wildrtrax (pronounced wild-r-tracks) is an R package containing functions to help manage and analyze environmental sensor data to and from WildTrax. It helps to simplify the entire data life cycle by offering tools for data pre-processing, wrangling, and analysis.
You can install the most recent version of wildrtrax directly from this repository with:
# install.packages("remotes")
remotes::install_github("ABbiodiversity/wildrtrax")The development version of this package contains experimental features and recent fixes. It can be installed with:
remotes::install_github("ABbiodiversity/wildrtrax@development")The development version of the package will be periodically merged and will be reflected in the Changelogs.
All functions begin with a wt_* prefix. Column names and metadata align with the WildTrax infrastructure. The goal is to follow the work flow of pre-processing, linking with WildTrax, download and analysis.
Explore data through Data Discover. No login required to access public data! Use wt_auth() to login to see data you have privileges and membership to.
library(wildrtrax)
library(sf)
# Search for public data without a boundary
wt_dd_summary(sensor = 'ARU', species = 'White-throated Sparrow')
# Apply an area of interest. Define a polygon or use a bbox from sf::st_bbox
my_aoi <- list(
c(-113.96068, 56.23817),
c(-117.06285, 54.87577),
c(-112.88035, 54.90431),
c(-113.96068, 56.23817)
)
wt_dd_summary(sensor = 'ARU', species = 'White-throated Sparrow', boundary = my_aoi)
# Alberta bounding box
abbox <- read_sf("...shp") |>
filter(Province == "Alberta") |> # A shapefile of Alberta
st_transform(crs = 4326) |>
st_bbox()
wt_dd_summary(sensor = "ARU", species = "White-throated Sparrow", boundary = abbox)Download data and run and a single-season single-species occupancy analysis. Consult APIs and Acoustic data wrangling for more information.
library(wildrtrax)
library(tidyverse)
# Login with your OAuth
Sys.setenv(WT_USERNAME = "*****", WT_PASSWORD = "*****")
# Authenticate to WildTrax
wt_auth()
# Get a project id
projects <- wt_get_projects("ARU") |>
filter(project == "Ecosystem Health 2023") |>
pull(project_id) |>
wt_download_report(sensor_id = "ARU", reports = "main")
# Format to occupancy for OVEN
dat.occu <- wt_format_occupancy(data = raw_data, species="OVEN", siteCovs=NULL)
# Run the model
unmarked::occu(~ 1 ~ 1, dat.occu)Conduct some pre-processing on various types of acoustic data. See more in Acoustic pre-processing.
library(wildrtrax)
library(tidyverse)
# Scan files and filter results
my_files <- wt_audio_scanner(path = ".", file_type = "wav", extra_cols = TRUE) |>
dplyr::mutate(hour = as.numeric(format(recording_date_time, "%H"))) |>
dplyr::filter(julian == 176, hour %in% c(4:8))
# Run acoustic indices and LDFCs
wt_run_ap(x = my_files, output_dir = paste0(root, 'ap_outputs'), path_to_ap = '/where/you/store/AP')
wt_glean_ap(x = my_files, input_dir = ".../ap_outputs", purpose = "biotic", include_ldfcs = TRUE)Evaluate the performance of BirdNET on a project, and search for false negatives missed by human taggers. See Classifiers Tutorial for more information.
library(wildrtrax)
library(tidyverse)
# OAuth logins only. Google OAuth2 will be supported soon.
Sys.setenv(WT_USERNAME = "*****", WT_PASSWORD = "*****")
# Authenticate to WildTrax
wt_auth()
my_reports <- wt_download_report(project_id = 1144, sensor_id = "ARU", reports = c("main", "ai"))
eval <- wt_evaluate_classifier(data = my_reports, resolution = "task", remove_species = TRUE, thresholds = c(0.01,0.99))
e1 <- wt_classifier_threshold(eval)
# Select the lowest threshold across classifiers and find additional species (false negatives)
wt_additional_species(my_reports, remove_species = TRUE, threshold = min(e1$threshold), resolution="task")The ultimate pipeline for your camera data work flows. See Camera data wrangling for more information.
library(wildrtrax)
library(tidyverse)
# OAuth logins only. Google OAuth2 will be supported soon.
Sys.setenv(WT_USERNAME = "*****", WT_PASSWORD = "*****")
# Authenticate to WildTrax
wt_auth()
# Get a camera project
raw <- wt_get_projects("CAM") |>
filter(project == "ABMI Ecosystem Health 2014") |>
pull(project_id) |>
wt_download_report(sensor_id = "CAM", reports = "main")
# Get individual species detections using various thresholds and time ranges
individual_detections <- wt_ind_detect(x = raw, threshold = 30, units = "minutes")Format tags from Kaleidoscope for a WildTrax project. Download data from a project into an NABAT acceptable format.
library(wildrtrax)
library(tidyverse)
input <- ".../bat.csv" # A Kaleidoscope output file
output <- ".../bats" # A folder to store the formatted csv
wt_kaleidoscope_tags(input, output, freq_bump = T)
## Authenticate to WildTrax, then upload the tags to a WildTrax project
wt_auth()
# Get a project id
projects <- wt_get_projects("ARU") |>
filter(project == "A bat project") |> # Enter your bat project name here
pull(project_id) |>
wt_download_report(sensor_id = "ARU", reports = "main")Download combined and formatted acoustic and point count data sets together.
library(wildrtrax)
library(tidyverse)
# An ARU project
an_aru_project <- wt_download_report(project_id = 620, sensor_id = 'ARU', reports = "main", )
# An ARU project as point count format
aru_as_pc <- wt_download_report(project_id = 620, sensor_id = 'PC', reports = "main", )To report bugs, request additional features, or get help using the package, please file an issue.
We encourage ongoing contributions and collaborations to improve the package into the future. The Alberta Biodiversity Monitoring Institute provides ongoing support, development and funding.
This R package is licensed under MIT license©2025 Alberta Biodiversity Monitoring Institute.
Please note that wildrtrax is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.