Rescale SDM predictions and (if applicable) associated uncertainties
Usage
ensemble_rescale(x, x.idx, y, y.abund = NULL, x.var.idx = NULL)
Value
The sf object x with the columns specified by x.idx and x.var.idx rescaled.
The agr attributes of x will be conserved
Arguments
x
object of class sf
x.idx
vector of column names or column indices;
indicates columns in x with prediction values that will be rescaled
y
rescaling method; must be either "abundance" or "sumto1".
See 'Details' section for descriptions of the rescaling methods
y.abund
numeric value; ignored if y is not "abundance"
x.var.idx
vector of column names or column indices;
indicates columns in x with variance values that will be rescaled.
If x.var.idx is specified, it must be the same length as x.idx.
Use x.var.idx = NULL (the default) if none of the predictions have associated uncertainty values;
see the 'Details' section for more information
Details
ensemble_rescale is intended to be used after overlaying predictions with
overlay_sdm and before creating ensembles with ensemble_create.
The provided rescaling methods are:
'abundance' - Rescale the density values so that the predicted abundance is y.abund
'sumto1' - Rescale the density values so their sum is 1
SDM uncertainty values must be rescaled differently than the prediction values.
Columns specified in x.var.idx must contain variance values.
These values will be rescaled using the formula var(c * x) = c^2 * var(x),
where c is the rescaling factor for the associated predictions.
If x.var.idx is not NULL, then the function assumes
x.var.idx[1] contains the variance values associated with the predictions in x.idx[1],
x.var.idx[2] contains the variance values associated with the predictions in x.idx[2], etc.
Use NA in x.var.idx to indicate a set of predictions that does not have
associated uncertainty values (e.g., x.var.idx = c(4, NA, 5))