Analysis of long-term historical agroecology at the archaeological site of Dhiban, Jordan (ca. 1000 BCE - 1450 CE)
Table of Contents
This repository contains the markdown and files needed to reproduce the analyses and visualizations found in Farahani, A. 2018. A 2500-Year historical ecology of agricultural production under Empire in Dhiban, Jordan. Journal of Anthropological Archaeology 52: 137-155..
The goal of this research was to identify and understand how socio-political changes over nearly 2500 years at the archaeological site of Dhiban in the Hashemite Kingdom of Jordan affected, and were affected by, agricultural production. Why agriculture? Simply put, people must eat to survive. And in empires, strategies of agricultural production often sit directly in-between the (coercive) power of state elites and the needs of local communities.
To investigate this relationship at Dhiban, I collected and analyzed over 200 archaeological sediment samples to identify archaeological plant remains: these are the actual physical residues (seeds, plant parts, etc.) of the crops these communities were growing. Statistical analysis of the data reveal shifts in the abundances of particular crops correlating to separate moments of nonlocal imperial intervention at Dhiban, in particular a greater emphasis on grape production during the Byzantine (ca. 330–635 CE) period, and of wheat and barley during the Mamluk (ca. 1250–1450 CE) period.
I argued that these shifts were more likely a product of changes in socio-cultural priorities (more importance on sheep and goat husbandry in the Mamluk period) and economic realities (greater local trade in the Byzantine period) in each of these time periods than due to climate change alone, although increasing aridity in the Mamluk period had some effect on the choices of farmers at Dhiban regarding what to grow. Obviously, there are different interpretations of these data.
Note that the data in this repository are recent up to the point of this specific study in 2018 -- newer samples, radiocarbon dates, and other data from archaeological research at Dhiban will be housed in a new repository (timeline and link TBD).
See the compiled analyses and figures here.
For more about the archaeological site of Dhiban, see:
There are two folders, one for data, and another for the markdown. See descriptions below:
-
dhb_c14_dates.csv
: The results of radiocarbon dating on archaeological short-lived annual plant remains, mainly the caryposes of wheat and barley (with some grape). Uncalibrated laboratory estimates and errors are provided as well as the laboratory number, the archaeological context (unit and locus), the δ13C values (where measured), and the 1σ and 2σ calibrated ranges. Note, however, that these calibrated ranges are outdated because of updates to the IntCal curve. It is recommended that they are re-calibrated with newer calibration curves (using OxCal,rcarbon
, etc.). -
dhb_raw_peb_data.csv
: The counts of the different archaeological plant remains recovered at the archaeological site of Dhiban, arranged by sample, and including some associated metadata, such as exact Cartesian location (x,y,z) of that sample, project sample number, volume of processed sediment for that sample, and inferred archaeological context (bin, pit, fill, etc.). -
dhb_seed_morph.csv
: Morphometric data of wheat (Triticum spp.) and barley (Hordeum vulgare subspp.) species recovered at (the archaeological site of) Dhiban. Includes length, width, thickness, permiter, area, and some other measures. -
estimate_s_output.csv
: As the filename states, this is the output of analyses conducted in the very helpful EstimateS software managed by Robert Colwell, which is used for "statistical estimation of species richness". This program was used to infer (crop) species richness for each archaeological time period. The calculation was based on taxonomic abundance by sample. Here richness refers to the ecological concept, namely the number of unique instances of some kind of biological entity (species, genus, etc.) in a defined area. Today, I would use the iNEXT package in R. In fact, with Reuven Sinensky, I already have (see here)!
-
supplement.Rmd
: An (R) Markdown that allows for interactive reproducibility of all of the figures and tables found in the manuscript, plus some additional analyses. -
supplement.html
: A compiled version of the Markdown in the event you just wanted to look at the analyses and figures .
You will need:
The packages needed for markdown compilation are listed at the beginning of the (R) markdown.
You can contribute to this repository by identifying coding errors, proposing more elegant syntax, or suggesting additional analyses / visualization. Any contributions you make are very welcome.
If you have a suggestion, first please open an issue with the tag "enhancement". Depending on how much work you've already done, I might recommend that you fork the repo and create a pull request.
If so:
-
Fork the project
-
Create a branch (
git checkout -b enhancement/amazinganalysis
) -
Commit your changes (
git commit -m 'Added an incredible new analysis'
) -
Push to the branch (
git push origin enhancement/amazinganalysis
) -
Open a pull request
Distributed under the MIT License. See LICENSE.txt
for more information.
Alan Farahani - alanfarahani@gmail.com
Project Link: https://github.com/alanfarahani/dhiban-bots-analysis
- README template
- 🌱🌿🌾