Data from: Assessing the effects of fire season, fire severity and rainfall on seedling establishment and mortality in dry eucalypt forests
Data files
Apr 27, 2026 version files 1.13 MB
-
Data_finalV2.csv
308.85 KB
-
Fire_tables1.csv
831 B
-
FireSeason_analysis.html
352.60 KB
-
FireSeason_analysis.qmd
33.92 KB
-
MortalityandWeather.csv
129.36 KB
-
README.html
23.61 KB
-
README.md
5.62 KB
-
year1matrix.csv
89.38 KB
-
year2matrix.csv
91.59 KB
-
year3matrix.csv
91.04 KB
Abstract
Fire plays a crucial role in shaping ecosystems globally, yet climate change is driving complex shifts in fire regimes. Fire-weather seasons are lengthening, with many fire-prone ecosystems experiencing increased drying and rising temperatures. Consequently, fire is occurring more frequently outside historical fire seasons, potentially exposing vulnerable species to risk of decline. Here, we examine how fire season and severity interact to influence perennial plant post-fire recruitment in the dry eucalypt jarrah forests of southwest Western Australia. We assessed seedling recruitment and mortality of all species (except graminoids) following 11 fires across two years. The burns occurred in either autumn or spring, and plots were stratified across three severity classes. Each site was surveyed three times post-fire. Seedling density was highest within the first year following autumn fires. However, by the final survey year, densities were highest after moderate and high-severity 2021 spring fires. Emergence following spring fires in both years was delayed with a large portion of seeds remaining viable, anticipating future winter rains. Seedling mortality was highest following autumn fires but differed significantly between burn years. Complex interactions among fire season, fire severity, and post-fire weather collectively shape recruitment patterns, yet our results suggest jarrah forest communities are resilient to variation in fire season and severity. As Southwest Western Australia continues to dry and annual rainfall declines, understanding how these interactions, combined with post-fire weather, drive ecosystem composition will be increasingly important.
Data required:
The data and scripts provided here allow for the full recreation of the analysis undertaken in “Assessing the effects of fire season, fire severity and rainfall on seedling establishment and mortality in dry eucalypt forests.”
There are 7 Excel sheets required to complete the full analysis.
- Fire_Tables1.csv - this provides the complete record of the data and the names of the fires measured in this manuscript.
- Data_finalV2.csv - this is the complete data set used in this analysis, including site and treatment information.
- Year1matrix.csv, Year2matrix.csv, and Year3matrix.csv - Species composition matrix for each of the three survey periods.
- MortalityandWeather.csv - plot level data for Rainfall and Max Temperature sourced from https://www.worldclim.org/data/monthlywth.html, merged with plot level mortality data, extracted from Data_final2.csv
- FireSeason_analysis.html - this is the full script for the analysis, including figure code and all model construction code.
- FireSeason_analysis.qmd - is a markdown file containing all analytical scripts.
Further information:
Below is the metadata for each column for each sheet above.
Fire_tables1.csv - summary table of the plot characteristics.
- Site Name - Burn and Site name
- Burn Date - Date which aligns with core ignitions
- Burn Year - the year in which the fire occurred.
- Burn Season - the broad season in which the fire occurred.
- Treated Area (Ha) - total area within the treatment block
- Burnt Area (Ha) - area that received low-severity fire or more.
- Proportion burnt( %) - actual area of the treatment cell burnt by fire. Proportion burnt( %)= (Burnt Area (Ha)/ Treated Area (Ha))x 100
- Number of low-severity plots - number of plots with a burn that were recorded as low severity
- Number of moderate severity plots - number of plots with a burn that were recorded as moderate severity
- Number of high-severity plots - number of plots with a burn that were recorded as high severity
Data_finalV2.csv - the final dataset used in the analysis comprises abundance and trait data.
- PlotID - Unique plot code comprised of Site name, severity class, and a number from 1 -10
- long - the Longitudinal location of the plot
- lat - the latitudinal location of the plot
- SeverityClass - the severity class of the plot
- BurnDate - Date which aligns with core ignitions
- Site - Burn and Site name
- Season - the broad season in which the fire occurred.
- BurnYear - the year in which the fire occurred.
- Plot - a plot number from 1-10
- Species - the specific species recorded and counted in each plot
- Year1 - the number of the specific species recorded in the first survey year
- Year2 - the number of the specific species recorded in the second survey year
- Year3 - the number of the specific species recorded in the third survey year
- Total - the highest recorded count of seedlings within the emergence period.
- Min - the count of seedlings in the emergence + 1 period.
- Mort - the proportional change (%) between Total and Min - is described by equation 1 in the research paper.
- Dead - the count of the difference between total and min.
- FireResponse - the species response to fire, either “Fire Killed” or “Resprouter”
- Dormancy - the seed dormancy class for the seeds of this species
- MortalitySurveyDate - the nominal date at which mortality is calculated (based on 12 months following emergence)
year1matrix.csv, year2matrix.csv, and year3matrix.csv - each of these three sheets is a species abundance matrix, one for each survey year.
- PlotID - Unique plot code comprised of Site name, severity class, and a number from 1 -10
- Season - the broad season in which the fire occurred.
- SeverityClass - the severity class of the plot
- Site - Burn and Site name
- BurnYear - the year in which the fire occurred.
The rest of the columns are species found in the specific plots .
MortalityandWeather.csv - this sheet is the weather data used in model 4, assessing the influence of temperature and rainfall on species-level mortality.
- PlotID - Unique plot code comprised of Site name, severity class, and a number from 1 -10
- long- Longitude of plot location
- lat - latitude of plot location
- Site - Burn and Site name
- BurnYear - the year in which the fire occurred.
- Rainfall_12mo - cumulative rainfall for the 12 months between E and E+1 (E and E+1 are defined in the manuscript.
- Temp_12mo - monthly mean temperature for the 12 months between E and E+1 (E and E+1 are defined in the manuscript.
- FireResponse - the species response to fire, either "Fire Killed" or "Resprouter"
- Dormancy - the seed dormancy class for the seeds of this species
- SeverityClass - the severity class of the plot
- Season - the broad season in which the fire occurred.
- MeanMort - the average proportional change (%) between Total and Min for all species within a plot - is described by equation 1 in the research paper.
Missing data
Missing data is indicated by “NA” in cells.
Code/Software
The main script is provided in this DRYAD link, provided as a .html. It is fully compiled to re-conduct the analysis as is included in the manuscript. All required data is also provided.