Regional Biomes outperform broader spatial units in capturing biodiversity responses to land-use change
Data files
Nov 22, 2024 version files 30.04 KB
-
README.md
5.02 KB
-
RegionalBiomes-Code.zip
25.02 KB
Abstract
Biogeographic context, such as biome type, has a critical influence on ecological resilience, as climatic and environmental conditions impact how communities respond to anthropogenic threats. For example, land-use change causes a greater loss of biodiversity in tropical biomes compared to temperate biomes. Furthermore, the nature of threats impacting ecosystems varies geographically. Therefore, monitoring the state of biodiversity at a high spatial resolution is crucial to capture variation in threat-responses caused by biogeographical context. However such fine-scale ecological data collection could be prohibitively resource intensive. In this study, we aim to find the spatial scale that could best capture variation in community-level threat responses whilst keeping data collection requirements feasible. Using a database of biodiversity records with extensive global coverage, we modelled species richness and total abundance (the responses) across land-use types (reflecting threats), considering three different spatial scales: biomes, biogeographical realms, and regional biomes (the interaction between realm and biome). We then modelled data from three highly sampled biomes to ask how responses to threat differ between regional biomes and taxonomic group. We found strong support for regional biomes in explaining variation in species richness and total abundance compared to biomes or realms alone. Our biome case studies demonstrate that there is variation in magnitude and direction of threat responses across both regional biomes and taxonomic group, although the interpretation is limited by sampling bias in the literature. All groups in tropical forest showed a consistently negative response, whilst many taxon-regional biome groups showed no clear response to threat in temperate forest and tropical grassland. Our results provide the first empirical evidence that the taxon-regional biome unit has potential as a reasonable spatial unit for monitoring how ecological communities respond to threats and designing effective conservation interventions to bend the curve on biodiversity loss.
README: Regional Biomes outperform broader spatial units in capturing biodiversity responses to land-use change
https://doi.org/10.5061/dryad.dr7sqvb5m
Description of the data and file structure
Analysing biodiversity responses to land use change across regional biomes
Files and variables
Data must be downloaded from original sources and placed in this folder.
PREDICTS data
This project uses data from the 2016 release of the PREDICTS database (Hudson et al., 2016). The dataset can be found at https://doi.org/10.5519/0066354
This analysis starts with the RDS format of the data, called 'database.rds'. Place this in a folder called 'PredictsData'
Terrestrial Ecoregions of the world Map
Ecoregion data was originally downloaded from the The Nature Conservancy's (TNC) conservation atlas (http://maps.tnc.org/gis_data.html). At the time of publishing, the link to this dataset is broken , but the same dataset can be downloaded from ResourceWatch here. Note, the labelling of ecoregions in this dataset follows the same format as the PREDICTS database. Other ecoregions maps are available, for example from WWF, but will require some editing to fit with the version of the PREDICTS database used here. All ecoregion maps are based on Olson et al., 2001.
References:
Lawrence Hudson; Tim Newbold; Sara Contu; Samantha L L Hill et al. (2016). The 2016 release of the PREDICTS database [SUPERSEDED] [Data set]. Natural History Museum. https://doi.org/10.5519/0066354
Olson, D. M., Dinerstein, E., Wikramanayake, E. D., Burgess, N. D., Powell, G. V. N., Underwood, E. C., D'Amico, J. A., Itoua, I., Strand, H. E., Morrison, J. C., Loucks, C. J., Allnutt, T. F., Ricketts, T. H., Kura, Y., Lamoreux, J. F., Wettengel, W. W., Hedao, P., Kassem, K. R. 2001. Terrestrial ecoregions of the world: a new map of life on Earth. Bioscience 51(11):933-938.
Code/software
Package requirements:
The versions listed here are the versions used at publication, but future versions may be usable.
- R Version - 4.0.4
- dplyr Version 1.1.3
- sf Version 1.0-16
- devtools Version 2.4.0
- predictsFunctions Version 1.0
- knitr Version 1.31
- tidyr Version 1.3.0
- reshape2 Version 1.4.4
- flextable Version 0.6.4
- officer Version 0.3.17
- lme4 Version 1.1-28
- sjPlot Version 2.8.7
- data.table Version 1.14.2
- ggplot2 Version 3.4.1
- performance Version 0.7.0
- StatisticalModels Version 0.1
- cowplot Version 1.1.1
- patchwork Version 1.1.1
Description of R Scripts
The analysis uses a series of R scripts. The entire analysis can be run by calling 'RunAllScriptsHere.R'. This should take approximately 2.5 hours depending on compute power.
01_CreateSiteMetrics.R - uses database.rds from Data/PredictsData folder - this script uses the mergesites function to group observations from a study to one row to give species richness and total abundance - here we also choose which columns to retain. Output: 02_PREDICTSDivMetrics.csv
02_PreProcessingModelData.R - remove 'urban' and 'Cannot decide' land use types - edit land use & use intensity factor names & add regional biome names, create landuse:use intensity factor for all 5 land use types - create taxa variable Output: 03_PREDICTSModelData.csv ; 03_PREDICTSModelData.rds
03_ExploreModelData.R - use this script to create summary tables of data - number of observations in each biome and regional biome - number of observations in each land use type by biome and regional biome - number of observations in each taxon group by biome and regional biome - it would be useful to save these tables as CSVs that can be brought back in later to subset
04_GlobalModels&Figs.R - run global model of species richness/abundance with land use change - create figures for: biodiversity change with land use - biodiversity change with land use and regional biome - biodiversity change with land use and regional biome and taxa
05_BiomeCaseStudies.R - Subset predicts database to 3 biomes, run regional biome model on species richness and abundance and plot predictions in change in metric over land-use change. Explore the impact of evening out sample sizes. Explore the impact of adding taxon to the species richness model.
PredictGLMERfunction.R - a copy of the PredictGLMER function from the StatisticalModels Package, called in case there are issues loading this package.