., 1994a. Fluvial palaeoenvironments in the intracratonic Amazonas Basin (Early Miocene-early Middle Miocene, Colombia). Palaeogeogr. Palaeoclimatol. Palaeoecol. 109, 1–54. DOI:

of

10.1016/0031-0182(94)90117-1 Hoorn, C., 1994b. An environmental reconstruction of the palaeo-Amazon River system (Middle–Late

ro

Miocene, NW Amazonia). Palaeogeogr. Palaeoclimatol. Palaeoecol. 112, 187–238. DOI:

-p

10.1016/0031-0182(94)90074-4

Hoorn, C., Guerrero, J., Sarmiento, G.A., Lorente, M.A., 1995. Andean tectonics as a cause for changing

7613(1995)023<0237:ATAACF>2.3.CO;2

re

drainage patterns in Miocene northern South America. Geology, 237–240. DOI: 10.1130/0091-

lP

Hoorn, C., Wesselingh, F.P., ter Steege, H., Bermudez, M.A., Mora, A., Sevink, J., Sanmartin, I., SanchezMeseguer, A., Anderson, C.L., Figueiredo, J.P., Jaramillo, C., Riff, D., Negri, F.R., Hooghiemstra, H.,

na

Lundberg, J., Stadler, T., Sarkinen, T., Antonelli, A., 2010. Amazonia through time: Andean uplift, climate change, landscape evolution, and biodiversity. Science 330, 927–931. DOI:

ur

10.1126/science.1194585

Jo

Horbe, A.M.C., Motta, M.B., de Almeida, C.M., Dantas, E.L., Vieira, L.C., 2013. Provenance of Pliocene and recent sedimentary deposits in western Amazonia, Brazil: consequences for the paleodrainage of the Solimões-Amazonas River. Sediment. Geol. 296, 9–20. DOI: 10.1016/j.sedgeo.2013.07.007 Horton, B.K., Parra, M., Salor, J.E., Nie, J., Mora, A., Torres, V., Stockli, D.F., Strecker, M.R., 2010. Resolving uplift of the northern Andes using detrital zircon age signatures. GSA Today 20, 7. Hovikoski, J., Räsänen, M.E., Gingras, M., Roddaz, M.,Brusset, S., Hermoza, W., Romero-Pittman, L., 2005. Miocene semidiurnal tidal rhythmites in Madre de Dios, Peru. Geology 33, 177–180. DOI: 10.1130/G22997.1 Hovikoski, J., Räsänen, M., Gingras, M., Lopez, S., Ranzi, A., Melo, J., 2007. Palaeogeographical implications of the Miocene Quendeque Formation (Bolivia) and tidally-influenced strata in the southwestern Amazonia. Palaeogeogr. Palaeoclimatol. Palaeoecol. 243, 23–41. DOI: 10.1016/j.palaeo.2006.07.013 23

Journal Pre-proof Iglikowska, A., Pawłowska, J., 2015. The Adaptations of the Foraminifera and Ostracoda to Fresh Water Colonisation, pp. 91–113. In: Zielinski, T., Weslawski., M., Kulioski, K. (Eds.) Impact of Climate Changes on Marine Environments. GeoPlanet: Earth and Planetary Sciences. Springer, Switzerland. Jaramillo, C.A., Dilcher, D.L., 2001. Middle Paleogene palynology of Central Colombia, South America: A study of pollen and spores from tropical latitudes. Palaeontogr. Abt. B 258, 87–213. DOI: 10.1127/palb/258/2001/87 Jaramillo, C., Hoorn, C., Silva, S., Leite, F., Herrera, F., Quiroz, L., Dino, R., Antonioli, L., 2010. The origins of the modern Amazon rainforest: Implications from the palynological and paleobotanical record.

of

In: Hoorn, M.C., Wesselingh, F.P. (Eds.) Amazonia, Landscape and Species Evolution. Wiley-

ro

Blackwell, Oxford, pp. 317–334.

-p

Jaramillo, C.A., Rueda, M.J., Torres, V., 2011. A palynological zonation for the Cenozoic of the Llanos and Llanos Foothills of Colombia. Palynology 35, 46–84. DOI: 10.1080/01916122.2010.515069

re

Jaramillo, C., Romero, I., D’Apolito, C., Bayona, G., Duarte, E., Louwye, S., Escobar, J., Luque, J., Carrillo-

lP

Briceño, J.D., Zapata, V., Mora, A., Schouten, S., Zavada, M., Harrington, G., Ortiz, J., Wesselingh, F.P. 2017. Miocene flooding events of western Amazonia. Sci. Ad. 3, 1–11. DOI: 10.1126/sciadv.160169

na

Kaandorp, R.J.G., Vonhof, H.B., Wesselingh, F.W., Romero Pittman, L., Kroon, D., van Hinte, J.E., 2005. Seasonal Amazonian rainfall variation in the Miocene Climate Optimum. Palaeogeogr.

ur

Palaeoclimatol. Palaeoecol. 221, 1–6. DOI: doi:10.1016/j.palaeo.2004.12.024

Jo

Kaandorp, R.J.G., Wesselingh, F.P., Vonhof, H.B., 2006. Ecological implications from geochemical records of Miocene Western Amazonian bivalves. J. South Am. Earth Sci., 21, 54–74. DOI: 10.1016/j.jsames.2005.07.012 Kachniasz, K.E., Silva-Caminha, S.A.F., 2016. Palinoestratigrafia da Formação Solimões: comparação entre bioestratigrafia tradicional e o método de Associações Unitárias. Rev. Bras. Palaontol. 19, 481–490. DOI: 10.4072/rbp.2016.3.12 Karanovic, I., 2012. Recent Freshwater Ostracods of the World. Crustacea, Ostracoda, Podocopida. Springer, Berlin-Heidelberg, 608 pp. Keyser, D., 1976. Zur Kenntnis der brackigen mangrovebewachsenen Weichböden Südwest-Floridas unter besonderer Berücksichtigung ihrer Ostracodenfauna. PhD thesis, University of Hamburg, 142 pp.

24

Journal Pre-proof Keyser, M., Roddaz, M., Bailleul, J., Antoine, P.-O., Adnet, S., Kim, J.H., van Soelen, E., Parra, F., Gérard, J., Caleron, Y., Garnaison, C., Sinninghe Damsté, J.S., Baby, P., 2014. Evidences for a Paleocene marine incursion in southern Amazonia (Madre de Dios Sub-Andean Zone, Peru). Palaeogeogr. Palaeoclimatol. Palaeoecol. 414, 451–471. DOI: 10.1016/j.palaeo.2014.09.027 Latrubesse, E., Cozzuol, M., Silva, S., Rigsby, C., Absy, M.L., Jaramillo, C., 2010. The Late Miocene paleogeography of the Amazon basin and the evolution of the Amazon River. Earth Sci. Rev. 99, 99– 124. DOI: 10.1016/j.earscirev.2010.02.005 Lawrence, R.L., Cox, R., Mapes, R.W., Coleman, D.S., 2011. Hydrodynamic fractionation of zircon age

of

populations. GSA Bulletin 123, 295–305. DOI: 10.1130/B30151.1

ro

Leandro, L.M., Viera, C.E.L., Santos, A., Fauth, G., 2019. Palynostratigraphy of two Neogene boreholes from the northwestern portion of the Solimões Basin. J. South Am. Earth Sci. 89, 211–218. DOI:

-p

10.1016/j.jsames.2018.11.016

Leite, F.P.R., Paz, J., do Carmo, D.A., Silva-Caminha, S.A., 2017. The effects of the inception of Amazonian

re

transcontinental drainage during the Neogene on the landscape and vegetation of the Solimões

lP

Basin, Brazil. Palynology 41, 412–422. DOI: 10.1080/01916122.2016.1236043 Linhares, A.P., Gaia, V.C.S., Ramos, M.I.F., 2017. The significance of marine microfossils for

na

paleoenvironmental reconstruction of the Solimões Formation (Miocene), western Amazonia, Brazil. J. South Am. Earth Sci. 79, 57–66. DOI: 10.1016/j.jsames.2017.07.007

ur

Linhares, A.P., Ramos, M.I.F., Gaia, V.C.S., Friaes, Y.S., 2019. Integrated biozonation based on palynology and ostracods from the Neogene of Solimões Basin, Brazil. J. South Am. Earth Sci. 91, 57–70. DOI:

Jo

10.1016/j.jsames.2019.01.015

Lorente, M., 1986. Palynology and Palynofacies of the Upper Tertiary in Venezuela, Berlin, J. Cramer, 222 pp.

Lumpkin, T.A., Plucknett, D.L., 1980. Azolla: Botany, Physiology, and Use as a Green Manure. Economi. Bot. 34, 111–153. Maia, R.G., Godoy, H.K., Yamaguti, H.S., Moura, P.A., Costa, F.S., Holanda, M.A., Costa, J.B.S., 1977. Projeto de carvão no Alto Solimões. Relatório Final. CPRM-DNPM, 137 pp. Malhi, Y., Roberts, J.T., Betts, R. A., Killeen, T.J., Li, W., Nobre, C.A., 2008. Climate Change, Deforestation, and the Fate of the Amazon. Science 319, 169–72. DOI: 10.1126/science.1146961

25

Journal Pre-proof Mapes, R.W., 2009, Past and present provenance of the Amazon River drainage basin [Ph.D. thesis]: University of North Carolina. Mason, C.C., Romans, B.W., Stockli, D.F., Mapes, R.W., Fildani, A., 2019. Detrital zircons reveal sea-level and hydroclimate controls on Amazon River to deep-sea fan sediment transfer. Geology, 47, 563– 567. DOIi: 10.1130/G45852.1 Medeiros, C.G., Carmo, D.A., Antonietto, L.S., Boush, L.E.P., 2019. The ostracods from Solimoes Formation, Brazil: an alternative biostratigraphic zonation for the Neogene of Amazonia. Revista Brasileira de Paleontologia, 22, 97–105. DOI: 10.4072/rbp.2019.2.02

of

Mora, A., Baby, P., Roddaz, M., Parra, M., Brusset, S., Hermazoa, W., Espurt, N., 2010. Tectonic history of the Andes and sub-Andean zones: implications for the development of the Amazon drainage basin.

ro

In: Hoorn, C., Wesselingh, F.P. (Eds.) Amazonia, Landscape and Species Evolution: A Look into the

-p

Past. Wiley-Blackwell, Oxford, 38–60.

Medeiros, C.G., 2017. Ostracodes da formação Solimões, Brasil: Contribuição à bioestratigrafia do

re

Neógeno da Amazônia. Master thesis, Universidade de Brasília, 394, 67 pp. Mulch, A., Uba, C.E., Strecker, M.R., Schoenberg, R., Chamberlain, C.P., 2010. Late Miocene climate

10.1016/j.epsl.2009.12.019

lP

variability and surface elevation in the central Andes. Earth Planet. Sci. Lett. 290, 173–182. DOI:

na

Muller, J., Di Giacomo, E., Van Erve, A., 1987. A palynologic zonation for the Cretaceous, Tertiary and

ur

Quaternary of Northern South America. A.A.S.P. Contribution Series 19, 7–76. Muñoz-Torres, F., Whatley, R., Harten, D. van, 1998. The endemic non-marine Miocene ostracod fauna

Jo

of the Upper Amazon Basin. Revista Española Micropaleontología 30, 89–105. Muñoz-Torres, F.A., Whatley, R.C., Harten, D. van, 2006. Miocene ostracod (Crustacea) biostratigraphy of the upper Amazon Basin and evolution of the genus Cyprideis. J. South Am. Earth Sci. 21, 75–86. DOI: 10.1016/j.jsames.2005.08.005 Nogueira, A.C.R., Silveira, R., Guimarães, J.T.F., 2013. Neogene-Quaternary sedimentary and paleovegetation history of the eastern Solimões Basin, central Amazon region. J. South Am. Earth Sci. 46, 89–99. DOI: 10.1016/j.jsames.2013.05.004 Nogueira, A.A.E., Ramos, M.I.F., 2016. The genus Perissocytheridea Stephenson, 1938 (Crustacea: Ostracoda) and evidence of brackish water facies along the Oligo-Miocene, Pirabas Formation, eastern Amazonia, Brazil. J. South Am. Earth Sci. 65, 101–121. DOI: 10.1016/j.jsames.2015.11.007

26

Journal Pre-proof Perez, N.D., Horton, B.K., 2014. Oligocene–Miocene deformational and depositional history of the Andean hinterland basin in the northern Altiplano plateau, southern Peru. Tectonics 33, 1819–47. DOI: 10.1002/2014TC003647 Pocknall, D.T., Jarzen, D.M., 2012. Grimsdalea magnaclavata Germeraad, Hopping & Muller: an enigmatic pollen type from the Neogene of northern South America. Palynology 36, 134–143. DOI: 10.1080/01916122.2011.642258 Pupim, F.N., Sawakuchi, A.O., Almeida, R.P., Ribas, C.C., Kern, A.K., Hartmann, G.A., Chiessi, C.M., Tamura, L.N., Minelo, T.D., Savian, J.F., Grohmann, C.H., Bertassoli Jr., D.J., Stern, A.G., Cruz, F.W.,

of

Cracraft, J., 2019. Chronology of Terra Firme formation in Amazonian lowlands reveals a dynamic Quaternary landscape. Quat. Sci. Rev. 210, 154–163. DOI: 10.1016/j.quascirev.2019.03.008

ro

Purper, I., 1979. Cenozoic Ostracodes of the Upper Amazon Basin, Brazil. Pesquisas, 12, 209–281.

-p

Räsänen, M.E., Salo, J.S., Kalliola, R.J., 1987. Fluvial perturbance in the western Amazon Basin: Regulation by long-term Sub-Andean tectonics. Science 238, 1398–1401. DOI:

re

10.1126/science.238.4832.1398

Räsänen, M., Linna, A., Irion, G., Rebata Hernani, L., Vargas Huaman, R., Wesselingh, F., 1998. Geología y

lP

geoformas de la zona de Iquitos. In: Kalliola, R., Flores Paitán, S. (Eds) Geoecología y desarollo Amazónico: estudio integrado en la zona de Iquito, Peru. Annales Turkuensis (AII) 114, 59–137.

na

Rebata H., L.A., Räsänen, M.E, Gingras, M.K., Vieira Jr., V., Barberi, M., Irion, G., 2006a. Sedimentology and ichnology of tide-influenced Late Miocene successions in western Amazonia: The gradational

ur

transition between the Pebas and Nauta formations. J. South Am. Earth Sci. 21, 96–119. DOI: 10.1016/j.jsames.2005.07.011

Jo

Rebata H., L.A., Gingras, M.K., Räsänen, M.E., Barberi, M., 2006b. Tidal-channel deposits on a delta plain from the Upper Miocene Nauta Formation, Marañón Foreland Sub-basin, Peru. Sedimentol. 53, 971–1013. DOI: 10.1111/j.1365-3091.2006.00795.x Ribas, C.C., Aleixo, A., Nogueira, A.C.R., Miyaki, C.Y., Cracraft, J., 2012. A palaeobiogeographic model for biotic diversification within Amazonia over the past three million years. Proc. R. Soc. B 279, 681– 689. DOI: 10.1098/rspb.2011.1120 Roddaz, M., Viers, J., Brusset, S., Baby, P., Boucayrand, C., Hérail, G., 2006. Controls on weathering and provenance in the Amazonian foreland basin: Insights from major and trace element geochemistry of Neogene Amazonian sediments. Chem. Geol. 226, 31– 65. DOI: 10.1016/j.chemgeo.2005.08.010

27

Journal Pre-proof Rossetti, D.F., Toledo, P.M., Góes, A.M., 2005. New geological framework for western Amazonia (Brazil) and implications for biogeography and evolution. Quaternary Res. 63, 78–89. DOI: 10.1016/j.yqres.2004.10.001 Rossetti, D.F., Cohen, M.C., Tatumi, S.H., Sawakuchi, A.O., Cremon, E.H., Mittani, J.C., Bertani, T.C., Munita, C.J.A.S., Tudela, D.R.G., Yee, M., Moya, G., 2015. Mid-Late Pleistocene OSL chronology in western Amazonia and implications for the transcontinental Amazon pathway. Sediment. Geol. 330, 1–15. DOI: 10.1016/j.sedgeo.2015.10.001 Rull, V., 1998. Biogeographical and evolutionary considerations of Mauritia (Arecaceae), based on

of

palynological evidence. Rev. Palaeobot. Palyno. 100, 109–122. DOI: 10.1016/S0034-6667(97)000602

ro

Rull, V., 2011. Neotropical biodiversity: timing and potential drivers. Trends Ecol. Evol. 26, 508–513. DOI:

-p

10.1016/j.tree.2011.05.011

Saylor, J.E., Horton, B.K., 2014. Nonuniform surface uplift of the Andean plateau revealed by deuterium

re

isotopes in Miocene volcanic glass from southern Peru. Earth Planet. Sci. Lett. 387, 120–131. DOI: 10.1016/j.epsl.2013.11.015

Peru. Palaeontology, 23, 97–124.

lP

Sheppard, L.M., Bate, R.H., 1980. Plio-Pleistocene ostracods from the Upper Amazon of Colombia and

na

Silva, S.A.F., 2004. Palinologia do Neógeno da Amazônia sul ocidental, Brasil. MSc thesis at the Instituto Nacional de Pesquisas da Amazônia (INPA), Manaus, Brazil.

ur

Silva-Caminha, S.A.F., Jaramillo, C., Absy, M.L., 2010. Neogene palynology of the Solimões Basin,

Jo

Brazilian Amazonia. Palaeontogr. Abt. B: Paleobot.-Palaeophyt. 283, 1–67. DOI: 10.1127/palb/284/2010/13

Smith, B.T., McCormack, J.E., Cuervo, A.M., Hickerson, M.J., Aleixo, A., Cadena, C.D., Perez-Emán, J., Burney, C.W., Xie, X., Harvey, M.G., Faircloth, B.C., Glenn, T.C., Derryberry, E.P., Prejean, J., Fields, S., Brumfield, R., 2014. The drivers of tropical speciation. Nature 515, 406–409. DOI: 10.1038/nature13687 Souza-Filho, J.P., Guilherme, E., 2015. A Paleontologia no Estado do Acre. In: Adamy, A. (Ed.), Geodiversidade Do Estado Do Acre. Porto Velho, pp. 147–158. Sundell, K.E., Saylor, J.E., 2017. Unmixing detrital geochronology age distributions. Geochem. Geophys. Geosy. 18, 2872–2886. doi:10.1002/2016GC006774.

28

Journal Pre-proof Tibaldi, A., Bonali, F.L., Carazzato, C., 2017. Structural control on volcanoes and magma paths from localto orogen-scale: The central Andes case. Tectonophysics 699, 16–41. DOI: 10.1016/j.tecto.2017.01.005 Vermeesch, P., 2013. Multi-sample comparison of detrital age distributions. Chem. Geol. 341, 140–146. DOI: 10.1016/j.chemgeo.2013.01.010 Vonhof, H.B., Wesselingh, F.P., Ganssen, G.M., 1998. Reconstruction of the Miocene western Amazonian aquatic system using molluscan isotopic signatures. Palaeogeogr. Palaeoclimatol. Palaeoecol. 141, 85–93. DOI: 10.1016/S0031-0182(98)00010-8

of

Vonhof, H.B., Wesselingh, F.P., Kaandorp, R.J.G., Davies, G.R., van Hinte, J.E., Guerrero, J., Räsänen, M., Romero-Pittman, L., Ranzi, A., 2003. Paleogeography of Miocene Western Amazonia: isotopic

ro

composition of molluscan shells constrains the influence of marine incursions. Geol. Soc. Am. Bull.

-p

115, 983–993. DOI: 10.1130/B25058.1

Wesselingh, F.P., 2007. Long-Lived Lake Molluscs as Island Faunas: A Bivalve Perspective. In: Renema, W.

re

(Ed.) Biogeography, Time, and Place: Distributions, Barriers, and Islands. Springer, Dordrecht, pp. 275–314.

lP

Wesselingh, F.P., Ramos, M.I.F., 2010. Amazonian aquatic invertebrate faunas (Mollusca, Ostracoda) and their development over the past 30 million years. In: Hoorn, C., Wesselingh, FP. (Eds.) Amazonia,

na

Landscape and Species Evolution: A Look into the Past. Wiley-Blackwell, Oxford, 302–316. Wesselingh, F.P., Räsänen, M. E., Irion, G., Vonhof, H.B., Kaandorp, R., Renema, W., Romero Pittman, L.,

ur

Gingras, M., 2002. Lake Pebas: a palaeoecological reconstruction of a Miocene, long-lived lake complex in western Amazonia. Cainozoic Res. 1, 35–81.

Jo

Wesselingh, F.P., Hoorn, M.C., Guerrero, J., Räsänen, M.E., Romero-Pittmann, L., Salo, J., 2006a. The stratigraphy and regional structure of Miocene deposits in western Amazonia (Peru, Colombia and Brazil), with implications for late Neogene landscape evolution. Scripta Geol. 133, 292–322. Wesselingh, F.P., Guerrero, J., Räsänen, M., Romero Pitmann, L., Vonhof, H., 2006b. Landscape evolution and depositional processes in the Miocene Amazonian Pebas lake/wetland system: evidence from exploratory boreholes in northeastern Peru. Scripta Geol. 133, 323–361. Whatley, R.C., Muñoz-Torres, F., Harten, D. van, 1998. The Ostracoda of an isolated Neogene saline lake in the western Amazon Basin, Peru. Bulletin du Centres de Recherches Elf Exploration-Production, Memoires 20, 231–245. Whatley, R.C., Muñoz-Torres, F., Harten, D. van, 2000. Skopaeocythere: a minute new limnocytherid (Crustacea, Ostracoda) from the Neogene of the Amazon Basin. Ameghiniana, 37, 163–167. 29

Journal Pre-proof Wrozyna, C., Meyer, J., Gross, M., Ramos, M.I.F., Piller, W.E., 2018. Definition of regional ostracod morphotypes by use of landmark-based morphometrics. Freshw. Sci. 37, 573–592. DOI: 10.1086/699482

Figure captures Fig.1. A: Simplified overview map showing northern South America with country borders and selected cities. The oceans are presented in blue, while the Andes are colored in light grey and the cratonic

of

mountains in dark grey. Lowlands are in white, showing major structural highs and sedimentary basins and the isopach thickness of the Solimões Formation is based on Maia et al. (1977) in

ro

western Brazil Overview topographical map overview including country borders and marking the

-p

window the study area; isopach thickness of the Solimões Formation is based on Maia et al. (1977); B: Isopach map of the Solimões Formation in the study area based on Maia et al. (1977) including

re

the position of the studied records (1AS..AM) of 4A and 51 (Hoorn, 1993; Vonhof et al., 2003); 7D and 8 (Linhares et al., 2017, 2019); 10 (Gross et al., 2014, 2017); 19 and 27 (Silva-Caminha et al.,

lP

2010); 31 and 34 (Kachniasz and Silva-Caminha, 2016); 32 (Silva, 2004); 33 (Leite et al., 2017; Meideiros et al., 2019, 2017) ; 51 and 52 (Leandro et al., 2019); 105 (Jaramillo et al., 2017) and the

na

herein studied core 14 in red; C: Regional elevation map of the studied area showing similar elevations between studied cores shown in Fig. 1b; d: Lithological column of the drilled core 1AS-

ur

14-AM based on Maia et al. (1977) with the position of the palynological, ostracod (depth

Jo

information) and sandstone for integrated detrital zircon samples (z1; z2; z3). Fig.2. A-D: Zircon measurements showing 206Pb/238U age spectra distributions of the entire population of each sample as well as combined results based on LA-ICPMS; E: cross-correlation coefficient between the combined signal and a Monte Carlo simulation calculated following Sundell and Saylor (2017); F: Selected zircon age measurements are summarized; G-H: additional precise age determinations and errors of the youngest detrital zircons based on SHRIMP analyses are shown for the uppermost sample z1 (35.5–45.5 m) and the lowermost sample z3 (84.5–94.5 m). Fig.3. Palynological results of AM14 shown in percentages; all data is based on the pollen and spore assemblages, except diagrams of Azolla sp. and foraminifera linings, which are based on the entire palynological counts. Details on each group are presented in Table 1 and supplementary Table2.

30

Journal Pre-proof Fig.4. Summary on the estimate ages based on palynological and ostracod biostratigraphy with the distribution of their most significant indicators and maximum depositional ages of the U/Pb measurements on detrital zircon grains in the centered lithology.

Table 1. Overview all the presumed ecology of detected palynomorphs and their known botanical

Jo

Ceiba sp.

re

lP

ur

na

Bombacacidites nacimientoenis Bombacacidites psilatus Byttneripollis ruedae Catostemma-type Clavainaperturits microclavatus Clavamonocolpites orinocoensis Crotopollenites annaemarie Cteholophonidites suigensis Echiperiporites estelae Heterocolpites incomptus Heterocolpites rotundus Heterocolpites verrucatus Ilexpollenites sp. Kuyisporites waterbolki Margicolporites vanwijhei Multimarinites vanderhammeri Perisyncolporites pokoryi Podocarpiidites sp. Polydopollis mariae Psilamonocolpites rincoii Retimonocolpites absyae Retitricolpites lorantae Retitricolpites simplex

ro

botanical affinity Arecaceae Arecaceae Arecaceae Bombacoideae Bombacoideae Bombacoideae Bombacoideae Bombacoideae Bombacoideae Bombacoideae Malvaceae Malvaceae Chlorantaceae Arecaceae Euphorbiaceae Apocynaceae Malvaceae Melastomataceae Melastomataceae Melastomataceae Aquifoliaceae Cyatheacae Fabaceae

-p

Palynological morpho-species Arecipites perfectus Arecipites polaris Arecipites regio Bombacacidites araucuarensis Bombacacidites brevis Bombacacidites concavoides Bombacacidites fossulatus Bombacacidites muinaneorum

of

affinity following Jaramillo et al. (2010).

Acanthaceae Malphgiaceae Podocarpaceae Fabaceae Arecaceae Arecaceae Bombacoideae Anacardiaceae?

Byttneria Hedyosmum

Gleissospermum Miconia?

Ilex sp.

TrichanteraBravaisia Podocarpus sp.

ecology rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest 31

Journal Pre-proof

Jo

of

ro

-p

re

ur

na

Ranuculacidites operculatus Retimonocolpites absyae Retimonocolpites maximus Retitricolporites kaasii Rugotricolporites arcus Syncolporites poricostatus Tetracolporopollenites transversalis Trichotomosulcus normalis Cosinipollenites oculanoctis Mauritiidites franciscoi Retitrecolporites irregularis Cichoreacidites longispinosus Cyperaceites neogenicus Echiperiporites spinosus Fenestrites spinosus Glencopollis curvimuratus

Sapindaceae-Proteaceae Anonnaceae Moraceae Moraceae Arecaceae Arecaceae Arecaceae Arecaceae Arecaceae Arecaceae Apocynaceae Sapotaceae Moraceae Euphorbiaceae Alchornea Arecaceae Arecaceae Euphorbiaceae Dalechampia Chrysobalanaceae Licania Myrtaceae Sapotaceae Arecaceae Onagraceae Ludwigia Arecaceae Maurita flexuosa Euphorbiaceae Amanoa Asteraceae Cyperaceae Asteraceae Asteraceae Polygonaceae Malvaceae Poaceae Polygalaceae Acanthaceae Teliostachya Lythraceae Crenea

lP

Proteacidites triangulatus Proxapertites tertiaria Psiladiporatus minimus Psiladiporatus moratus Psilamonocolpites amazonicus Psilamonocolpites fossulatus Psilamonocolpites grandis Psilamonocolporites medius Psilamonocolpites nanus Psilamonocolpites rinconii Psilastephanoporites herengreenii Psilatricolporites obesus Psiladiporites redundantis

Janufouria seamrogiformis Monoporoporites annulatus Psilastephanocolporites fissilis Retitrescolpites traversei Verrutricolpites rotundiporus

rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest rainforest wetland wetland wetland herbs herbs herbs herbs herbs herbs herbs herbs herbs herbs

Highlights 

Zircon ages indicate late Miocene age for the top 95 m of the Solimões Formation.



Detrital zircon grains provenance shows an Andean source.



Foraminifera are present Shallow slightly saline ponds persisted in the late Miocene deposits of the Solimões Basin. 32

Journal Pre-proof 

Ostracod and palynology indicate dynamic environmental shifts. Dynamic environments shifted from rivers with wetlands to small lakes.



Palynological and ostracod biostratigraphy show agreements/disagreements. Radiometric zircon ages verify palynological and ostracod biostratigraphic concepts.

Highlights (clear version) Zircon ages indicate late Miocene age for the top 95 m of the Solimões Formation.



Detrital zircon grains show an Andean source.



Shallow slightly saline ponds persisted in the late Miocene of the Solimões Basin.



Dynamic environments shifted from rivers with wetlands to small lakes.



Radiometric zircon ages verify palynological and ostracod biostratigraphic concepts.

Jo

ur

na

lP

re

-p

ro

of



33