Predicting plant performance for the ecological restoration of grasslands: the role of regenerative traits

Abstract

Active restoration of grassland ecosystems often requires the addition of species as seeds. However, knowledge of seed germination and initial performance of native species are still lacking in many regions, including the Brazilian grasslands. Regenerative traits (such as seed traits) have the potential to predict performance parameters related to the initial establishment of a species in a given environment. This study aimed to test the potential of regenerative traits as predictors of the performance of grassland species, information which may be useful for ecological restoration purposes. Using 15 grassland species, we carried out three parallel experiments at different conditions: cabinet, greenhouse, and field. Three performance parameters were evaluated: final germination percentage (FGP), germination rate index (GRI), and survival rate. The suitability of seed traits (mass, area, and shape) as predictors of performance parameters was evaluated for each experiment using linear regression models. Performance parameters varied among the species and experiments, but seed mass and shape slightly influenced GRI. Seed area had a positive effect on GRI, and there was a significant interaction between seed circularity and mass, which improved the GRI of round, heavy seeds. FGP and survival rate were not predicted by any of the traits. Our study highlights the importance of testing fitness–trait relationships under field conditions, which may reflect the real challenges species will face in restoration projects. Our results reinforce the need of considering regenerative traits and interaction factors to improve the understanding and prediction of the initial performance of species.