Supplementary Information - Phytoliths as indicator of plant water availability: the case of millets cultivation in the Indus Valley civilization

Supplementary Information for D'Agostini et al., 2022

This repository contains all the supplementary information cited in the manuscript. This version is an update of the previous one.

Publication:

Title: Phytoliths as indicator of plant water availability: the case of millets cultivation in the Indus Valley Civilisation

Authors: D’Agostini F., Ruiz-Pérez J., Madella M., Vadez V., Kholova J., Lancelotti C.

Year: submitted on 18 August 2022

Journal: Paleobotany and Palynology

Contents:

Figure S1: Silica skeleton size ratio violin boxplots -  Violin boxplots of Silicon skeleton size ratio of finger millet, pearl millet and sorghum by water treatments. White diamond = mean.

File S1: SummaryDataset - Excel file that includes the database of the modern replicas with the specification of the samples in use; physiological data; phytolith data for modern and archaeological assemblages; metadata.

File S1.1: Modern - dataset that includes the database of the modern replicas used for building the model.

File S1.2: Archaeological - dataset that includes the database of the archaeological replicas tested by the model.

File S1.3: Phytoliths2019 - dataset includes the database of phytoliths used for the analysis of concentration and of ratio sensitive to fixed morphotypes.

File S2: Phytolith description - It includes a table with the description of each morphotype category and additional phytolith images.

File S3: R script - R code that includes all the scripts used for statistics and data visualisation.

File S4: Metadata - table that includes the descriptions of all the variables of the datasets.

Abstract:

The interpretation of crop water management practices has been central to the archaeological debate on agricultural strategies and is crucial where the type of water strategy can provide fundamental explanations for the adoption and use of specific crops. Traces of water administration are difficult to detect and are mostly indirect, in the form of water harvesting or distribution structures. Attempts have been made to infer plant water availability directly from archaeobotanical remains. Current evidence suggests that the ratio of sensitive to fixed phytolith morphotypes can be used as a proxy for water availability in C₃ crops, as well as in sorghum and maize. Nevertheless, the controversy on whether genetically and environmentally controlled mechanisms of biosilica deposition are directly connected to water availability in C₄ crops is open, and several species remain to be tested for their phytolith production in relation to water levels. This research aims at clarifying whether leaf phytolith assemblages and concentration, silica skeleton size and ratio of sensitive to fixed morphotypes can be related to different water regimes in Eleusine coracana Gaertn., Pennisetum glaucum (L.) R. Br., and Sorghum bicolor (L.) Moench. We cultivated 5 traditional landraces for each species in lysimeters, under different watering conditions and analyzed their phytolith content/production in leaves. results show higher proportions of long cells, bulliforms and stomata produced in well water conditions. The model built on the basis of phytolith composition is then applied to interpret archaeological phytolith assemblages recovered from a single phase at four different sites of the Indus Civilisation: Harappa, Kanmer, Shikarpur and Alamgirpur. The results show that most probably C₄ crops grew under water stress conditions, providing new data on the interpretation of ancient agricultural management in the Indus Valley.