One potentially useful addition to the structured randomisations is to use the abundance based species richness estimators (Chao1, ACE) implemented in #420 to define the target richness scores.
This would simply involve the calculation of the relevant index for a single neighbour set of sp_self_only(). This means the incidence based indices cannot be used as they cannot be calculated for a single group (unless we add a spatial condition to define the richness targets).
These targets will usually exceed the observed richness, but the randomisations can already handle such cases. We can use the observed richness for cases where the estimators return undef.
The richness_multiplier and richness_addition parameters can apply in the usual way.
An alternate approach is to use the confidence intervals to define a target richness range. This would probably need to be in a separate randomisation.
One potentially useful addition to the structured randomisations is to use the abundance based species richness estimators (Chao1, ACE) implemented in #420 to define the target richness scores.
This would simply involve the calculation of the relevant index for a single neighbour set of sp_self_only(). This means the incidence based indices cannot be used as they cannot be calculated for a single group (unless we add a spatial condition to define the richness targets).
These targets will usually exceed the observed richness, but the randomisations can already handle such cases. We can use the observed richness for cases where the estimators return undef.
The richness_multiplier and richness_addition parameters can apply in the usual way.