R Files — occupancy modeling pipeline

R scripts and R Markdown notebooks for processing camera-trap photo identification data and fitting multi-city occupancy models (including analyses built around the autoOcc package).

Scripts are numbered to reflect the usual data → models → summaries → figures order. Paths are relative to this folder: inputs are expected under ../Input Data/, and outputs under ../Output/, ../Species Occupancy/, or paths noted in each file.

Requirements

• R (4.x recommended)
• Core packages used across the pipeline include dplyr, tidyr, ggplot2, lubridate, purrr, sf, terra, stringr / forcats, and others as referenced in each notebook.
• Occupancy modeling (3_OccupancyModeling.Rmd): install autoOcc via devtools::install_github("mfidino/autoOcc").
• Species richness (2_SpeciesRichness.Rmd): iNEXT, googlesheets4 (if used), terra.
• Site clustering / impervious extraction steps use igraph, sf, and terra (1_SiteDataActivity.Rmd).

Install missing packages in R with install.packages(...) before knitting or sourcing.

Pipeline overview

Step	File	Role
1	1_CleanForOccupancy.Rmd	Parse all city photo database CSVs (commas inside {...} fields), keep key columns, assign seasons, and output one consolidated multi_city_data_cleaned.csv.
1	1_SiteDataActivity.Rmd	Build all-city site activity and site metadata products: standardize active histories, cluster nearby cameras (~150 m), and extract imperviousness from NLCD raster buffers. Outputs all_city_active_dates.csv and camera_sites_with_impervious.csv.
2	2_ConvertToOccupancy.Rmd	Build species-level occupancy inputs and reports under ../Species Occupancy/ (also references running models in some versions—check the notebook).
2	2_SpeciesRichness.Rmd	Species richness analyses (e.g. iNEXT) parallel to strict occupancy prep.
3	3_OccupancyModeling.Rmd	Parameterized occupancy models: set params (path, sp_name, range, by, city, conf) when rendering. Writes per-city / species outputs under the path you pass.
4	4_SummarizeModelResults.Rmd	Aggregate AIC / model comparison tables from ../Output/ using species and city lists from inputs.
5	5_OccupancyModelResults.Rmd	Explore and summarize parameter estimates (example workflow includes PACA-focused filters).
5	5_OccupancyAllParams.Rmd	Cross-threshold / multi-species parameter summaries (e.g. model m2).
5	5_GenerateOccupancyFigures.Rmd	Pull prediction CSVs and build figure-ready summaries for chosen cities, species, and confidence thresholds.

Run earlier numbered steps before later ones unless you are only refreshing a self-contained notebook (e.g. figures from existing Output/).

Data layout (typical)

• ../Input Data/ — Raw and intermediate CSVs (photo_database_fin_*.csv, cleaned variants, active-date files, NLCD raster, etc.). Filenames match those referenced in each script.
• ../Output/ — Model results, parameters, and summaries produced by 3_, 4_, and 5_ steps (often organized by city and confidence).
• ../Species Occupancy/ — Occupancy report outputs from 2_ConvertToOccupancy.Rmd when that path is used.

Adjust the filepath / input_path variables at the top of each notebook if your directory layout differs.