Data from: Quantifying the hydrological niche of upland swamp plant communities using indicator species
Data files
Jun 05, 2026 version files 2.29 MB
-
moddat.csv
3.65 KB
-
README.md
3.14 KB
-
scr_Dryad.R
12.85 KB
-
topmod
1.25 MB
-
topmod3
1.02 MB
Abstract
This repository contains the data used in a study seeking to use indicator species to quantify the hydrological niche of the corresponding plant communities they are indicative of within Coastal Upland Swamps of the Sydney Basin Bioregion, Australia. Swamp vegetation organises into five recognised communities, occupying distinct positions along a hydrological gradient. However, longwall mining reduces the duration of root zone saturation, potentially impacting the observed vegetation.
We explored this question by first modelling the relative frequency (presence in 16 contiguous 0.5 x 0.5 m quadrats) of 20 vascular plant indicator species (four for each community) from 11 unmined sites across four swamps. We quantified the hydrological niche for each community using the average number of days-per-year of root zone saturation (≥30 cm below surface) from time series soil moisture data (predictor variable).
We then used the fitted model to predict the frequency of indicator species based on the observed root zone saturation at a mine-impacted swamp.
Dataset DOI: 10.5061/dryad.p8cz8wb3p
Description of the data and file structure
The data were collected from 14 sites across 5 upland swamps (4 unmined swamps with natural hydrology and 1 mine impacted swamp) on the Woronora Plateau, Sydney Basin Bioregion, Australia. The data consist of:
- soil moisture saturation data (mean fraction of time saturated at 30 cm below ground across 3-5 year periods). This is our predictor variable in analysis (notated 'ave30').
- the relative frequency of 20 plant community indicator species from 16 contiguous quadrats, co-located with each soil moisture monitoring probe. The relative frequency of indicators from each of five communities was the response variable (notated 'freq').
Files and variables
File: topmod
Description: This is an R object containing a fitted regression model output created by R function brms::brm(). To open, requires an active R session with R package brms loaded. The model specific to this object is the 5 community full upland plant indicator species model.
File: moddat.csv
Variables
- record: unique row ID
- site: site of veg and soil moisture sampling
- com_ind: plant community indicator, one of five different plant communities modelled: bt = banksia thicket, rh = restioid heath, ch = cyperoid heath; ttt = ti tree thicket; sl = sedgeland
- freq: frequency of plant communities indicator species in 16 contiguous 0.5 x 0.5 m quadrats
- ave30: average fraction of time saturated at the base of the root zone (30 cm below ground) measured over a 3 year period of record.
- swamp: identity of swamp where the monitoring site was located
- surv: survey type - used in sensitivity tests - see R script for details
- com_3: simplified community descriptor pooling the three dry-adapted communities as a single composite, levels (drier composite, dc, ch, ttt).
File: topmod3
Description: This is an R object containing a fitted regression model output created by R function brms::brm(). To open, requires an active R session with R package brms loaded. The model specific to this object is the 3 community simplified model.
File: scr_Dryad.R
Description: This is an R script file with the code needed to produce the results in the manuscript.
Code/software
moddat.csv can be viewed in Excel or other spreadsheet application.
topmod and topmod3 must be viewed in an active R session, and requires the package brms is loaded.
scr_Dryad.R is an R script and can be run from an R session. To run the analysis, simply save all files in a Windows Folder and double click this file (assumes R for Windows is installed on the machine).
Access information
Other publicly accessible locations of the data:
- NA
Data was derived from the following sources:
- Mason, T. J., M. Krogh, G. C. Popovic, W. Glamore, and D. A. Keith. 2021. Persistent effects of underground longwall coal mining on freshwater wetland hydrology. Science of the Total Environment 772.