Skip to contents

run_simulation

Source: R/run_simulation.R
run_simulation.Rd

Run simulation.

Usage

run_simulation(
  seafloor,
  fishpop,
  nutrients_input = 0,
  movement = "rand",
  parameters,
  torus_diffusion = TRUE,
  max_i,
  min_per_i,
  seagrass_each = 1,
  save_each = 1,
  burn_in = 0,
  return_burnin = TRUE,
  to_disk = FALSE,
  path_disk = NULL,
  verbose = TRUE
)

Arguments

seafloor

Data.frame with seafloor.

fishpop

Data.frame with fish population.

nutrients_input

Vector with nutrient input for each time step.

movement

String specifying movement algorithm.

parameters

List with all model parameters.

torus_diffusion

Logical if diffusion uses torus

max_i

Integer with maximum number of simulation time steps.

min_per_i

Integer to specify minutes per i.

seagrass_each

Integer how often (each i * x) seagrass dynamics will be simulated.

save_each

Numeric how often data should be saved.

burn_in

Numeric with time steps used to burn in (no fish).

return_burnin

If FALSE, all time steps < burn_in are not saved.

to_disk

Logical if TRUE, results are written into a text file.

path_disk

String with path to result text file.

verbose

Logical if TRUE, progress reports are printed.

Value

mdl_rn

Details

This is the main function of the simulation model. Besides running all sub-processes, the function also includes some basic checks to make sure the model does not crash. However, this does not ensure that all parameter values "make sense". The function returns a mdl_rn object, which stores the model run results and additionally information about the model run itself.

The functions is a 'wrapper' around the following sub-processes: (i) nutrient input, (ii) seagrass growth, (iii) detritus mineralization, (iv) movement of individuals, (v) respiration of individuals, (vi) growth of individuals, (vii) mortality of individuals, (viii) diffusion of nutrients/detritus, and ix) nutrient output.

If nutrients_input is zero (default), no nutrient input is simulated. To simulate no nutrient output, set the nutrients_loss parameter to zero.

The movement argument allows to either specify random movement of individuals ('rand'), attracted movement towards the artificial reef cells of individuals ('attr') or a movement behavior based on their biosenergetics ('behav').

seagrass_each allows to simulate all seagrass sub-processes only each specified time step.

If save_each > 1, not all iterations are saved in the final mdl_rn object, but only each time step specified by the argument. max_i must be dividable by save_each without rest.

If burn_in > 0, all sub-processes related to fish individuals are not simulated before this time step is reached.

References

For a detailed model description, please see Esquivel, K.E., Hesselbarth, M.H.K., Allgeier, J.E., 2022. Mechanistic support for increased primary production around artificial reefs. Ecological Applications e2617. <https://doi.org/10.1002/eap.2617>

Examples

if (FALSE) {
reef <- matrix(data = c(-1, 0, 0, 1, 1, 0, 0, -1, 0, 0),
ncol = 2, byrow = TRUE)

seafloor <- setup_seafloor(dimensions = c(50, 50), grain = 1,
reef = reef, starting_values = default_starting)
fishpop <- setup_fishpop(seafloor = seafloor,
starting_values = default_starting, parameters = default_parameters)

run_simulation(seafloor = seafloor, fishpop = fishpop, parameters = parameters,
max_i = 1000, min_per_i = 120, save_each = 10)
}