Dataset DOI: 10.5061/dryad.fj6q5748r

Description of the data and file structure

This

This dataset is comprised of several .csv files and an R script built to synthesize a stochastic Integral Projection Model (IPM) from census data, and to analyse the outputs of the IPM simulations under a range of climatic and harvesting scenarios.

Included in this folder are:

• "A_pearsonii_censuses_1999_2003.csv" - the raw census data required to fit the vital rate models.
• "Knersvlakte_climate_variables_1950_2100.csv" - the annual climate class transitions required to construct Markovian climate matrices.
• "A_pearsonii_init_popsizes.csv" - plant density estimates across the study region, used to initialize the IPM.
• "A_pearsonii_pop_projections.csv" - results of the population simulations published in the article linked to this dataset.
• "A_pearsonii_popsize_results_new.csv" - a template data frame to store the results of a new set of simulations, if required.
• "Engelen et al 2026 Argyroderma pearsonii IPM.R" - the R script used to synthesize the IPM, run the population simulations, and analyse the outputs:

Files and variables

File: A_pearsonii_init_popsizes.csv

Description: Argyroderma pearsonii population densities recorded during the census years 1999-2003. Used to initialize the IPM simulations.

Variables

• site: site name (1-6)
• plot: 0.25 m² plot name within each site
• year: year of census (1999-2003)
• n: number of individual plants counted in the plot
• density_m2: number of individual plants per m² (i.e. 4 x n)

File: A_pearsonii_popsize_results_new.csv

Description: A template for storing Argyroderma pearsonii population size calculations during the simulations, should the user wish to repeat them.

Variables

• p_h: probability of annual harvesting event occurrence (i.e., 0.1 = once every 10 years on average, 1 = once per year).
• p_nh: probability of no annual harvest (i.e., 1 - p_h)
• scenario: type of harvesting scenario (1 = plants non-selective, 2 = plants size-selective, 3 = fruits)
• Columns named 0-100: population size in each of the 100 annual transitions iterated by the IPM. These will be blank cells until populated by running the simulation script.

File: Knersvlakte_climate_classifications_ERA5_Land_CMIP6.csv

Description: Annual climate classifications used to parameterize Markov transition matrices for the stochastic simulation of climate change effects on Argyroderma pearsonii populations. Classifications are based on mean annual temperature and total annual precipitation of the Knersvlakte (extent = 31.42:31.35S, 18.64:18.69E), derived from monthly ERA5-Land reanalysis climate data (1950-2023), and CMIP6 UKESM1-0-LL climate projections for Shared Socioeconomic Pathways SSP126 and SSP245 (2015-2100). Processed source data for the climate classifications are available from the authors upon request. The raw climatic data can be found at the Copernicus Climate Data Store (see Access information). 

Variables

• year: year of annual climate aggregate (1950 - 2100)
• dataset: climate scenario for which data were recorded (i.e., historical, SSP126, SSP245, or SSP585)
• climate: annual climate classification of the year given by the year column relative to the median conditions of the census period (warm-dry, cool-wet, cool-dry, warm-wet)
• climate_next: annual climate classification of the year following the year given by the year column

File: A_pearsonii_pop_projections.csv

Description: The complete set of Argyroderma pearsonii population size projections presented in the manuscript linked to this dataset.

Variables

• p_h: probability of annual harvesting event occurrence (i.e., 0.1 = once every 10 years on average, 1 = once per year).
• p_nh: probability of no annual harvest (i.e., 1 - p_h)
• scenario: type of harvesting scenario (1 = plants non-selective, 2 = plants size-selective, 3 = fruits)
• Columns named 0-100: population size in each of the 100 annual transitions iterated by the IPM

File: A_pearsonii_censuses_1999_2003.csv

Description: Argyroderma pearsonii annual census data for six populations from 1999-2003. Missing values are NAs (e.g., leaf sizes of individuals that did not survive, or seed numbers of individuals that failed to set fruit).

Variables

• year: year of census (1999-2003)
• site: site name (1-6)
• plot: 0.25 m² plot name within each site
• grid_row: grid cell row within each plot
• grid_col: grid cell column within each plot
• lat: site latitude in decimal degrees
• lon: site longitude in decimal degrees
• lfsz: sum of leaf diameters taken across all leaf pairs in mm in year t
• ln_lfsz: natural log of lfsz
• lfsz_next: sum of leaf diameters taken across all leaf pairs in mm in year t + 1
• ln_lfsz_next: natural log of lfsz_next
• plsz: maximum diameter taken across the plant in mm in year t
• ln_plsz: natural log of plsz
• plsz_next: maximum diameter taken across the plant in mm in year t + 1
• ln_plsz_next: natural log of plsz_next
• branch: quantity of leaf pairs present on an individual (one leaf pair = "single", multiple leaf pairs = "branched")
• pup: did an individual transition from a single to a branched state between years t and t + 1? 1 = yes, 0 = no.
• pdown: did an individual transition from a branched to a single state between years t and t + 1? 1 = yes, 0 = no.
• surv: did an individual that was alive in year t survive to year t +1? 1 = yes, 0 = no.
• fl: did an individual flower in year t? 1 = yes, 0 = no.
• fr: did an individual set fruit in year t? 1 = yes, 0 = no.
• ncap_fact: quantity of fruit capsules produced. Can be "single" or "multiple".
• nseeds: total number of seeds produced by an individual.
• rec: is an individual a new recruit? 1 = yes, 0 = no

File: Engelen_et_al_2026_Argyroderma_pearsonii_IPM.R

Description: R code required to implement harvesting simulations, formulate vital rate models, synthesise and run the IPM, and analyse the outputs. This script was compiled for R version 4.3.2, and requires installation of the packages "glmmTMB", "ggplot2", "ggpubr", "gamlss", "gamlss.tr", "ipmr", and "tidyverse".

Code/software

Any software capable of reading .csv files (e.g., R, Microsoft Excel) is required to view the data. R statistical software is required to run the attached script and replicate the analysis.

Access information

This repository contains derived annual climate classification outputs generated from ERA5-Land and CMIP6 climate variables. No original source gridded data are redistributed here, nor are they attainable from the derived outputs. Original source datasets should be cited independently according to their respective licenses.

Climate classification data were derived from the following sources under CC-BY license:

• Copernicus Climate Change Service (2022): ERA5-Land monthly averaged data from 1950 to present. Copernicus Climate Change Service (C3S) Climate Data Store (CDS). DOI: 10.24381/cds.68d2bb30

• Copernicus Climate Change Service, Climate Data Store, (2021): CMIP6 climate projections. Copernicus Climate Change Service (C3S) Climate Data Store (CDS). DOI: 10.24381/cds.c866074c