Detect stationarity in animal movement data

This is a test repository. Our final contribution to the EMAC23 Coding Challenge can be found in this repository.

EMAC23 Coding Challenge: https://www.moveapps.org/#news

Use Cases: https://www.moveapps.org/assets/EMAC23_CodingChallenge_Details2.eeb3420c.pdf -> Challenge 1: Stationarity

Assume that stationarity is definite, e. g.:

  • animal is deceased
  • animal has dropped tag

Data used:

  • Kruger African Buffalo, GPS tracking, South Africa (Movebank ID 1764627)
  • Straw-colored fruit bats (Eidolon helvum) in Africa 2009-2014 (Movebank ID 404939825)
  • Migration timing in white-fronted geese (data from Klzsch et al. 2016) (Movebank ID 133992043)

The used datasets are transformed to RDS-format using the MoveApps platform.

Move the downloaded transformed data into a directory called data outside of this repository.

The script stationarity_analysis.R

  • loads a dataset,
  • retrieves some general characteristics of the loaded dataset,
  • processes the given data for each individual for a given time period,
  • computes aggregated distances for each individual,
  • plots the aggregated distances over time for a selected individual
  • plots the last coordinates on a map for a selected individual
  • and finally creates a movement summary table for all individuals in the given dataset.

Below are a time series and a map example for an individual from the buffalo dataset:

time_series example

map example