Data from: Demographic processes and fire regimes interact to influence plant population persistence under changing climates
Data files
Nov 22, 2024 version files 16.29 MB
-
dryad_data.zip
16.29 MB
-
README.md
5.80 KB
Abstract
Individual and interactive effects of changing climate and shifting fire regimes are influencing many plant species across the globe. Climate change will likely have significant impacts on plant population viability over time by altering environmental conditions and wildfire regimes as well as influencing species demographic traits. However, the outcomes of these complex interactions for different plant functional types under future climate conditions have been rarely examined. We used a proof-of-concept case-study approach to model multiple plant species across two functional plant types, obligate seeder, and facultative resprouter, to examine the interactive effects of demographic shifts and fire regime change on population persistence across two landscapes of over 7,000 km2 in temperate southeastern Australia. Our approach involves a novel combination of a fire regime simulation tool with a spatially explicit population viability analysis model. We simulated fire regimes under six different future climates representing different temperature and precipitation shifts and combined them with 16 hypothetical plant demographic change scenarios, characterised by changes to individual or multiple plant demographic processes. Plant populations were more likely to decline or become extinct due to changes in demographic processes than in the fire regime alone. Although both functional types were vulnerable to climate-induced changes in demography, obligate seeder persistence was also negatively influenced by future fire regimes characterised by shorter fire intervals. Integrating fire regime simulations with spatially explicit population viability analyses increased our capacity to identify those plant functional types most at risk of extinction, and why, as fire regimes change with climate change. This flexible framework is a first step in exploring the complex interactions that will determine plant viability under changing climates and will improve research and fire management prioritisation for species in the future.
https://doi.org/10.5061/dryad.mkkwh717x
This dataset includes code and some example data. The code primarily relates to the STEPS analysis with code and functions.
Description of the data and file structure
The folder structure is as follows:
Grids -
--> severity: Contains example fire severity layers for the Grampians and Blue Mountains regions for one year of fires as produced by the FROST model. These files could be replicated for every year of the simulation in order to run an example full analysis, or could be substituted with similar data e.g. a tif with the disturbance metric of choice. This is an input in the STEPS analysis.
Both files have the following attributes:
-
CRS= EPSG:3111 - GDA94/Vicgrid
-
pixel size= 180 by 180 meters
-
Values are between 1 to 5, 5 being the highest severity, 1 being the lowest severity.
--> habitat_BO: Contains a habitat quality layer for Banksia ornata which was developed using Maxent and presence records downloaded from the Atlas of Living Australia (ALA), see publication for reference details.
-
CRS= EPSG:3111 - GDA94/Vicgrid
-
pixel size= 180 by 180 meters
-
Values are between 0 to 1, 1 being the highest habitat quality, 0 being the lowest habitat quality.
--> habitat_HS: Contains a habitat quality layer for Hakea rostrata which was developed using Maxent and presence records downloaded from the Atlas of Living Australia (ALA), see publication for reference details.
-
CRS= EPSG:3111 - GDA94/Vicgrid
-
pixel size= 180 by 180 meters
-
Values are between 0 to 1, 1 being the highest habitat quality, 0 being the lowest habitat quality.
--> habitat_LE: Contains a habitat quality layer for Leucopogon esquamatus which was developed using Maxent and presence records downloaded from the Atlas of Living Australia (ALA), see publication for reference details.
-
CRS= EPSG:3111 - GDA94/Vicgrid
-
pixel size= 180 by 180 meters
-
Values are between 0 to 1, 1 being the highest habitat quality, 0 being the lowest habitat quality.
--> habitat_BS: Contains a habitat quality layer for Banksia serrata which wasdeveloped using Maxent and presence records downloaded from the Atlas of Living Australia (ALA), see publication for reference details.
-
CRS= EPSG:3111 - GDA94/Vicgrid
-
pixel size= 180 by 180 meters
-
Values are between 0 to 1, 1 being the highest habitat quality, 0 being the lowest habitat quality.
All habitat folders contain 1 tif file with habitat quality maps which were built using code from https://esajournals.onlinelibrary.wiley.com/doi/full/10.1002/ecm.1486
This is an input in the STEPS analysis.
Fire_regime_variables
This folder contains the summarised outputs of the FROST simulations for each replicate and climate model.
- Model= location of fire regime data
- cat= category of simulations i.e., CF_CW equates to current fuel and current weather, FF_FW equates to future fuel and future weather.
- rep= replicate number
- mean_interval= mean fire interval in years
- mean_high_int_int= mean high intensity fire interval in years
- mean_int_native= mean interval in native vegetation only in years
- mean_annual_extent_highint= mean annual extent of high intensity fire in hectares
- annual_extent= annual extent of wildfire in hectares
- season_midpoint= The number of days into the fire season where the area burnt hits 50% of the season total
Scenarios
This folder contains the parameters for each species for one climate model and one set of demographic changes (10% reductions in all parameters). This is an input to the STEPS analysis.
- Species= Species of interest
- climate= CF_CW equates to current fuel and current weather, FF_FW equates to future fuel and future weather.
- start_ind_all= Number of individuals at the start of the simulation
- max_ind_plants= the max number of individuals in the simulation i.e., the ceiling.
- Rmax= not used in these simulations but indicates population growth (rmax>1) or population decline (rman<1)
- MTSS= Maximum training sensitivity plus specificity, not used in these simulations but is an output of Maxent modelling
- fire_rep = fire replicate number
- clim_model= climate model used
- Scenario= Numeric indicator of scenario number
Remaining columns relate to the specific transition matrix of the species. Every value that needed to be changed within the simulation e.g. seed survival, has a column and its associated value depending on the scenario. Transition and survival values are between 0-1 and represent a proportion. Reproduction columns are in whole numbers and represent how many individuals (in this case seeds) are produced per adult. The terminology NF and F indicate if the value is used when the cell is burnt (F=fire) or non burnt (N=NF).
Scripts
This folder contains 8 scripts related to the STEPS analysis, please see the Code section below for details.
Sharing/Access information
Key packages used in this analysis are: \
STEPS: https://doi.org/10.1111/2041-210X.13354
Code/Software
Scripts are all contained within the scripts folder. All code is written in R version 4.3.2.
Each species has 2 scripts- “run_simulations” and “custom_functions”
The “custom_functions” script is sourced within the “run_simulations” file.
The working directory is based on the folder structure within this supplementary data.
All required R libraries are listed within the R scripts.