Processing and Analysing Animal Trajectories.

move2

The goal of the move2 package is to facilitate handling movement data by creating the move2 class that extents the functionality of the sf package. It facilitates import of animal movement data from movebank. The move2 package is designed as a successor to the move package, but improves in speed and functionality by being redesigned from the ground up. More documentation can be found on the website.

Installation

The most recent released version of move2 can be installed directly from CRAN:

install.packages("move2")

You can also install the development version of move2 like so:

devtools::install_git('https://gitlab.com/bartk/move2.git')

Example

By using units it is possible to track changes through analysis, note that ggplot directly has the correct units on the y axis

require(units)
require(dplyr)
require(ggplot2)
require(move2)
require(rnaturalearth)
track <- movebank_download_study("Galapagos Albatrosses",
  sensor_type_id = "gps"
)
track
#> A <move2> with `track_id_column` "individual_local_identifier" and
#> `time_column` "timestamp"
#> Containing 28 tracks lasting on average 37.1 days in a
#> Simple feature collection with 16414 features and 18 fields (with 386 geometries empty)
#> Geometry type: POINT
#> Dimension:     XY
#> Bounding box:  xmin: -91.3732 ymin: -12.79464 xmax: -77.51874 ymax: 0.1821983
#> Geodetic CRS:  WGS 84
#> # A tibble: 16,414 × 19
#>   sensor_type_id individual_local_identifier eobs_battery_voltage
#>          <int64> <fct>                                       [mV]
#> 1            653 4264-84830852                               3686
#> 2            653 4264-84830852                               3701
#> 3            653 4264-84830852                               3701
#> 4            653 4264-84830852                               3691
#> # ℹ 16,410 more rows
#> # ℹ 16 more variables: eobs_fix_battery_voltage [mV],
#> #   eobs_horizontal_accuracy_estimate [m], eobs_key_bin_checksum <int64>,
#> #   eobs_speed_accuracy_estimate [m/s], eobs_start_timestamp <dttm>,
#> #   eobs_status <ord>, eobs_temperature [°C], eobs_type_of_fix <fct>,
#> #   eobs_used_time_to_get_fix [s], ground_speed [m/s], heading [°],
#> #   height_above_ellipsoid [m], timestamp <dttm>, visible <lgl>, …
#> First 4 track features:
#> # A tibble: 28 × 52
#>   deployment_id  tag_id individual_id animal_life_stage attachment_type
#>         <int64> <int64>       <int64> <fct>             <fct>          
#> 1       2911170 2911124       2911090 adult             tape           
#> 2       2911150 2911126       2911091 adult             tape           
#> 3       2911167 2911127       2911092 adult             tape           
#> 4       2911168 2911129       2911093 adult             tape           
#> # ℹ 24 more rows
#> # ℹ 47 more variables: deployment_comments <chr>, deploy_on_timestamp <dttm>,
#> #   duty_cycle <chr>, deployment_local_identifier <fct>,
#> #   manipulation_type <fct>, study_site <chr>, tag_readout_method <fct>,
#> #   sensor_type_ids <chr>, capture_location <POINT [°]>,
#> #   deploy_on_location <POINT [°]>, deploy_off_location <POINT [°]>,
#> #   individual_comments <chr>, individual_local_identifier <fct>, …
ggplot() +
  geom_sf(data = ne_coastline(returnclass = "sf", 10)) +
  theme_linedraw() +
  geom_sf(data = track) +
  geom_sf(data = mt_track_lines(track), aes(color = `individual_local_identifier`)) +
  coord_sf(
    crs = sf::st_crs("+proj=aeqd +lon_0=-83 +lat_0=-6 +units=km"),
    xlim = c(-1000, 600),
    ylim = c(-800, 700)
  )
#> In total 386 empty location records are removed before summarizing.

track %>%
  ggplot() +
  geom_point(aes(
    x = `eobs_used_time_to_get_fix`,
    y = (`eobs_battery_voltage` - `eobs_fix_battery_voltage`) / `eobs_used_time_to_get_fix`
  )) +
  xlab("Time to fix") +
  ylab("Voltage drop rate")

Other packages

Several other packages exist for dealing with movement data. For an overview see the CRAN task view. Here we make a quick comparison to some other packages that define movement datasets and the differences to move2:

• move: The move package is based on sp and therefore the S4 classes. This class system is very capable but it sometimes results in the data being more difficult to handle. Furthermore as sf improves on spmove2 will be able to improve in speed.
• sftrack: Does not include trajectory information. Furthermore the sft_group attribute is quite large (in memory due to being a character) and relatively slow.
• amt/trajr/ctmm: These do not extend a spatial class meaning other spatial functions do not work on these trajectories. Furthermore these objects do not keep track of trajectory level information (e.g. sex, data of birth & capture location).

Compared to the other packages move2 tries to be flexible enough that it is easy to use in other software but still retains the formal properties and all meta data of trajectories. Therefore we avoid more complicated data structures (e.g. Kranstauber et al, 2011).

Future

Currently the move2 package focuses on core functionality, ideas for additional packages exist:

• moveAcc/moveIMU: probably using vctrs/pillar for storing bursts of acceleration or orientation data.
• moveUd: for example Brownian bridges
• moveSim: for simulation trajectories
• moveShiny: for simple quick apps using shiny modules for visualizing and exploring trajectories

Additional functionality in move2

There are ideas for additional functionality they are not directly implemented but recorded here to keep them on the radar. Specific suggestions or information what is used are welcome.

• Filter functions we might want to implement:
  • filter_duplicate_locations_minimal_travel
  • filter_minimal_travel: Minimally x meter travel between locations
  • filter_minimal_distance: Minimally x meter distance between locations
  • filter_outlier_quick_return: Combination of ~180 degree turns, high speed and comparable distances?
• Import data from PostgreSQL database

Development

By installing the package todor it is possible to find all to-dos as markers through the rstudio addins menu, or run todor::todor_package_addin(). Furthermore we use precommit to validate code before committing. Testing is done through testthat.

Dependencies

depgraph::plot_dependency_graph(".", suggests = FALSE)