Installation

Installing from CRAN:

#Install from CRAN:

# From CRAN would be
install.packages("suwo")

#load package
library(suwo)

Install the latest development version from GitHub:

install.packages("suwo", repos = c(
  'https://ropensci.r-universe.dev',
  'https://cloud.r-project.org'
))

#load package
library(suwo)

Basic workflow for obtaining nature media files

Obtaining nature media using suwo follows a basic sequence. The following diagram illustrates this workflow and the main functions involved:

Here is a description of each step:

Obtain metadata:

1. Queries regarding a species are submitted through one of the available query functions (query_repo_name()) that connect to five different online repositories (Xeno-Canto, Inaturalist, GBIF, Macaulay Library and WikiAves). The output of these queries is a data frame containing metadata associated with the media files (e.g., species name, date, location, etc, see below).

Curate metadata:

1. If multiple repositories are queried, the resulting metadata data frames can be merged into a single data frame using the merge_metadata() function.

2. Check for duplicate records in their datasets using the find_duplicates() function. Candidate duplicated entries are identified based on matching species name, country, date, user name, and geographic coordinates. User can double check the candidate duplicates and decide which records to keep, which can be done with remove_duplicates().

3. Download the media files associated with the metadata using the download_media() function.

4. Users can update their datasets with new records using the update_metadata() function.

Obtaining metadata: the query functions

The following table summarizes the available suwo query functions and the types of metadata they retrieve:

These are some example queries:

1. Images of Sarapiqui Heliconia (Heliconia sarapiquensis) from iNaturalist (we print the first 4 rows of each output data frame):

# Load suwo package
library(suwo)

Warning in library(suwo): package 'suwo' already present in search()

h_sarapiquensis <- query_inaturalist(species = "Heliconia sarapiquensis",
                                     format = "image")

✔ Obtaining metadata (30 matching records found) 🎉

head(h_sarapiquensis, 4)

2. Harpy eagles (Harpia harpyja) audio recordings from WikiAves:

WikiAves is protected by Cloudflare’s bot detection, so query_wikiaves() requires authentication cookies obtained beforehand with access_wikiaves(); see the query_wikiaves() section below for details:

# get cookies for WikiAves (only needs to be done once per session)
access_wikiaves()

ℹ Chrome not found on port 9333 -- launching it now... 

ℹ On Cloudflare challenge page -- waiting up to 30s for it to clear (solve manually in the Chrome window) 

✔ Cookies successfully retrieved from Chrome. 🌈

✔ Saved to the `wikiaves_cookies` environment variable for this session. 😸

h_harpyja <- query_wikiaves(species = "Harpia harpyja", format = "sound")

✔ Obtaining metadata (80 matching records found) 😸

head(h_harpyja, 4)

3. Common raccoon (Procyon lotor) videos from GBIF:

p_lotor <- query_gbif(species = "Procyon lotor", format = "video")

✖ No matching records found 😵

head(p_lotor, 4)

NULL

By default all query function return the 14 most basic metadata fields associated with the media files. Here is the definition of each field:

• repository: Name of the repository
• format: Type of media file (e.g., sound, photo, video)
• key: Unique identifier of the media file in the repository
• species: Species name associated with the media file (Note taxonomic authority may vary among repositories)
• date*: Date when the media file was recorded/photographed (in YYYY-MM-DD format or YYYY if only year is available)
• time*: Time when the media file was recorded/photographed (in HH:MM format)
• user_name*: Name of the user who uploaded the media file
• country*: Country where the media file was recorded/photographed
• locality*: Locality where the media file was recorded/photographed
• latitude*: Latitude of the location where the media file was recorded/photographed (in decimal degrees)
• longitude*: Longitude of the location where the media file was recorded/photographed (in decimal degrees)
• file_url: URL link to the media file (used to download media files)
• file_extension: Extension of the media file (e.g., .mp3, .jpg, .mp4)
• observation_url: URL link to the original observation page in the repository (used to check the original metadata and media file)

* Can contain missing values (NAs)

Users can also download all available metadata by setting the argument all_data = TRUE. These are the additional metadata fields, on top of the basic fields, that are retrieved by each query function:

The code above examplifies the most common use of query functions, which applies also to the function query_gbif(). The following sections provide more details on the two query functions that require special considerations: query_macaulay() and query_xenocanto().

query_macaulay()

Interactive retrieval of metadata

query_macaulay() is the only interactive function. This means that when users run a query the function opens a browser window to the Macaulay Library’s search page, where the users must download a .csv file with the metadata. Here is a example of a query for strip-throated hermit (Phaethornis striigularis) videos:

p_striigularis <- query_macaulay(species = "Phaethornis striigularis",
                                 format = "video")

ℹ A browser will open the macaulay library website. Save the .csv file ('export' button) to this directory: 
/home/m/Dropbox/R_package_testing/suwo/vignettes/ 

ℹ (R is monitoring for new CSV files. Press ESC to stop the function)

ℹ File:  
ML__2026-08-12T16-45_stther2_video.csv 

✔ 30 matching records found 🎊

Users must click on the “Export” button to save the .csv file with the metadata:

Note that for bird species the species name must be valid according to the Macaulay Library taxonomy (which follows the Clements checklist). For non-bird species users must use the argument taxon_code. The species taxon code can be found by running a search at the Macaulay Library’s search page and checking the URL of the species page. For instance, the taxon code for jaguar (Panthera onca) is “t-11032765”:

Once you have the taxon code, you can run the query as follows:

jaguar <- query_macaulay(taxon_code = "t-11032765",
                                 format = "video")

ℹ A browser will open the macaulay library website. Save the .csv file ('export' button) to this directory: 
/home/m/Dropbox/R_package_testing/suwo/vignettes/ 

ℹ (R is monitoring for new CSV files. Press ESC to stop the function)

ℹ File:  
ML__2026-08-12T16-45_t-11032765_video.csv 

✔ 37 matching records found 🥇

Here are some tips for using this function properly:

• Valid bird species names can be checked at suwo:::ml_taxon_code$SCI_NAME
• The exported csv file must be saved in the directory specified by the argument path of the function (default is the current working directory)
• The function will not proceed until the file is saved (press ESC to stop the function)
• Do not overwritte files : if the file is saved overwriting a pre-existing file (i.e. same file name) the function will not detect it
• Users must log in to the Macaulay Library/eBird account in order to access large batches of observations

After saving the file, the function will read the file and return a data frame with the metadata. Here we print the first 4 rows of the output data frame:

head(p_striigularis, 4)

repository	format	key	species	date	time	user_name	country	locality	latitude	longitude	file_url	file_extension	observation_url
Macaulay Library	video	660763140	Phaethornis striigularis	2026-06-30	10:28	Carlos Viquez	Panama	Camino a Cascada de Chorcha	8.411990	-82.21932	https://cdn.download.ams.birds.cornell.edu/api/v1/asset/660763140/	mp4	https://macaulaylibrary.org/asset/660763140
Macaulay Library	video	656764975	Phaethornis striigularis	2026-03-12	09:37	Anthony Marella	Colombia	Hotel Tinamú Birding Nature Reserve	5.064686	-75.59220	https://cdn.download.ams.birds.cornell.edu/api/v1/asset/656764975/	mp4	https://macaulaylibrary.org/asset/656764975
Macaulay Library	video	656313242	Phaethornis striigularis	2026-05-01	16:50	Jessy Lopez Herra	Costa Rica	Bijagua–Tapir Valley	10.716530	-85.01138	https://cdn.download.ams.birds.cornell.edu/api/v1/asset/656313242/	mp4	https://macaulaylibrary.org/asset/656313242
Macaulay Library	video	654011852	Phaethornis striigularis	2026-03-15	06:56	Bret Whitney	Mexico	Camino La Guadalupe–La Reforma	17.847230	-96.03763	https://cdn.download.ams.birds.cornell.edu/api/v1/asset/654011852/	mp4	https://macaulaylibrary.org/asset/654011852

Bypassing record limit

Even if logged in, a maximum of 10000 records per query can be returned. This can be bypassed by using the argument dates to split the search into a sequence of shorter date ranges. The rationale is that by splitting the search into date ranges, users can download multiple .csv files, which are then combined by the function into a single metadata data frame. Of course users must download the csv for each data range. The following code looks for photos of costa’s hummingbird (Calypte costae). As Macaulay Library hosts more than 30000 costa’s hummingbird records, we need to split the query into multiple date ranges:

# test a query with more than 10000 results paging by date
cal_cos <- query_macaulay(
  species = "Calypte costae",
  format = "image",
  dates = c(1976, 2020, 2022, 2024, 2025, 2026)
)

ℹ A browser will open the macaulay library website. Save the .csv file ('export' button) to this directory: 
/home/m/Dropbox/R_package_testing/suwo/vignettes/ 

ℹ (R is monitoring for new CSV files. Press ESC to stop the function)

• Query 1 of 5 (1976-2019):

ℹ File:  
ML__2026-08-12T16-46_coshum_photo.csv 

• Query 2 of 5 (2020-2021):

ℹ File:  
ML__2026-08-12T16-46_coshum_photo2.csv 

• Query 3 of 5 (2022-2023):

ℹ File:  
ML__2026-08-12T16-47_coshum_photo.csv 

• Query 4 of 5 (2024):

ℹ File:  
ML__2026-08-12T16-48_coshum_photo.csv 

• Query 5 of 5 (2025-2026):

ℹ File:  
ML__2026-08-12T16-48_coshum_photo2.csv 

✔ 41040 matching records found 🎊

Users can check at the Macaulay Library website how many records are available for their species of interest (see image below) and then decide how to split the search by date ranges accordingly so each sub-query has less than 10000 records.

query_macaulay() can also read metadata previously downloaded from Macaulay Library website. To do this, users must provide 1) the name of the csv file(s) to the argument files and 2) the directory path were it was saved to the argument path.

# opens (or reuses) a Chrome window, waits for the Cloudflare challenge to clear
access_wikiaves()

h_harpyja <- query_wikiaves(species = "Harpia harpyja", format = "sound")

#  set your Xeno-Canto key as environmental variable (run it on the console)
# Sys.setenv(xc_api_key = "YOUR_API_KEY_HERE")

# query Xeno-CAnto
t_tricolor <- query_xenocanto(species = "Thyroptera tricolor")

ℹ Obtaining metadata: 

✔ 6 matching sound files found 🥳

# we remove urls to avoid CRAN issues
head(t_tricolor[, grep("url", names(t_tricolor), invert = TRUE)], 4)

# assuming you already set your API key as in previous code block
a_hahneli <- query_xenocanto(
  species = 'sp:"Ameerega hahneli" cnt:"French Guiana" q:"A"')

ℹ Obtaining metadata: 

✔ 3 matching sound files found 🌈

# we remove urls to avoid CRAN issues
head(a_hahneli[, grep("url", names(a_hahneli), invert = TRUE)], 4)

Update metadata

The update_metadata() function allows users to update a previous query to add new information from the corresponding repository of the original search. This function takes as input a data frame previously obtained from any query function (i.e. query_reponame()) and returns a data frame similar to the input with new data appended.

To show case the function, we first query metadata of Eisentraut’s Bow-winged Grasshopper sounds from iNaturalist. Let’s assume that the initial query was done a while ago and we want to update it to include any new records that might have been added since then. The following code removes all observations recorded after 2024-12-31 to simulate an old query:

# initial query
c_eisentrauti <- query_inaturalist(species = "Chorthippus eisentrauti")

✔ Obtaining metadata (124 matching records found) 🌈

head(c_eisentrauti, 3)

# exclude new observations (simulate old data)
old_c_eisentrauti <-
  c_eisentrauti[c_eisentrauti$date <= "2024-12-31" | is.na(c_eisentrauti$date),
                ]

# update "old" data
upd_c_eisentrauti <- update_metadata(metadata = old_c_eisentrauti)

✔ Obtaining metadata (124 matching records found) 🥳

✔ 111 new entries found 🥳

# compare number of records
nrow(c_eisentrauti) == nrow(upd_c_eisentrauti)

[1] TRUE

The function update_metadata() only accepts metadata data frames originating from a single repository. For workflows involving multiple repositories, users can apply the function separately to each subset of metadata; see the examples in the function documentation for an illustration of this approach.

Combine metadata from multiple repositories

The merge_metadata() function allows users to combine metadata data frames obtained from multiple query functions into a single data frame. The function will match the basic columns of all data frames. Data from additional columns (for instance when using all_data = TRUE in the query) will only be combined if the column names from different repositories match. The function will return a data frame that includes a new column called source indicating the name of the original metadata data frame:

truf_xc <- query_xenocanto(species = "Turdus rufiventris")

ℹ Obtaining metadata: 

✔ 492 matching sound files found 🎊

truf_gbf <- query_gbif(species = "Turdus rufiventris", format = "sound")

✔ Obtaining metadata (788 matching records found) 🥳

! 2 observations do not have a download link and were removed from the results (included as an attribute called 'excluded_results'). 

truf_ml <- query_macaulay(species = "Turdus rufiventris",
                          format = "sound")

ℹ A browser will open the macaulay library website. Save the .csv file ('export' button) to this directory: 
/home/m/Dropbox/R_package_testing/suwo/vignettes/ 

ℹ (R is monitoring for new CSV files. Press ESC to stop the function)

ℹ File:  
ML__2026-08-12T16-50_rubthr1_audio.csv 

✔ 1473 matching records found 😀

# merge metadata
merged_metadata <- merge_metadata(truf_xc, truf_gbf, truf_ml)

# we remove urls to avoid CRAN issues
head(merged_metadata[, grep("url", names(merged_metadata), invert = TRUE)], 4)

Note that in such a multi-repository query, all query functions use the same search species (i.e. species name) and media format (e.g., sound, image, video). To facilitate this, users can set the global options species and format so they do not need to specify them in each query function:

# query at multiple repositories setting global options
options(species = "Turdus rufiventris", format = "sound")
truf_xc <- query_xenocanto() # assuming you already set your API key
truf_gbf <- query_gbif()
truf_ml <- query_macaulay()

# merge metadata
merged_metadata <- merge_metadata(truf_xc, truf_gbf, truf_ml)

Find and remove duplicated records

When compiling data from multiple repositories, duplicated media records are a common issue, particularly for sound recordings. These duplicates occur both through data sharing between repositories like Xeno-Canto and GBIF, and when users upload the same file to multiple platforms. To help users efficiently identify these duplicate records, suwo provides the find_duplicates() function. Duplicates are identified based on matching species name, country, date, user name, and locality. The function uses a fuzzy matching approach to account for minor variations in the data (e.g., typos, different location formats, etc).The output is a data frame with the candidate duplicate records, allowing users to review and decide which records to keep.

In this example we look for possible duplicates in the merged metadata data frame from the previous section:

# find duplicates
dups_merged_metadata <- find_duplicates(merged_metadata)

ℹ 686 potential duplicates found 

# look first 6 columns
head(dups_merged_metadata)

Note that the find_duplicates() function adds a new column called “duplicate_group” to the output data frame. This column assigns a unique identifier to each group of potential duplicates, allowing users to easily identify and review them. For instance, in the example above, records from duplicated group 95 belong to the same user, were recorded on the same date and time and in the same country:

subset(dups_merged_metadata, duplicate_group == 95)

In this case all the observations seem to refer to the same media file. Therefore only one copy is needed. Also note that the locality is not exactly the same for these records, but the fuzzy matching approach used by find_duplicates() was able to identify them as potential duplicates. By default, the criteria is set to country > 0.8 & locality > 0.5 & user_name > 0.8 & time == 1 & date == 1 which means that two entries will be considered duplicates if they have a country similarity greater than 0.8, locality similarity greater than 0.5, user_name similarity greater than 0.8, and exact matches for time and date (similarities range from 0 to 1). These values have been found to work well in most cases. Nonetheless, users can adjust the sensitivity based on their specific needs using the argument criteria.

Once users have reviewed the candidate duplicates, they can apply the remove_duplicates() function to eliminate unwanted duplicates from their metadata data frames. This function takes as input a metadata output data frame from find_duplicates():

# remove duplicates
dedup_metadata <- remove_duplicates(dups_merged_metadata)

ℹ 311 duplicates removed 

The output is a data frame similar to the input but without the specified duplicate records:

# look at first 4 columns of deduplicated metadata
# we remove urls to avoid CRAN issues
head(dedup_metadata[, grep("url", names(dedup_metadata), invert = TRUE)], 4)

When duplicates are found, one observation from each group of duplicates is retained in the output data frame. However, if multiple observations from the same repository are labeled as duplicates, by default (same_repo = FALSE) all of them are retained in the output data frame. This is useful as it can be expected that observations from the same repository are not true duplicates (e.g. different recordings uploaded to Xeno-Canto with the same date, time and location by the same user), but rather have not been documented with enough precision to be told apart. This behavior can be modified. If same_repo = TRUE, only one of the duplicated observations from the same repository will be retained in the output data frame (and all other excluded). The function will give priority to repositories in which media downloading is more straightforward (i.e. Xeno-Canto, GBIF), but this can be modified with the argument repo_priority.

Further identification of duplicates can be done by checking the file size of the media (ideally including it as a criterion in the ‘criteria’ argument). To obtain the file size, the media files need to be downloaded first with the function download_media() which returns the column file_size which can be used for this purpose.

Download media files

The last step of the workflow is to download the media files associated with the metadata. This can be done using the download_media() function, which takes as input a metadata data frame (obtained from any query function or any of the other metadata managing functions) and downloads the media files to a specified directory. For this example we will download images from a query on zambian slender Caesar (Amanita zambiana) (a mushroom) on GBIF:

# query GBIF for Amanita zambiana images
a_zam <- query_gbif(species = "Amanita zambiana", format = "image")

✔ Obtaining metadata (7 matching records found) 🥇

# create folder for images
out_folder <- file.path(tempdir(), "amanita_zambiana")
dir.create(out_folder)

# download media files to a temporary directory
azam_files <- download_media(metadata = a_zam, path = out_folder)

Downloading media files:

✔ All files were downloaded successfully 😸

The output of the function is a data frame similar to the input metadata but with two additional columns indicating the file name of the downloaded files (‘downloaded_file_name’) and the result of the download attempt (‘download_status’, with values “success”, ‘failed’, ‘already there (not downloaded)’ or ‘overwritten’).

Here we print the first 4 rows of the output data frame:

head(azam_files, 4)

… and check that the files were saved in the path supplied:

fs::dir_tree(path = out_folder)

/tmp/Rtmp8Kji8K/amanita_zambiana
├── Amanita_zambiana-GBIF3759537817-1.jpeg
├── Amanita_zambiana-GBIF3759537817-2.jpeg
├── Amanita_zambiana-GBIF4430877067-1.jpeg
├── Amanita_zambiana-GBIF4430877067-2.jpeg
├── Amanita_zambiana-GBIF4430877067-3.jpeg
├── Amanita_zambiana-GBIF5069132689-1.jpeg
├── Amanita_zambiana-GBIF5069132689-2.jpeg
├── Amanita_zambiana-GBIF5069132691.jpeg
├── Amanita_zambiana-GBIF5069132696-1.jpeg
├── Amanita_zambiana-GBIF5069132696-2.jpeg
├── Amanita_zambiana-GBIF5069132732.jpeg
└── Amanita_zambiana-GBIF5104283819.jpeg

Note that the name of the downloaded files includes the species name, an abbreviation of the repository name and the unique record key. If more than one media file is associated with a record, a sequential number is added at the end of the file name.

This is a multipanel plot of 6 of the downloaded images (just for illustration purpose):

# create a 6 pannel plot of the downloaded images
opar <- par(mfrow = c(2, 3), mar = c(1, 1, 2, 1))

for (i in 1:6) {
img <- jpeg::readJPEG(file.path(out_folder, azam_files$downloaded_file_name[i]))
  plot(
    1:2,
    type = 'n',
    axes = FALSE
  )
  graphics::rasterImage(img, 1, 1, 2, 2)
  title(main = paste(
    azam_files$country[i],
    azam_files$date[i],
    sep = "\n"
  ))
}

# reset par
par(opar)

Users can also save the downloaded files into sub-directories with the argument folder_by. This argument takes a character or factor column with the names of a metadata field (a column in the metadata data frame) to create sub-directories within the main download directory (suplied with the argument path). For instance, the following code searches/downloads images of longspined porcupinefish (Diodon holocanthus) from GBIF, and saves images into sub-directories by country (for simplicity only 6 of them):

# query GBIF for longspined porcupinefish images
d_holocanthus <- query_gbif(species = "Diodon holocanthus", format = "image")

✔ Obtaining metadata (4436 matching records found) 😸

! 1 observation does not have a download link and was removed from the results (included as an attribute called 'excluded_results'). 

# keep only JPEG records (for simplicity for this vignette)
d_holocanthus <- d_holocanthus[d_holocanthus$file_extension == "jpeg", ]

# select 6 random JPEG records
set.seed(666)
d_holocanthus <- d_holocanthus[sample(seq_len(nrow(d_holocanthus)), 6),]

# create folder for images
out_folder <- file.path(tempdir(), "diodon_holocanthus")
dir.create(out_folder)

# download media files creating sub-directories by country
dhol_files <- download_media(metadata = d_holocanthus,
                             path = out_folder,
                             folder_by = "country")

Downloading media files:

✔ All files were downloaded successfully 🌈

fs::dir_tree(path = out_folder)

/tmp/Rtmp8Kji8K/diodon_holocanthus
├── Australia
│   └── Diodon_holocanthus-GBIF1100594205.jpeg
├── Costa Rica
│   └── Diodon_holocanthus-GBIF6235469997.jpeg
├── Mexico
│   └── Diodon_holocanthus-GBIF5167665103.jpeg
├── Nicaragua
│   └── Diodon_holocanthus-GBIF2529270882.jpeg
├── Saint Vincent and the Grenadines
│   └── Diodon_holocanthus-GBIF6422591620.jpeg
└── United States of America
    └── Diodon_holocanthus-GBIF3112903569.jpeg

In such case the ‘downloaded_file_name’ column will include the sub-directory name:

dhol_files$downloaded_file_name

[1] "Mexico/Diodon_holocanthus-GBIF5167665103.jpeg"                          
[2] "Costa Rica/Diodon_holocanthus-GBIF6235469997.jpeg"                      
[3] "Nicaragua/Diodon_holocanthus-GBIF2529270882.jpeg"                       
[4] "United States of America/Diodon_holocanthus-GBIF3112903569.jpeg"        
[5] "Australia/Diodon_holocanthus-GBIF1100594205.jpeg"                       
[6] "Saint Vincent and the Grenadines/Diodon_holocanthus-GBIF6422591620.jpeg"

This is a multipanel plot of the downloaded images (just for fun):

# create a 6 pannel plot of the downloaded images
opar <- par(mfrow = c(2, 3), mar = c(1, 1, 2, 1))

for (i in 1:6) {
img <- jpeg::readJPEG(file.path(out_folder, dhol_files$downloaded_file_name[i]))
  plot(
    1:2,
    type = 'n',
    axes = FALSE
  )
  graphics::rasterImage(img, 1, 1, 2, 2)
  title(main = paste(
    substr(dhol_files$country[i], start = 1, stop = 14),
    dhol_files$date[i],
    sep = "\n"
  ))
}

# reset par
par(opar)

R version 4.6.1 (2026-06-24)
Platform: x86_64-pc-linux-gnu
Running under: Ubuntu 22.04.5 LTS

Matrix products: default
BLAS:   /usr/lib/x86_64-linux-gnu/blas/libblas.so.3.10.0 
LAPACK: /usr/lib/x86_64-linux-gnu/lapack/liblapack.so.3.10.0  LAPACK version 3.10.0

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
 [3] LC_TIME=es_CR.UTF-8        LC_COLLATE=en_US.UTF-8    
 [5] LC_MONETARY=es_CR.UTF-8    LC_MESSAGES=en_US.UTF-8   
 [7] LC_PAPER=es_CR.UTF-8       LC_NAME=C                 
 [9] LC_ADDRESS=C               LC_TELEPHONE=C            
[11] LC_MEASUREMENT=es_CR.UTF-8 LC_IDENTIFICATION=C       

time zone: America/Costa_Rica
tzcode source: system (glibc)

attached base packages:
[1] stats     graphics  grDevices utils     datasets 
[6] methods   base     

other attached packages:
 [1] knitr_1.51             suwo_0.2.2            
 [3] getPass_0.2-4          RecordLinkage_0.4-12.6
 [5] ff_4.5.3               bit_4.6.0             
 [7] RSQLite_3.53.3         DBI_1.3.0             
 [9] beepr_2.0              cli_3.6.6             
[11] styler_1.11.0          testthat_3.3.2        
[13] pkgstats_0.2.4         pkgcheck_0.1.3.013    
[15] maps_3.4.3             leaflet_2.2.3         
[17] devtools_2.5.2         usethis_3.2.1         
[19] crayon_1.5.3           jsonlite_2.0.0        
[21] RCurl_1.98-1.19        curl_7.1.0            
[23] remotes_2.5.0         

loaded via a namespace (and not attached):
  [1] RColorBrewer_1.1-3  warbleR_1.1.37     
  [3] rstudioapi_0.19.0   audio_0.1-12       
  [5] magrittr_2.0.5      rmarkdown_2.31     
  [7] farver_2.1.2        fs_2.1.0           
  [9] vctrs_0.7.3         memoise_2.0.1      
 [11] htmltools_0.5.9     signal_1.8-1       
 [13] sass_0.4.10         parallelly_1.48.0  
 [15] bslib_0.11.0        htmlwidgets_1.6.4  
 [17] desc_1.4.3          httr2_1.3.0        
 [19] covr_3.6.5          lubridate_1.9.5    
 [21] cachem_1.1.0        commonmark_2.0.0   
 [23] lifecycle_1.0.5     pkgconfig_2.0.3    
 [25] Matrix_1.7-6        R6_2.6.1           
 [27] fastmap_1.2.0       rcmdcheck_1.4.0    
 [29] future_1.75.0       digest_0.6.39      
 [31] spelling_2.3.2      ps_1.9.3           
 [33] rprojroot_2.1.1     pkgload_1.5.3      
 [35] textshaping_1.0.5   crosstalk_1.2.2    
 [37] timechange_0.4.0    httr_1.4.8         
 [39] compiler_4.6.1      proxy_0.4-29       
 [41] bit64_4.8.2         withr_3.0.3        
 [43] S7_0.2.2            backports_1.5.1    
 [45] praise_1.0.0        viridis_0.6.5      
 [47] pkgbuild_1.4.8      R.utils_2.13.0     
 [49] MASS_7.3-66         lava_1.9.2         
 [51] rappdirs_0.3.4      sessioninfo_1.2.4  
 [53] rjson_0.2.23        chromote_0.5.1     
 [55] tools_4.6.1         otel_0.2.0         
 [57] fftw_1.0-9          future.apply_1.20.2
 [59] lintr_3.4.0         nnet_7.3-21        
 [61] R.oo_1.27.1         glue_1.8.1         
 [63] callr_3.8.0         promises_1.5.0     
 [65] R.cache_0.17.0      grid_4.6.1         
 [67] rsconnect_1.10.1    checkmate_2.3.4    
 [69] generics_0.1.4      diffobj_0.3.8      
 [71] gtable_0.3.6        R.methodsS3_1.8.2  
 [73] class_7.3-24        websocket_1.4.4    
 [75] NatureSounds_1.0.5  data.table_1.18.4  
 [77] xml2_1.6.0          stringr_1.6.0      
 [79] pillar_1.11.1       later_1.4.8        
 [81] splines_4.6.1       dplyr_1.2.1        
 [83] lattice_0.22-9      survival_3.8-9     
 [85] tidyselect_1.2.1    pbapply_1.7-4      
 [87] gridExtra_2.3.1     svglite_2.2.2      
 [89] xfun_0.60           brio_1.1.5         
 [91] rex_1.2.2           stringi_1.8.7      
 [93] yaml_2.3.12         xopen_1.0.1        
 [95] kableExtra_1.4.1    evaluate_1.0.5     
 [97] codetools_0.2-20    dtw_1.23-3         
 [99] evd_2.3-7.1         tibble_3.3.1       
 [ reached 'max' / getOption("max.print") -- omitted 29 entries ]