Daniela Linero, MSc - National Audubon Society
Jorge Velásquez, Ph.D. - National Audubon Society
Camilo Correa-Ayram, Ph.D. - Departamento de Ecología y Territorio, Universidad Javeriana
\
Colombia's ecosystems are home to the highest richness of bird species worldwide, yet they are being lost rapidly due to the growing threats of climate and land-use change. The National Audubon Society seeks to counteract the effects of these threats by supporting the establishment of a representative and ecologically connected system of protected areas in Colombia. The purpose of this project is to identify, for the first time on a national scale, the most critical places for conserving and restoring the connectivity among Colombian protected areas. We will harness the power of Azure cloud computing to build high-resolution species distribution models for ten priority birds based on open-source occurrence data and environmental GIS layers. With the resulting maps, we will apply least-cost and circuit theory algorithms to model large-scale connectivity priorities among national and sub-national protected areas.
¦
+---data
¦ +--- 01_cleanOccurrences : Species occurrences obtained from eBird
¦ +--- 02_SDMs : Raw and processed environmental data
¦ +--- species_distributions : Shapefiles of the species distributions
¦
+---outputs
¦ +--- 01_cleanOccurrences : Clean occurrences
¦ +--- 02_SDMs : Results of species distribution models under different frameworks
¦
+---scripts
¦ +--- 01_cleanOccurrences.R : R script to clean occurrences
¦ +--- 01_cleanOccurrences_markdown.html : Pdf file describing cleaning procedure
¦ +--- 01_cleanOccurrences_markdown.Rmd : Markdown file describing cleaning procedure
¦ +--- 01_cleanOccurrences_markdown.tex : Latex file associated with markdown
¦ +--- 02_Build_sampling_probability_map.R : R script to build sampling probability map based on birds occurrences
¦ +--- 02_Prepare_ESA_data.R : R script to prepare land cover data for SDMs
¦ +--- 02_SDMs_correcting_samplingBias.R : SDM models under different frameworks to correct for sampling bias
¦ +--- 02_SDMs_correcting_samplingBias_loop.R : Loop that accelerates SDM models to correct for sampling bias
¦ +--- 02_SDMs_models.R : SDM models following the simplest steps of Wallace
¦ +--- 02_SDMs_models_ESA.R : SDM models incorporating land cover data
¦ +--- 02_SDMs_thinning.R : Script to do spatial thinning of occurrences
¦ +--- 01_cleanOccurrences.R : Script to clean occurrences
¦ +--- 03_composite_corridor_maps.R : Reclassify corridors for each species to create composite maps
¦ +--- 03_creation_resistance_surfaces.R : Create movement resistance maps for each focal species
¦ +--- 03_Selecting_cores_&_study_areas.R : Selecting protected areas to connect for each species
¦
¦
¦ README.md : Description of the repository
¦ Connectivity.Rproj : RStudio project file
Feel free to email me at [email protected]