Source code for image processing and statistical analysis, and the patch size data files for the manuscript:
Power laws and critical fragmentation in global forests. (2016) Saravia L.A. Doyle S.R. Bond-Lamberty B. bioRxiv 091751; doi: http://dx.doi.org/10.1101/091751
Zip Files:
Patch Sizes distribution by region, year and threshold from 2000 to 2015: these are binary files consisting on a vector of doubles, the R command used to read it is readBin(connection_file, "double", n = 10^8)
Africa_PatchSizes.zip
Northamerica_PatchSizes.zip
Southamerica_PatchSizes.zip
Eurasia_PatchSizes.zip
Oceania_PatchSizes.zip
Southasia_PatchSizes.zip
LargestPatchAnimations.zip Animations of the two largest patches for all the regions defined in the paper.
PercolationAnimations.zip Animation demonstrating percolation in a forest/non-forest model
Study_areas_definition.zip These files needed to be unziped in a folder and the location added to the script data_input_script.m file
R markdown files:
Threshold_sensitivity.Rmd # Fitting heavy tails to modis VCF, Smax & RSmax patch analysis.
Download_modis.Rmd # Download modis files MOD44B version 051
Map_study_areas.Rmd # Generate maps of study areas
Map_Max_patches.Rmd # Generate GIF animations of maximum patch dynamics
R Files
map_fun.r # Map raster functions
power_fun.r # Function to fit continuos heavy tail distributions
MATLAB Files
Extract patch sizes from bin, and other analysis not used in the paper
data_input_script.m
GCF_spatial_analyses.m
PYTHON Files
powlawfit.py : Call the Python powerlaw
package from command line
DATA Tables: csv and tab separated text files with model fits for the patch size distribution models.