09: Troubleshooting / caveats

When creating your own simulation, you may encounter issues / errors to which the solution may not be immediately obvious. Here we will cover a basic strategy on how to debug with the example of a simple typo in one of the species traits.

Debugging

The following code will result in an error:

library(metaRange)
library(terra)
set_verbosity(0)
raster_file <- system.file("ex/elev.tif", package = "terra")
r <- rast(raster_file)
r <- scale(r, center = FALSE, scale = TRUE)
r <- rep(r, 10)
landscape <- sds(r)
names(landscape) <- c("habitat_quality")

sim <- create_simulation(
    source_environment = landscape,
    ID = "example_simulation",
    seed = 1
)

sim$add_species(name = "species_1")
sim$add_traits(
    species = "species_1",
    population_level = TRUE,
    abundance = 100,
    reproduction_rtae = 0.5,
    carrying_capacity = 1000
)
sim$add_process(
    species = "species_1",
    process_name = "reproduction",
    process_fun = function() {
        ricker_reproduction_model(
            self$traits$abundance,
            self$traits$reproduction_rate,
            self$traits$carrying_capacity * self$sim$environment$current$habitat_quality
        )
    },
    execution_priority = 1
)
sim$begin()

#> Warning: package 'metaRange' was built under R version 4.3.3
#> metaRange version: 1.1.4
#> terra 1.7.55
#> Error: Not compatible with requested type: [type=NULL; target=double].

And it may not immediately be obvious what the problem is. The first step to narrow down the problem is to enable extensive verbosity. So, if we run the code again, but this time with set_verbosity(2), we get the following output:

set_verbosity(2)

sim <- create_simulation(
    source_environment = landscape,
    ID = "example_simulation",
    seed = 1
)
#> number of time steps: 10
#> time step layer mapping: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10
#> added environment
#> class       : SpatRasterDataset 
#> subdatasets : 1 
#> dimensions  : 90, 95 (nrow, ncol)
#> nlyr        : 10 
#> resolution  : 0.008333333, 0.008333333  (x, y)
#> extent      : 5.741667, 6.533333, 49.44167, 50.19167  (xmin, xmax, ymin, ymax)
#> coord. ref. : lon/lat WGS 84 (EPSG:4326) 
#> source(s)   : memory 
#> names       : habitat_quality
#> 
#> created simulation: example_simulation

sim$add_species(name = "species_1")
#> adding species
#> name: species_1
sim$add_traits(
    species = "species_1",
    population_level = TRUE,
    abundance = 100,
    reproduction_rtae = 0.5,
    carrying_capacity = 1000
)
#> adding traits:
#> [1] "abundance"         "reproduction_rtae" "carrying_capacity"
#> 
#> to species:
#> [1] "species_1"
#> 
sim$add_process(
    species = "species_1",
    process_name = "reproduction",
    process_fun = function() {
        ricker_reproduction_model(
            self$traits$abundance,
            self$traits$reproduction_rate,
            self$traits$carrying_capacity * self$sim$environment$current$habitat_quality
        )
    },
    execution_priority = 1
)
#> adding process: reproduction
#> to species:
#> [1] "species_1"
#>

The setup look ok so far. Now we start the simulation.

sim$begin()

#> 
#> Starting simualtion.
#> 
#> passed initial sanity checks.
#> 
#> start of time step: 1
#> |- species_1 : reproduction
#> Error: Not compatible with requested type: [type=NULL; target=double].

We can see that the error occurs in the first time step, in the reproduction process of “species_1”. With this information, we can now insert a browser() function, which stops the code execution once it is called and lets us inspect the process and the variables of the function / environment it has been called from.

set_verbosity(2)

sim <- create_simulation(
    source_environment = landscape,
    ID = "example_simulation",
    seed = 1
)

sim$add_species(name = "species_1")
sim$add_traits(
    species = "species_1",
    population_level = TRUE,
    abundance = 100,
    reproduction_rtae = 0.5,
    carrying_capacity = 1000
)
sim$add_process(
    species = "species_1",
    process_name = "reproduction",
    process_fun = function() {
        browser()

        ricker_reproduction_model(
            self$traits$abundance,
            self$traits$reproduction_rate,
            self$traits$carrying_capacity * self$sim$environment$current$habitat_quality
        )
    },
    execution_priority = 1
)
sim$begin()

In the browser, we are conceptually inside the reproduction process of “species_1”. This means we can make use of the self keyword to inspect the state of the species.

As a first step, we might want to call ls() to see which objects we can inspect.

# type this in the console,
# once the browser has halted the code execution
ls()

#> [1] "initialize" "name"       "print"      "processes"  "sim"       
#> [6] "traits"

Since the error was about an wrong type being passed to the reproduction function, we can inspect the traits of the species to see if they are as we would expect them to be. This means we can just type self$traits in the console to see them and we may notice that the reproduction_rate is misspelled as reproduction_rtae.

# type this in the console,
# once the browser has halted the code execution
self$traits

#> abundance :  num [1:90, 1:95] 100 100 100 100 100 100 100 100 100 100 ...
#> carrying_capacity :  num [1:90, 1:95] 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 ...
#> reproduction_rtae :  num [1:90, 1:95] 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 ...

We can quit the browser by typing Q and then Enter in the console. Now we can remove the browser() call from the code again and proceed to fix the typo.

General caveats

While metaRange can be used to simulate a wide range of scenarios, there are some caveats to keep in mind.

Different scales of the environment and the species

Since the size and resolution of the environment also describes the spatial size of each population (i.e. one grid cell = one population), it is important to choose scales that are appropriate for the species. This is especially important to keep in mind when simulating multiple species, since they may have different spatial requirements.
Gene flow

While it is planned for future versions, metaRange does (currently) not support gene flow during dispersal.
Spatial distortion

Since metaRange uses raster data to represent the environment, it is important to keep in mind that the raster is a 2D representation of a 3D world. The larger the geographic extent of the environment, the more distorted the raster will be (also depending on the map projection and the resolution).

Fallert, S. and Cabral, J.S.

Debugging

General caveats