Data and R scripts for the paper 'Regional variations in the demographic response to the arrival of rice farming in prehistoric Japan'

This repository contains data and scripts used in the following paper:

Crema, E. R., Carrignon, S., Shoda, S., & Stevens, C. J., & Shoda, S. (In press). Regional variations in the demographic response to the arrival of rice farming in prehistoric Japan. Antiquity.

The repository is organised into five main directories: data, analyses, results, figures, tables, and src. The data folder contains all relevant radiocarbon and settlement data, analyses contains core R scripts for executing all analyses and estimates, results contains R image files of all outputs, figures and tables contain all figures and tables for the manuscript and the supplementary materials as well as R scripts required to generate them, and src contains additional custom utility R functions.

Analyses Summary

Settlement Data Analyses

Settlement data were obtained from the 「縄文・弥生集落データベース」("Jomon-Yayoi settlement database") of the National Museum of Japanese History, and stored as a CSV file (data/site_raw.csv). The file data/data_prep_sites.R contains data wrangling R scripts that convert relevant data into an R data.frame stored in the R image file data/sitedata.RData. Scripts for the composite Kernel Density estimates (see manuscript and ESM for details) are stored in the R file figures/figure2.R. Pipeline: data/site_raw.csv → data/sitedata.RData.

Bayesian Analyses of Radiocarbon Dates

14C data were obtained from radiocarbon database of the National Museum of Japanese History. We joined the cleaned and translated version of the database (c14db_1.1.0.csv, obtained here) to the full Japanese version of the database (.xslx files in /data/rekihaku_downloads, obtained here) to extract relevant field for identifying anthropogenic contexts. The file data/data_prep_c14.R contains pre-processing R scripts for generating the core dataset used in this paper, stored in the R image file data/c14data.RData. Bayesian analyses were carried out using the nimbleCarbon R package, which contains custom probability distributions and utility functions for the NIMBLE probabilistic programming language. The files analyses/icar500.R and analyses/icar750.R contain R scripts for fitting the Bayesian models for the 500 and 750 yrs time intervals. Posterior samples are stored in the R image files results/icar_c14doubleRes500.RData and results/icar_c14doubleRes750.RData. Pipelines:

data/rekihaku_downloads/1200_100_T.xlsx & data/rekihaku_downloads/2000_1201_T.xlsx & data/rekihaku_downloads/3000_2001_T.xlsx & data/rekihaku_downloads/5000_3001_T.xlsx → data/rekihaku_downloads/bindCSV.R → data/rekihaku_downloads/binded.csv.
data/rekihaku_downloads/binded.csv & data/rekihaku_downloads/c14db_1.1.0.csv → data/data_prep_c14.R → data/c14data.RData.
data/c14data.RData → analyses/icar500.R → results/icar_c14doubleRes500.RData
data/c14data.RData → analyses/icar750.R → results/icar_c14doubleRes750.RData

Absolute Population Estimates

Absolute population estimates were calculated using a modified version of the equation introduced by Koyama (Koyama, S.1978. Jomon Subsistence and Population. Senri Ethnological Studies, 2, 1–65.) with an updated dataset. The core calculations are included as an R script in the file analyses/pop_dens_est.R, with the outputs stored in the CSV files results/pop_estimate_compare.csv (comparison of the different estimates discussed in the supplementary materials) and results/pop_estimate_region.csv (estimates used in table 1 of the manuscript). Raw input data required for the calculations are Koyama's original data (data/koyama_popestimate_1984.csv), the number of archaeological sites published by the Japanese Agency of Cultural Affairs (data/maizobunkazai_2017.csv), and a lookup table for matching administrative units (prefectures) to the regions used in this paper (data/prefecture_data.csv). Pipeline: data/maizobunkazai_2017.csv & data/prefecture_data.csv & data/koyama_popestimate_1984.csv → analyses/pop_dens_est.R → results/pop_estimate_compare.csv & results/pop_estimate_region.csv.

Figures and Tables

Main (figures/figure1.pdf ~ figures/figure6.pdf) and supplementary (figures/figureS1.pdf ~ figures/figureS3.pdf) are generated using the Rscript in figures/figures_main.R and figures/figures_esm.R. Pipelines:

Main Figures: data/c14data.RData & data/sitedata.RData & results/icar_c14doubleRes500.RData & results/icar_c14doubleRes750.RData → figures/figures_main.R → figures/figure1.pdf ~ figures/figure6.pdf
Supplementary Figures: results/pop_estimate_compare.csv → figures/figures_esm.R → figures/figureS1.pdf ~ figures/figureS3.pdf

File Structure

data

c14data.RData
data_prep_c14.R
data_prep_sites.R
data_summary.R
koyama_popestimate_1984.csv
maizobunkazai_2017.csv
prefecture_data.csv
sitedata.RData
site_raw.csv

/data/rekihaku_downloads

1200_100_T.xlsx
2000_1201_T.xlsx
3000_2001_T.xlsx
5000_3001_T.xlsx
c14db_1.1.0.csv
binded.csv
bindCSV.R

analyses

icar500.R
icar750.R
pop_dens_est.R

results

icar_c14doubleRes500.RData
icar_c14doubleRes750.RData
pop_estimate_compare.csv
pop_estimate_region.csv

figures

figures_main.R
figures_esm.R
figure1.pdf ~ figure6.pdf
figureS1.pdf ~ figureS3.pdf

tables

tables_main.R
tables_esm.R
table1.csv
tableS1.csv
tableS2.csv

src

dbscanID.R

R Session Info

attached base packages:
[1] parallel  stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
 [1] dbscan_1.1-11       spdep_1.2-8         spData_2.2.2        coda_0.19-4        
 [5] sf_1.0-13           rnaturalearth_0.3.2 latex2exp_0.9.6     RColorBrewer_1.1-3 
 [9] here_1.0.1          rcarbon_1.5.1       nimbleCarbon_0.2.4  nimble_1.0.1       
[13] gridExtra_2.3       dplyr_1.1.2         ggplot2_3.4.2      

loaded via a namespace (and not attached):
 [1] tidyselect_1.2.0       fastmap_1.1.1          spatstat.geom_3.2-2   
 [4] pracma_2.4.2           spatstat.explore_3.2-1 digest_0.6.31         
 [7] rpart_4.1.19           lifecycle_1.0.3        spatstat.data_3.0-1   
[10] magrittr_2.0.3         compiler_4.3.0         rlang_1.1.1           
[13] doSNOW_1.0.20          tools_4.3.0            igraph_1.5.0          
[16] utf8_1.2.3             yaml_2.3.7             knitr_1.43            
[19] sp_1.6-1               classInt_0.4-9         abind_1.4-5           
[22] KernSmooth_2.23-20     withr_2.5.0            purrr_1.0.1           
[25] numDeriv_2016.8-1.1    grid_4.3.0             polyclip_1.10-4       
[28] fansi_1.0.4            e1071_1.7-13           colorspace_2.1-0      
[31] progressr_0.13.0       scales_1.2.1           iterators_1.0.14      
[34] spatstat.utils_3.0-3   spatstat_3.0-6         cli_3.6.1             
[37] rmarkdown_2.21         generics_0.1.3         rstudioapi_0.14       
[40] httr_1.4.6             DBI_1.1.3              proxy_0.4-27          
[43] stringr_1.5.0          splines_4.3.0          spatstat.model_3.2-4  
[46] s2_1.1.4               vctrs_0.6.3            boot_1.3-28.1         
[49] Matrix_1.5-4           jsonlite_1.8.4         tensor_1.5            
[52] elevatr_0.4.5          foreach_1.5.2          units_0.8-2           
[55] snow_0.4-4             goftest_1.2-3          glue_1.6.2            
[58] spatstat.random_3.1-5  codetools_0.2-19       stringi_1.7.12        
[61] gtable_0.3.3           deldir_1.0-9           munsell_0.5.0         
[64] tibble_3.2.1           pillar_1.9.0           htmltools_0.5.5       
[67] R6_2.5.1               wk_0.7.3               rprojroot_2.0.3       
[70] evaluate_0.21          lattice_0.21-8         class_7.3-21          
[73] Rcpp_1.0.11            spatstat.linnet_3.1-1  nlme_3.1-162          
[76] spatstat.sparse_3.0-2  mgcv_1.8-42            xfun_0.39             
[79] pkgconfig_2.0.3

Funding

This research was funded by the ERC grant Demography, Cultural Change, and the Diffusion of Rice and Millets during the Jomon-Yayoi transition in prehistoric Japan (ENCOUNTER) (Project N. 801953, PI: Enrico Crema) and by a Philip Leverhulme Prize (PLP-2019-304) in archaeology awarded to Enrico Crema.

Licence

CC-BY 3.0

ercrema / yayoi_demo