Back EppAm

The Early Peopling of Amazonia and the Beginning of Plant Domestication

The Early Peopling of Amazonia and the Beginning of Plant Domestication
This project investigates: 1) the social-ecological dynamics of earliest people of Southwestern Amazonia and 2) the role these early Holocene populations played in the domestication of Neotropical plants.

1 Context and overall objectives of the project

This project investigates two relevant topics in the prehistory of the Neotropics: 1) the social-ecological dynamics of earliest people of Southwestern Amazonia and 2) the role these early Holocene populations played in the domestication of Neotropical plants. SW Amazonia has a great potential for reconstructing early human-environment interactions in South America and offers a unique opportunity for exploring the domestication of several Amazonian plants and the origin of agriculture in the Neotropical lowlands. Of all the domesticated crops of the Americas, about half seem to have originated in the Amazon Basin. In particular, based on genetic studies that look at the similarities between domesticated plants and their closest wild relatives, SW Amazonia has been proposed as the potential area for the domestication of cassava, peanuts, jack bean, two species of chilli pepper and peach palm. So far, the genetic evidence is not supported by archaeological data, which is difficult to produce as organic remains are poorly preserved in the dry/humid environments of SW Amazonia. A specific innovative aspect of this proposal is the use of plant micro-remains (mostly phytoliths) from recently discovered early and mid-Holocene archaeological sites (shell middens) in order to study plant exploitation strategies, even in these difficult preservation environments.

Figure 1. A forest Island covering a mid-Holocene shell midden

 

Phytoliths will be also used to reconstruct vegetation changes troughtout the Holocene in the Llanos de Moxos (LM), where the shell middens are. The Holocene environmental history of most of the LM is largely unknown, as there is no paleoecological archive from the LM going back to the mid-Holocene. The main reason for this lack of data is that the lacustrine sediments from the hundreds of lakes that dot the central and southern LM are extremely difficult to core due to their stiff clay sediments and, if cored, provide very shallow sedimentary archives, spanning only a few thousand years.  Because of the lack of suitable lakes to carry out Holocene environmental reconstructions in most of the LM, the present study is based on stratigraphic archives built by past river inundations. These archives contain many phytolith-rich paleosols, which we use for paleoenvironmental and paleoclimatic reconstructions.

The project provides important contributions to the long-term debate of the origin of domesticated plants in the Amazon and the peopling of Amazonia. The nature of the early occupation in SW Amazonia, including the antiquity, settlements patterns, subsistence strategies, processes of social change and environmental impacts, is largely unknown. Understanding the nature of the first human settlements is also a much-needed step in order to understand the region’s cultural diversity. Moreover, this work will help assess the environmental impact of early Amazonians. This data is key in order to correctly interpret paleo-ecological archives, in particular lacustrine sediments where pollen and charcoal signals could be the results of human disturbance. This research integrates archaeological evidence from hunter-gatherers/early cultivators with multiproxy environmental and archaeobotanical data. The new data arising from the project contributes to the understanding of the early Holocene environment, population dynamics and the economy of ancient South Americans, which will be valuable to policy makers at a continental and global level: 1) understanding environmental change and the way it has affected different societies in the past is key to inform future planning, 2) traditional ecological knowledge (TEK) on early uses of domesticated plants can support the work of NGOs and local government to achieve the sustainable exploitation of Amazonia’s natural resources, 3) collected data will improve current environmental modelling.

 

2 Work performed and main results achieved so far

The research work has been organized following two parallel lines of investigation: reconstructing human-environmental interactions and unveiling the time and the processes of plant domestication. Two field seasons have taken place, one during the summer of 2016, immediately preceding the starting of the project, and the second during the summer of 2017.

 2.1 Holocene landscape evolution in the Llanos de Moxos

Most of the field work of 2016 has been focused on gathering new data about fluvial dynamics in the Llanos de Moxos. Rivers are the main actor shaping the landscape, and no study of human-environment interactions can be made without considering the role of rivers and how they have contributed to environmental change during the Holocene. Field and remote sensing observations of small rivers along the Bolivian Andean piedmont have permitted to discover a new process where, frequent (sometimes on yearly basis) blockages of the river course due to the accumulation of woods (logjams) cause large scale forest die-off events, with important consequences in terms of forest biodiversity and flood risk in a large area of the LM. This discovery has been published in 2017 in the journal Earth System Dynamics, title: “River logjams cause frequent large-scale forest die-off events in southwestern Amazonia”. In studies published before the starting of this project, the presence of several paleosols below the alluvial plains of the LM was already documented. During the 2016 fields work, the sampling of these paleosols along a 300 km long transect from the Andes to the central LM was completed. The analysis of 36 stratigraphic profiles plus 50 radiocarbon ages from paleosols intercalated with fluvial sediments permitted to reconstruct the most important changes in floodplain dynamics on a millennial scale and the links between pre-Columbian cultural processes and environmental change. These findings have been published in the journal Quaternary Science Review in 2018, title: “Alluvial plain dynamics and human occupation in SW Amazonia during the Holocene: A paleosol-based reconstruction”.  These paleosol are currently the only paleoecological archive available for the reconstruction of past vegetation. The analyses of stable carbon isotopes, from 36 paleosols, and biogenic silica, from 29 paleosols, show that the patchwork of forests and savannahs that we see today was established after the 4 kyr BP climate change. During the dry period between 8 and 4 kyr BP, most of the central Llanos de Moxos, nowadays covered with seasonally flooded savannah, were covered by Cerrado-like savannah in the west and by forest in the east. However, results also suggest that, at both regional and local scales, vegetation cover has been influenced by changes in topography resulting from the region’s river dynamics. The paper presenting these results is currently under second round of revisions in the journal Global and Planetary Change and should be published between the end of 2018 and the beginning of 2019.

 

Figure 2. Material at 1.5 meters depth under a forest island. Very rich in organic matter and burnt earth.

 

2.2 Early Holocene peopling of SW Amazonia and the domestication of Neotropical plants.

2017 field work has focused on surveying and sampling of forest islands (Figure 1 and 2). Forest Islands are small patches of forest surrounded by savannah under which most of the Moxos’ early and mid-Holocene archaeological sites are found.

 

Figure 3. Examples of phytoliths of economic species found in the studied aracheological sites: A) Marantacea, B) Squash, C) Mays, D) Cassava.

 

Umberto Lombardo (PI) also mantains a web blog about the Amazon Basin as seen from the Llanos de Moxos (part of EppAm). You can follow Umberto clicking here.

Principal researchers

Umberto Lombardo and Marco Madella

Researchers

Javier Ruiz-Perez

Funding for this project was from the European Commission H2020 Marie Skłodowska-Curie actions (MSCA)