Data from: Dispersal and connectivity in increasingly extreme climatic conditions
Data files
Apr 23, 2024 version files 1.97 GB
Abstract
While climate change has been shown to impact several life-history traits of wild-living animal populations, little is known about its effects on dispersal and connectivity.
Here, we capitalize on the highly variable flooding regime of the Okavango Delta to investigate impacts of changing environmental conditions on dispersal and connectivity of the endangered African wild dog (Lycaon pictus). Based on remote sensed flood extents observed over 20 years, we derive two extreme flood scenarios: a minimum and a maximum flood extent; representative of very dry and very wet environmental periods. These conditions are akin to those anticipated under increased climatic variability, as it is expected under climate change. Using a movement model parametrized with GPS data from dispersing individuals, we simulate 12,000 individual dispersal trajectories across the ecosystem under both scenarios and investigate patterns of connectivity.
Across the entire ecosystem, surface water coverage during maximum flood extent reduces dispersal success (i.e., the propensity of individuals to disperse between adjacent subpopulations) by 12% and increases dispersal durations by 17%. Locally, however, dispersal success diminishes by as much as 78%. Depending on the flood extent, alternative dispersal corridors emerge, some of which in the immediate vicinity of human-dominated landscapes. Notably, under maximum flood extent, the number of dispersing trajectories moving into human-dominated landscapes decreases by 41% at the Okavango Delta’s inflow, but increases by 126% at the Delta’s distal end. This may drive the amplification of human-wildlife conflict.
Whilst predicting the impacts of climate change on environmental conditions on-the-ground remains challenging, our results highlight that environmental change may have significant consequences for dispersal patterns and connectivity, and ultimately, population viability. Acknowledging and anticipating such impacts will be key to effective conservation strategies and to preserve vital dispersal corridors in light of climate change and other human-related landscape alterations.
README: Data from: Dispersal and connectivity in increasingly extreme climatic conditions
https://doi.org/10.5061/dryad.z34tmpgnm
This repository contains all R
-code and data to reproduce the analyses and visualizations from Hofmann et al., 2024. It is recommended to explore the data through the provided R
-Scripts. A general design principle was to compartmentalize all analyses and simulations to reduce computational requirements. As such, there are often parent files (in .rds
format) that provide overviews and bundle further data files as tidyverse
data-tibbles. All R
-codes are extensively documented, giving detailed insights into the processing and analytical steps.
Description of the File Structure
The file structure of the repository is as follows:
├── 02_R-Scripts
│ ├── 00_Functions.cpp # Custom C++ functions
│ ├── 00_Functions.R # Custom R functions
│ ├── 01_Analysis # R codes for the simulation analysis
│ │ ├── 01_LandCover.R # Preparation of land cover data
│ │ ├── 02_SourceAreas.R # Generation of source areas from which to simulate dispersal
│ │ ├── 03_Simulation.R # Dispersal simulation under a minimum and maximum flood
│ │ ├── 04_HeatmapsBetweenness.R # Generation of heatmaps and betweenness maps from simulations
│ │ ├── 05_Interpatch.R # Computing interpatch connectivity from simulations
│ │ ├── 06_ExtentFlood.R # Computing flood extent
│ │ └── 07_Distance.R # Identify simulated individuals in the vicinity to humans
│ ├── 02_Visualization # R codes for visualizations
│ │ ├── Distance.R # Figure S9
│ │ ├── Egression.R # Figure S5
│ │ ├── Floodmaps.R # Figure 3
│ │ ├── GraphicalAbstract.R # Figure 1
│ │ ├── InterpatchTable.R # Figure 5, Figure S2
│ │ ├── Metrics.R # Figures 6, S3, S4, S6, S7, S8, S10
│ │ ├── Model.R # Figure S1
│ │ └── StudyArea.R # Figure 2
│ └── 99_SessionInformation # Folder containing R session information
│ └── ...
├── 03_Data
│ ├── 01_RawData # Unprocessed raw data
│ │ ├── FLOODMAPS # Remote sensed floodmaps in .tif raster format. Filename indicates remote sensing date (YYYY.MM.DD)
│ │ ├── GLOBELAND # Static water map (used to represent areas that are not affected by flooding) in .tif raster format
│ │ ├── MERIT # MERIT river data in .tif raster format
│ │ └── MODIS # Modis Continuous Vegetation Fields data in .tif raster format
│ ├── 02_CleanData # Cleaned and pre-processed data
│ │ ├── Africa.shp # Shapefile of African continent used for visualizations
│ │ ├── AreasOfInterest.shp # Shapefile of areas of interest within which we compared human wildlife conflict
│ │ ├── Cutlines.shp # Shapefile of cutlines used to delineate egression zones
│ │ ├── DistanceToHumans.tif # Distance to human influence (villages, roads, agriculture) as .tif raster format
│ │ ├── DistanceToWater.tif # Distance to water (major waters and rivers) in .tif raster format
│ │ ├── Faults.shp # Shapefile of major fault lines near the Okavango Delta for visualizations
│ │ ├── GammaDistribution.rds # Gamma distribution used to sample step-lengths for the step-selection simulation as an R-file (.rds)
│ │ ├── HumanInfluence.tif # Human influence estimates in .tif raster format
│ │ ├── KAZA.shp # Shapefile of KAZA-TFCA borders for visualizations
│ │ ├── MababeDepression.shp # Shapefile of MababeDepression lines for visualizations
│ │ ├── MajorRivers.shp # Shapefile of major river lines for visualizations
│ │ ├── MajorWaters.shp # Shapefile of major water bodies for visualizations
│ │ ├── MovementModel.rds # Movement model using which dispersal is simulated as an R-file (.rds)
│ │ ├── Protected.shp # Shapefile of protected areas for visualizations
│ │ ├── ReferenceRaster.tif # Reference .tif raster (used to crop other rasters)
│ │ ├── ReferenceShape.shp # Reference .shp shapefile (used to crop other shapes)
│ │ ├── Roads.shp # Shapefile of roads for visualizations
│ │ ├── Scaling.rds # Values used to scale extracted covariates
│ │ ├── ShrubCover.tif # Shrub cover in .tif raster format
│ │ ├── SourceAreas.shp # Shapefile of source areas from which dispersal was simulated
│ │ ├── TreeCover.tif # Tree cover in .tif raster format
│ │ ├── Villages.shp # Shapefile of villages for visualizations
│ │ └── WaterCover.tif # Water cover during the extreme scenarios in .tif raster format
│ └── 03_Results # Model and simulation results
│ ├── 99_Betweenness # Folder containing betweenness estimates for each source area in .tif raster format
│ ├── 99_Distances # Folder containing estimates of human wildlife conflict (i.e. low distance to humans) for each source area in .tif raster format
│ ├── 99_Heatmaps # Folder containing heatmaps for each source area in .tif raster format
│ ├── 99_Simulations # Folder containing simulated dispersal trajectories as R-files (.rds)
│ ├── BetweennessGlobal.tif # Consolidated betweenness estimates (from 99_Betweenness) across different source areas in .tif raster format
│ ├── BetweennessLocal.tif # Consolidated betweenness estimates (from 99_Betweenness) per source area in .tif raster
│ ├── DispersalSimulations.rds # Consolidated dispersal simulations (from 99_Simulations) as .rds R-files.
│ ├── Distance.rds # Consolidated estimates of human wildlife conflict (i.e. low distance to humans, from 99_Distances) as .rds R-file
│ ├── Distance.tif # Consolidated estimates of human wildlife conflict (i.e. low distance to humans, from 99_Distances) in .tif raster format
│ ├── DistanceAOI.rds # Human wildlife conflict within areas of interest as .rds R-file
│ ├── HeatmapsBetweennessGlobal.rds # Consolidated heatmaps and betweenness maps across source areas as .rds R-files
│ ├── HeatmapsBetweennessLocal.rds # Consolidated heatmaps and betweenness maps per source area as .rds R-files
│ ├── HeatmapsGlobal.tif # Consolidated heatmaps (from 99_Heatmaps) across different source areas in .tif raster format
│ ├── HeatmapsLocal.tif # Consolidated heatmaps (from 99_Heatmaps) per source area in .tif raster
│ └── InterpatchConnectivityBootstrapped.rds # Estimates of interpatch connectivity as .rds R-files
└── README.md # This readme
Description of Data Files
01_RawData
Raw data was further processed
-
FLOODMAPS/YYYY.MM.DD.tif
- 0: Water
- 127: Cloud cover
- 255: Dryland
-
GLOBELAND/Water.tif
- 0: Dryland
- 1: Water
-
MERIT/Rivers.tif
- 0: Dryland
- 1: River (only rivers that are > 10 meters wide)
-
MODIS/Shrubs.tif
- 0-100: Percentage cover by shrubs
- 200: Invalid / water
-
MODIS/Trees.tif
- 0-100: Percentage cover by trees
- 200: Invalid / water
02_CleanData
Clean data was not further processed. Some of this data was obtained already cleaned from other data-sources (see below).
Shapefiles
Shapefiles are in EPSG:4326 projection and can be loaded using terra::vect()
.
-
Africa.shp:
- ID: Running number to identify individual polygons
- CODE: Country code
- COUNTRY: Name of the country
-
AreasOfInterest.shp:
- ID: Running number to identify individual polygons
- Name: Name of the area covered by each polygon
-
Cutlines.shp
(generated from theSourceAreas.R
script):- FID: Running number to identify individual cutlines
-
Faults.shp
:- id: Running number to identify individual fault lines
- Name: Name of the fault line
-
KAZA.shp
:- Name: Name of the area
-
MababeDepression.shp
:- Name: Name of the area
-
MajorRivers.shp
:- Name: Name of each river
-
MajorWaters.shp
:- Name: Name of each water source
- Category: Category of the water source
-
Protected.shp
:- Name: Name of the protected area
- IUCN: IUCN category of the protected area
- Country: Country in which the protected area lies
- Desig: Designation of the protected area (reclassified from the world database of protected areas into national park, protected area, forest reserve)
- Values: Numerical representation of the designation (3 = national park, 2 = protected area, 1 = forest reserve)
-
SourceAreas.shp
(generated from theSourceAreas.R
script):- ID: Running number to identify the source areas
- Type: Category of the source area
- Main: Areas within the main study area
- Buffer: Egression zone
-
Villages.shp
:- name: name of the village
- place: category of the village (Village or City)
Raster data
Raster data is in EPSG:4326 projection and can be loaded using terra::rast()
.
-
DistanceToHumans.tif
:- Values: Distance (in meters) to the nearest human influenced grid cell (roads, settlement, agriculture)
-
DistanceToWater.tif
:- Values: Distance (in meters) to the nearest grid cell covered by water for a minimum, average, and maximum flood scenario.
-
HumanInfluence.tif
:- Values: Relative strength of human influence (roads, settlements, and agriculture). The derivation of this layer is described by Hofmann et al., 2021.
-
ShrubCover.tif
:- Values: Proportion of shrub cover (0 to 1) for a minimum, average, and maximum flood scenario.
-
TreeCover.tif
:- Values: Proportion of tree cover (0 to 1) for a minimum, average, and maximum flood scenario.
-
WaterCover.tif
:- Values: Proportion of water cover (0 or 1) for a minimum, average, and maximum flood scenario.
R-Data
R-Data files are in .rds
format and can be loaded into R using readr::read_rds()
.
-
MovementModel.rds
(i.e. the dispersal model):- Model of the class
glmmTMB
, which requires theglmmTMB
R
-package to be installed and loaded. This model was used to predict selection scores. Covariates extracted during the simulation need to be scaled using the scaling parameters stored inScaling.rds
.
- Model of the class
-
GammaDistribution.rds
(used to propose random steps):- Shape: Estimated shape parameter of the step-length distribution
- Scale: Estimated scale parameter of the step-length distribution
-
Scaling.rds
(used to scale extracted covariates):- center: value by which covariates are shifted
- scale: value by which covariates are scaled
03_Results
The following files will all be generated when running through the analyses. Intermediate results are stored in the 99_...
folders, and only later consolidated into single files. The .rds
files give overviews over the exact simulation parameters and associated files. For instance, the file DispersalSimulations.rds
consolidates the individual simulations stored under 99_Simulations
.
Overviews / Consolidated files:
-
DispersalSimulation.rds
:- x: Simulated longitude
- y: Simulated latitude
- absta_: absolute turning angle (in radians)
- ta_: relative turning angle (in radians)
- sl_: step length (in meters)
- Timestamp: POSIXct timestamp
- BoundaryHit: Logical indicator if the simulated individual has hit a map boundary or not
- inactive: Logical indicator during what time a step is taken
- 0 = Inactive
- 1 = Active
- TrackID: Identifier of the simulated track
- SimID: Identifier of the simulation number
- FloodLevel: Flood scenario under which dispersal was simulated (either minimum or maximum flood, see
Water.tif
) - SourceArea: Source area from which individuals were simulated (i.e., the id of the
SourceArea.shp
) - StepNumber: Step number of the simulated step (1-2000)
-
Distance.rds
:- FloodLevel: Flood scenario (either minimum or maximum flood, see
Water.tif
) - SourceArea: Source area from which individuals were simulated (i.e., the id of the
SourceArea.shp
) - Filename: Filepath to where the raster files are stored.
- Level: Whether the metric is computed for a specific source area (Local) or if it was compute across all source areas (Global).
- FloodLevel: Flood scenario (either minimum or maximum flood, see
-
DistanceAOI.rds
:- FloodLevel: Flood scenario (either minimum or maximum flood, see
Water.tif
) - Name: Name of the area of interest, see
AreasOfInterest.shp
- Number: Density of trajectories within the area of interest
- SE: Standard error of the density of trajectories within the area of interest
- Percent: Percent change of the density from the minimum to maximum flood
- combined: Latex code for the number and standard error. This can be included in reports
- FloodLevel: Flood scenario (either minimum or maximum flood, see
-
HeatmapsBetweennessGlobal.rds
:- Steps: Number of steps after which the metric was computed (500, 1000, or 2000)
- FloodLevel: Flood scenario (either minimum or maximum flood, see
Water.tif
) - FilenameHeatmap: Filepath to where the raster file of the resulting heatmap is stored
- FilenameBetweenness: Filepath to where the raster file of the resulting betweenness map is stored
- Heatmap: Heatmap as raster file
- Betwenness: Betweenness map as raster file
-
HeatmapsBetweennessLocal.rds
:- Steps: Number of steps after which the metric was computed (500, 1000, or 2000)
- SourceArea: Source area from which individuals were simulated (i.e., the id of the
SourceArea.shp
) - FloodLevel: Flood scenario (either minimum or maximum flood, see
Water.tif
) - FilenameHeatmap: Filepath to where the raster file of the resulting heatmap is stored
- FilenameBetweenness: Filepath to where the raster file of the resulting betweenness map is stored
- Heatmap: Heatmap as raster file
- Betwenness: Betweenness map as raster file
-
InterpatchConnectivityBootstrapped.rds
:- SourceArea: Origin of the inter-patch connection (see
SourceAreas.shp
) - CurrentArea: Target of the inter-patch connection (see
SourceAreas.shp
) - FloodLevel: Considered flood level (min or max)
- Type: Type of the interpatch connection
- Dispersal = Movement between two main source areas
- Egression = Movement from a main source area to an egression zone
- DispersalSuccess: Number of simulated individuals moving between the specified areas
- SDDispersalSuccess: SD of simulated individuals moving between the specified areas
- DispersalDuration: Average minimum duration (in steps) it takes before simulated individuals successfully move from the source to the current area
- DispersalDuration: SD of the minimum duration (in steps) it takes before simulated individuals successfully move from the source to the current area
- SourceArea: Origin of the inter-patch connection (see
Unconsolidated files (these files should not be directly accessed):
-
99_Betweenness/Betweenness_Steps_FloodLevel.tif
:- Values: Betweenness values calculated from simulated dispersers across all source areas
-
99_Betweenness/Betweenness_Steps_SourceArea_FloodLevel.tif
:- Values: Betweenness values calculated from simulated dispersers separately for each source area
-
99_Betweenness/Distance_SourceArea_FloodLevel.tif
:- Values: Number of simulated coordinates within 500 meters of a human influenced grid cell computed separately for each source area
-
99_Heatmaps/Heatmap_Steps_FloodLevel.tif
:- Values: Traversal frequency calculated from simulated dispersers across all source areas
-
99_Heatmaps/Heatmap_Steps_SoufceArea_FloodLevel.tif
- Values: Traversal frequency calculated from simulated dispersers separately for each source area
-
99_Simulations/FloodLevel_SourceArea_Replicate.rds
- x: Simulated longitude
- y: Simulated latitude
- absta_: absolute turning angle (in radians)
- ta_: relative turning angle (in radians)
- sl_: step length (in meters)
- Timestamp: POSIXct timestamp
- BoundaryHit: Logical indicator if the simulated individual has hit a map boundary or not
- inactive: Logical indicator if the step is during wild dogs' active (1) or inactive (0) phase
- TrackID: Identifier of the simulated track
-
99_BetweennessGlobal.tif
- Values: Betweenness across all source areas
-
99_BetweennessLocal.tif
- Values: Betweenness for each source area separately
-
99_HeatmapsGlobal.tif
- Values: Betweenness across all source areas
-
99_HeatmapsLocal.tif
- Values: Heatmaps for each source area separately
Sharing/Access information
Links to other publicly accessible locations of the data:
- Description of the spatial data preparation: Hofmann et al., 2021
- Description of the dispersal model: Hofmann et al., 2023