14  Publish your data

This tutorial explains how to create a GeoLocator Data Package (GeoLocator DP) from a GeoPressureTemplate project, upload it to Zenodo, and validate the draft before publication.

Note

For an introduction to this format, visit the GeoLocator DP website. The standard and the Master GeoLocator Data Package are described in Nussbaumer et al. (2026), Scientific Data.

14.1 Install GeoLocatoR

GeoLocatoR website

We use the GeoLocatoR package to create and validate the data package.

# install.packages("pak")
pak::pkg_install("GeoPressure/GeoLocatoR")
library(GeoLocatoR)

14.2 Step 1. Create the package from GeoPressureTemplate

Read your project and build a geolocator-dp object.

pkg <- read_geopressuretemplate()
Warning: ! Dropped 1 column from "edges" not defined by its schema: name.
ℹ "edges" declares `fieldsMatch: superset`, which allows only the columns its
  schema defines.

By default, read_geopressuretemplate() reads data/interim/*.RData first and falls back to raw data for tags without interim files.

  • Use from to force a specific source ("interim" or "raw-tag").
  • Prefix a file/folder with _ to exclude a tag temporarily.

Inspect the package summary:

print(pkg)
## 
## ── A GeoLocator Data Package `pkg` (v1.3) ──────────────────────────────────────
## 
## ── Metadata
## Note: All green texts are fields of `pkg` which can be accessed with
## `pkg$field`.
## • temporal: "2017-06-20" to "2022-06-11"
## • taxonomic: "Acrocephalus arundinaceus" and "Limnothlypis swainsonii"
## • numberTags: tags: 2, measurements: 2, light: 1, pressure: 2, activity: 1,
##   temperature_external: 2, paths: 2
## 
## ── Resources: (7 resources)
## • `tags()` (n=2)
## • `observations()` (n=4)
## • `measurements()` (n=336,562)
## • `staps()` (n=38)
## • `twilights()` (n=632)
## • `paths()` (n=418)
## • `edges()` (n=308)

14.3 Step 2. Complete tags and observations

At this point, tags and observations will certaintly still be missing key metadata (for example ringing information).

The easiest way to fix this is to export the tables, edit them manually and then save them back.

write.csv(tags(pkg), file = "data/tags.csv", row.names = FALSE)
write.csv(observations(pkg), file = "data/observations.csv", row.names = FALSE)

Rebuild and validate after editing:

pkg <- read_geopressuretemplate()
## ⠙ Reading 1/2 interim tags.
## ✔ Read 2 interim tags.
## 
## Warning: ! Dropped 1 column from "edges" not defined by its schema: name.
## ℹ "edges" declares `fieldsMatch: superset`, which allows only the columns its
##   schema defines.
## ✔ Reading tags from './data/tags.csv'.
## ✔ Reading observations from './data/observations.csv'.
validate_gldp(pkg)
## 
## ── Check Profile
## ✔ Profile checks passed.
## 
## ── Check Resources tags
## ✔ Table tags is consistent with the schema.
## 
## ── Check Resources observations
## ✔ Table observations is consistent with the schema.
## 
## ── Check Resources measurements
## ✔ Table measurements is consistent with the schema.
## 
## ── Check Resources staps
## ✔ Table staps is consistent with the schema.
## 
## ── Check Resources twilights
## ✔ Table twilights is consistent with the schema.
## 
## ── Check Resources paths
## ✔ Table paths is consistent with the schema.
## 
## ── Check Resources edges
## ✔ Table edges is consistent with the schema.
## 
## ── Check Foreign Keys
## ✔ Foreign key checks passed.
## 
## ── Check Coherence
## ✔ Coherence checks passed.
## 
## ── Check Metadata
## ! pkg$title is missing or empty.
## ! Missing roles in pkg$contributors.
## ! Expected pkg$record_type to be "dataset"; got NA.
## ! No relatedIdentifiers present. Add any resources related to the data (e.g. papers, project pages, derived datasets, etc.).
## ! Missing selected community <https://zenodo.org/communities/geolocator-dp/>.
## ! Missing or invalid pkg$codeRepository; expected a GitHub repository URL.
## 
## ── Final: ──────────────────────────────────────────────────────────────────────
## ✔ Package validation succeeded.

14.4 Step 3. Run visual checks

In this final stage, we check the coverage, tag count, and deployment/retrieval consistency to avoid outliers or missing information. To do so, plot each variable and check that (1) all the ring numbers are present, (2) all measurements are correct and no information is missing and (3) trajectories are correct.

plot(pkg, "ring")

plot(pkg, "coverage")
Warning: Removed 2 rows containing missing values or values outside the scale range
(`geom_tile()`).

plot(pkg, "map")

14.5 Step 4. Write the package files

Now that we have checked the data, we can export pkg into the csv file that will be uploaded on Zenodo.

write_gldp(pkg, "data/datapackage")

14.6 Step 5. Create the Zenodo draft record

Create a new draft on Zenodo, complete all metadata fields, BUT, do not submit it just yet.

WarningMandatory checklist

Complete all items below. If one or more are missing, the record may be returned for revision and not accepted in the community.

Review the GeoLocator DP curation policy before submission.

14.7 Step 6. Validate the Zenodo draft

Before submitting, read the draft back and validate exactly what Zenodo contains.

Reading a draft requires a Zenodo API token. Create one at Zenodo token settings and save it in your keyring. key_set() asks for the token, so it is not saved in your R history:

keyring::key_set("ZENODO_TOKEN")
pkg <- read_zenodo(
  "10.5072/zenodo.472302",
  draft = TRUE, # Draft record
  sandbox = TRUE # Remove if using production Zenodo
)
Important

Run the official validation check on the draft package:

validate_gldp(pkg)

If validation reports errors, fix the metadata/files and run the check again until valid.

14.8 Step 7. Submit

Once all checks are valid and metadata are complete, submit the Zenodo record! 🎉 A curator will review the data and get back to you if any revision is required.