Increased disturbances across the western United States lead to a greater demand for native seed for restoration and rehabilitation projects. Seed selection tools may facilitate restoration projects by ranking seed source choices for current and future climates, indicating appropriate collection sites for current and future climates, and identifying collection gaps. These can make the seed mix selection and pre-project collection choices easier for restoration practitioners who may otherwise have only generalized seed transfer zones to guide them. We compared three seed selection tools, Seed Selector (SS)/Climpart, Climate Distance Mapper (CDM), and Climate Smart Restoration Tool (CSRT), using three case studies on the Colorado Plateau. We analyzed tool outputs to identify the best tool for each function, and to determine the frequency with which tool recommendations aligned in several outputs: climate similarity values quantifying climate similarity between a seed source location and project site; seed source rankings; area of appropriate collection sites; and relative size of clustered species-specific seed zones generated by each tool. Due to tool limitations, not every tool performs each function and only the tools capable of performing a given function were included in analyses.
Results/Conclusions
On average, CDM calculated higher climate similarity values than SS, and there was a significant difference between average climate similarity values in CDM and SS, likely from differing underlying climate data. However, source rankings largely aligned. For each species, sources with the 3 highest climate similarity values were ranked highly in both CDM and SS in 55% of the tests. When tools indicated appropriate collection sites, SS ranked the largest portions of the Colorado Plateau as 70%+ climate similar to the reference site and CSRT had the smallest portions, since the latter limited calculations by specific species’ traits. To identify collection gaps, Climpart randomly assigned climate centers and generated clustered seed zones of approximately equal size, while CDM incorporated user-imported source coordinates as climate centers and had the largest range in seed zone sizes. Overall, CDM performed all three functions well, though SS required less time to learn and was not constrained to specific ecoregions. As more seed sources and commercial releases become available to restoration practitioners, these tools will become increasingly useful, especially as other ecologically-important factors (such as adaptive traits specific to sources) are incorporated. They could also help create seed mixes adapted to both current and future climates to ensure maximum restoration success.