Skip to contents

Estimates, for each species, the rates of colonisation \(c\) and local extirpation \(e\) of a two-state Markov model of presence in sites over time, and from them the two axes of metapopulation persistence of Januario et al. (2025): the equilibrium occupancy \(\epsilon = \mathrm{logit}(c / (c + e))\) and the resilience \(\tau = \log(c + e)\), the log of the rate of turnover. The estimates are the maximum likelihood ones: \(c\) is the share of the site-intervals in which an absent species arrived, and \(e\) the share of those in which a present one was lost.

Usage

persistence(x, window = NULL, ...)

Arguments

x

An mc_evolving or mc_sim object, or an array (sites x species x times) of abundances or presences.

window

Number of the last intervals (generations, or intervals between records) to use; NULL for all.

...

Unused.

Value

A data frame with one row per species (lineage, for mc_evolving): species, the number of intervals used, colonisation, extirpation, the equilibrium occupancy \(c / (c + e)\) as occupancy, epsilon and tau. For mc_evolving, also extant and the good species each extant lineage belongs to (good_species). Rates are NA without any interval in the corresponding state, and epsilon is infinite when one of the rates is 0.

Details

For simulate_evolving_metacommunity(), transitions come from its series, between consecutive generations, for every lineage (extinct ones included, with the sites lost in the generation they died out); a site that a mother passed to a new species counts neither as an extirpation nor as exposure. For simulate_metacommunity(), they come from the abundances at record_steps (history), between consecutive records, which should be evenly spaced (seeding from the pool, seed_steps, counts as colonisation). An array of sites x species x times can be given directly.

References

Januario, M., Pinsky, M. L., & Rabosky, D. L. (2025). The metapopulation bridge to macroevolutionary speciation rates: a conceptual framework and empirical test. Ecology Letters, 28, e70021. doi:10.1111/ele.70021

Examples

set.seed(1)
ls <- landscape_grid(5, disp_scale = 0.5)
ancestor <- data.frame(optimum = 50, sigma = 20, r_max = 5, row.names = "sp1")
sim <- simulate_evolving_metacommunity(ls, ancestor, dispersal = 0.01, extirpation = 0.02,
                                       n_steps = 200,
                                       speciation = speciation_control("fission", rate = 0.02))
head(persistence(sim, window = 50))
#>   species intervals colonisation extirpation occupancy     epsilon       tau
#> 1     sp1        50   0.07993967  0.09466437 0.4578340 -0.16906551 -1.745234
#> 2     sp2        50   0.09105691  0.08346457 0.5217519  0.08706251 -1.745707
#> 3     sp3        50   0.09843750  0.08789386 0.5282927  0.11329183 -1.680229
#> 4     sp4        50   0.08345979  0.06643952 0.5567723  0.22807282 -1.897791
#> 5     sp5         0           NA          NA        NA          NA        NA
#> 6     sp6        50   0.04871795  0.07659574 0.3887680 -0.45249400 -2.076935
#>   extant good_species
#> 1   TRUE          sp1
#> 2   TRUE          sp2
#> 3   TRUE          sp3
#> 4   TRUE          sp4
#> 5  FALSE         <NA>
#> 6   TRUE          sp6