A variety of methods exist for implementing naive sampling schemes
for polygon-based products, such as provisional and empirical Seed
Transfer Zones (pSTZs and eSTZs), as well as Ecoregion (e.g., Omernik or
Baileys). The sampling of these products is implemented using a single
function, safeHavens PolygonBasedSample, which considers
environmental traits of seed transfer zones and the size and number of
polygon patches for both ecoregions and STZs.
The function requires modes for two main scenarios: when a user wants more collections than zones exist in the data, the function needs to use an ‘increase’ method. When a user wants fewer collections than there are zones, a decrease method needs to be implemented. Three options are available for both scenarios.
- number of polygons per class
- total area of each class
- simple climate parameters for each class
Additionally, this function can be applied to the clusters identified by the Isolation by Distance, Resistance, and Environmental workflows when a user wants to decouple the clustering process from the number of samples.
See ?PolygonBasedSample for notes on implementation.