Project Status: Active – The project has reached a stable, usable
state and is being actively




The goal of the package mobsim is to facilitate understanding of scale-dependent biodiversity changes.

The package includes functions to simulate species distributions in space with controlled abundance distributions as well as controlled intraspecific aggregation. For analysis there are functions for species rarefaction and accumulation curves, species-area relationships, endemics-area relationships and the distance-decay of community similarity.

A detailed introduction of the package is available at bioRxiv.


# The easiest way to get mobsim is to install from CRAN:

# Or the development version from GitHub:
# install.packages("devtools")
devtools::install_github("MoBiodiv/mobsim", build_vignettes = TRUE)

Please enter bug reports on github.

Getting help

You can get an overview of the available functions in mobsim:


Or have a look at tutorials in the vignette:



Here is an example of how to simulate two communities, which just differ in their spatial aggregation of species, but have the same species abundance distribution and the same total number of individuals.

Simulation of communities

comm_rand <- sim_poisson_community(s_pool = 30, n_sim = 300)
comm_agg <- sim_thomas_community(s_pool = 30, n_sim = 300, sigma = 0.05, mother_points = 1)
par(mfrow = c(1,2))

Analysis of spatially-explicit community data

mobsim mobsim offer functions to analyse spatially-explicit community data. For example the species-area relationship of a community can be easily evaluated.

sar_rand <- divar(comm_rand)
sar_agg <- divar(comm_agg)
plot(m_species ~ prop_area, data = sar_rand, type = "b", log = "xy",
     xlab = "Proportion of area sampled",ylab = "No. of species",
     ylim = c(3,30))
lines(m_species ~ prop_area, data = sar_agg, type = "b", col = 2)
legend("bottomright", c("Random","Aggregated"), col = 1:2, lwd = 2)

Sampling of communities

Simulated or observed communities can be also sampled inorder to test whether biodiversity changes are correctly detected and revealed by any sampling design.

par(mfrow = c(1,2))
samples_rand <- sample_quadrats(comm_rand, avoid_overlap = TRUE)
samples_agg <- sample_quadrats(comm_agg, avoid_overlap = TRUE)

MoBiodiv/MoBspatial documentation built on May 22, 2024, 7:44 p.m.