Add pseudo-absence points to a point data setSource:
For most engines, background or pseudo-absence points are
necessary. The distinction lies in how the absence data are handled. For
poisson distributed responses, absence points are considered background
points over which the intensity of sampling (
lambda) is integrated (in
a classical Poisson point-process model).
In contrast in
binomial distributed responses, the absence information is
assumed to be an adequate representation of the true absences and treated by
the model as such... Here it is advised to specify absence points in a way
that they represent potential true absence, such as for example through
targeted background sampling or by sampling them within/outside a given
add_pseudoabsence( df, field_occurrence = "observed", template = NULL, settings = getOption("ibis.pseudoabsence") )
tibbleobject containing point data.
charactername of the column containing the presence information (Default:
SpatRasterobject that is aligned with the predictors (Default:
NULL). If set to
pseudoabs_settings()has to be a
data.frame containing the newly created pseudo absence points.
pseudoabs_settings() object can be added to setup how absence
points should be sampled. A
bias parameter can be set to specify a
bias layer to sample from, for instance a layer of accessibility. Note that
when modelling several datasets, it might make sense to check across all
datasets whether certain areas are truly absent. By default, the
pseudo-absence points are not sampled in areas in which there are already
This method removes all columns from the input
df object other
field_occurrence column and the coordinate columns (which
will be created if not already present).
Stolar, J., & Nielsen, S. E. (2015). Accounting for spatially biased sampling effort in presence‐only species distribution modelling. Diversity and Distributions, 21(5), 595-608.
Bird, T.J., Bates, A.E., Lefcheck, J.S., Hill, N.A., Thomson, R.J., Edgar, G.J., Stuart-Smith, R.D., Wotherspoon, S., Krkosek, M., Stuart-Smith, J.F. and Pecl, G.T., 2014. Statistical solutions for error and bias in global citizen science datasets. Biological Conservation, 173, pp.144-154.