Exercise A: Segmentation

With the skills from Input: Segmentation you can now implement the segmentation algorithm described in Laube and Purves (2011) to either your own movement data or to a different wild boar using different sampling intervals.

Once you have completed this prepartions, commit your changes with a meaningful commit message. Before committing, make sure your location data is ignored! Then, test your connection to Github by pushing your changes to your remote repository.

Task 1: Calculate distances

Now, you can Step a): Specify a temporal window v and Step b): Measure the distance to every point within v, which you had used with sabi, on on your own movement data or to a different wild boar using different sampling intervals.

Once you have completed the task, commit your changes with a meaningful commit message and test your connection to Github by pushing your changes to your remote repository.

Task 2: Specify and apply threshold d

After calculating the Euclidean distances to positions within the temporal window v in task 1, you can explore these values (we stored them in the column stepMean) using summary statistics (histograms, boxplot, summary()): This way we can define a reasonable threshold value to differentiate between stops and moves. There is no “correct” way of doing this, specifying a threshold always depends on data as well as the question that needs to be answered. In this exercise, use the mean of all stepMean values.

Store the new information (boolean to differentiate between stops (TRUE) and moves (FALSE)) in a new column named static.

Commit your changes with a meaningful commit message.

Task 3: Visualize segmented trajectories

Now visualize the segmented trajectory spatially. Just like last week, you can use ggplot with geom_path(), geom_point() and coord_equal(). Assign colour = static within aes() to distinguish between segments with “movement” and without.

Commit your changes with a meaningful commit message.

Task 4: Segment-based analysis

In applying Laube and Purves (2011), we’ve come as far as step b) in Figure 16.1. In order to complete the last steps (c and d), we need a unique ID for each segment that we can use as a grouping variable. The following function does just that (it assigns unique IDs based on the column static which you created in Task 2). You will learn about functions next week. For now, just copy the following code chunk into your script and run it.

rle_id <- function(vec) {
    x <- rle(vec)$lengths
    as.factor(rep(seq_along(x), times = x))
}

You can use the newly created function rle_id to assign unique IDs to subtrajectories (as shown below). Visualize the moving segments by colourizing them by segment_ID. Then use segment_ID as a grouping variable to determine the segments duration and remove short segments (e.g. segments with a duration < 5 Minutes)

your_data_frame <- your_data_frame |>
    mutate(segment_id = rle_id(static))

Commit your changes with a meaningful commit message.