Tree Equity and Avian Species Richness in Tucson, Arizona

Summary

This project uses community science observations of birds to investigate the relationship between tree equity scores (TES) and avian species richness in the semi-arid city of Tucson, Arizona, USA. Tree equity data was sourced from the City of Tucson open data portal, while avian occurrence data was acquired from the Global Biodiversity Information Facility (GBIF). Avian species richness was calculated by aggregating and summarizing all GBIF observations for each neighborhood and determining the total number of unique species observed. A Poisson regression within a generalized linear modeling framework was employed, using the log of neighborhood area as an offset term to account for variations in neighborhood size.

Folders

• data: raw data and coordinate bounds
• output: modified data, output plots
• scripts: (descriptions below)

Scripts

• query_gbif.R: Function to download records from GBIF for a set of coordinates and list of taxa
• download_data.R: Download GBIF observation data
• download_large_data.R: Processing of two large manually downloaded datasets
• extract_gbif_data.R: Extract individual neighborhood data files downloaded from GBIF
• assemble_data.R: Assemble GBIF files, add neighborhood ID, create rows if a neighborhood has no obs.
• neighborhood_richness.R: Summary of GBIF data and calculate neighborhood species richness
• spatial_join_model.py: Model parameters used in ArcGIS Pro for spatial join between GBIF and TES datasets
• urbanforest_dataset.R: Combined neighborhood avian species richness and tree equity values
• regression_analyses.R: Generalized Linear Model

Dependencies

• tidyverse: Data wrangling and visualization
• dplyr: Data wrangling (included within tidyverse, but sometimes best to load separately)
• ggplot2: Data visualization (included within tidyverse, but sometimes best to load separately)
• sf: Point filtering
• raster: Work with raster data
• rgbif: Search and retrieve data from GBIF