ResNatSeed: An R package and shiny web app to predict the REStoration potential of NATive SEEDs using topographic factors

Grasslands ecological restoration relies on native seeds. The methods of assessment usually used are expert-based and focused on the similarity between the climatic and topographic factors of the two sites. The aim of the work was to develop a tool in R environment able to predict the suitability of a seed mixture for the receiving site, starting from some easily measurable topographic factors: elevation, slope, and aspect. The modeling process used a training database containing vegetation and topographic information, allowing the modeling of each species abundance across the three factor gradients. A Suitability Index is then computed through the composition of a seed mixture (or that of the donor site) and on the topography of the restoration site, based on the previous models. An open access Shiny application was also set up to provide an easy-to-use tool for suitability modeling. This method can practically help practitioners during restoration programs using native seeds, reducing the subjectivity when choosing potential donor grasslands for valuable seeds.