A set of simple decision matrices for prioritizing collection of rare plant species for ex situ conservation

Abstract

Because it is virtually impossible to collect seed or tissue for ex situ conservation banks from every known population of rare plant species, it is important to rank populations systematically in terms of their priority for collection. The New England Wild Flower Society, which maintains a regional seed bank, developed a set of three complementary decision matrices in spreadsheet form by which to prioritize all occurrences of all state-listed rare plant species in New England in terms of their urgency and feasibility for collection. Data on 4333 occurrences, spanning 456 taxa, were collated from six state Natural Heritage Programs. The first decision matrix ranked taxa in terms of their amenability to storage or propagation at ex situ institutions, and determined whether any known New England occurrences were reproductive. The second matrix further ranked taxa in terms of their global and regional rarity and the viability and genetic and geographic representation of collections already present in the bank. The third matrix scored individual occurrences within each taxon in terms of the presence of imminent threat, reproductive status, vigor, protection status, potential genetic distance from other occurrences, availability of landowner permission, and their current status in the bank. Occurrences were then sorted in ascending order by total matrix score; those with low scores were at the top of the list for collection priority. 3743 occurrences were deemed eligible for collection. Scores ranged from 14.5 to 182, and were influenced most strongly by the number of occurrences per taxon. Clear breakpoints were apparent in the distribution of scores, with clusters of uncommon taxa at the low end of the scale and a long tail created by taxa with more numerous occurrences in New England. These breakpoints could potentially be used to prioritize groups of occurrences that should receive the first attention for collection, while postponing collection of higher-scoring groups. Fewer than 1% of occurrences were misclassified, according to post hoc inspection. This simple set of decision matrices can be adapted by a wide range of institutions involved in ex situ conservation.