Simulating individual movement in fish

Data on the positions and movement statistics of 60 individual three-spined stickleback (Gasterosteus aculeatus) fish are available from the University of Lincoln Repository, https://eprints.lincoln.ac.uk/id/eprint/55426/ (as an object in R data format, fishData.rds).

Load the object

fishData <- readRDS(file.choose())

Look at its structure. It is a list of length 60 (i.e., it contains data for 60 fish)...

length(fishData)

[1] 60

... and for each fish it contains the following elements (e.g., for fish 1)

str(fishData[[1]])

List of 6
 $ smoothed.position:List of 3
  ..$ x: num [1:36000] 81.5 74.4 67.8 61.9 57.1 ...
  ..$ y: num [1:36000] 41.9 39 36.7 35.6 36 ...
  ..$ z: num [1:36000] 37.7 35.3 33.2 31.6 30.9 ...
 $ rho.s            : num 0.858
 $ shape            : num 2.7
 $ rate             : num 0.775
 $ kappa            :List of 2
  ..$ k: num 2.83
  ..$ d: num 1.98
 $ beta             :List of 2
  ..$ k: num 0.452
  ..$ d: num 2.26

These elements are as follows:

• smoothed.position: the [x,y,z] positions of the fish after smoothing.
• rho.s: the Spearman's rank correlation coefficient describing step length autocorrelation at lag 1.
• shape, rate: the parameters of the best-fitting gamma distribution for step length.
• kappa, beta: the coefficient (k) and exponent (d) of the power-law relationship between step length and the concentration (κ) and ovalness (β) parameters of the best-fitting Kent distributions.

Example usage

plot(fishData[[1]]$smoothed.position$x[1:360], fishData[[1]]$smoothed.position$y[1:360], xlab="x (mm)", ylab="y (mm)", asp=1)