Inter- and Intra-specific variability in functional traits of seedlings planted in successional tropical dry forest

Abstract: The dry tropical forests are among the most threatened ecosystems in the world; they have been degraded and removed due to commercial values and cattle ranges. Restoration projects have gained global importance, including in Costa Rica where large restoration efforts have been started since 1970s. However, previous restoration projects have scored limited success in several areas due to species mismatch with the prevailing site conditions. Thus, trait-based approach has been emerged as a way to identify candidate species for restoration. A trait is a morphological, anatomical, biochemical, physiological and/or phenological feature measurable at the individual level; whereas a functional trait is any trait which impacts fitness indirectly by its effects on growth, reproduction and survival. The aims of the study were to examine variation in traits within (intra-specific) and between (inter-specific) species in relation to above- (light) and below-ground (nutrient and water) resources; and if functional traits are good predictors of survival and growth of planted seedlings of four tree species, Astronium graveolens (RonRon), Dalbergia retusa (Cocobolo), Hymenaea courbaril (Guapinol), and Thouinidium decandrum (Sardino). For this purpose, a factorial experiment involving two levels of forest type (old versus young) and trenching (with versus without) was established in the forest experimental station, Horizontes, a part of Conservation area of Guanacaste. There were four plots in each forest type, and four seedlings of each species were planted in May 2013 in each sub-plot. Leaf- and stem functional traits and survival rate were assessed two years post-planting while root collar diameter and height of seedlings were recorded for two consecutive years post-planting. Two-way analysis of variance (ANOVA) was performed to determine significant differences in intra-specific trait variabilities while three-way ANOVA was performed to determine differences in traits and seedling performance among species and treatments. To examine whether traits can predict field performance of planted seedlings, correlation analyses were performed between traits and seedling growth variables. The results showed that intra-specific leaf and stem traits did not vary significantly between forest types and trenching treatment for all the studied species, except A. graveolens for which specific leaf area was significantly higher (p = 0.015) for seedlings planted under old- than young forest. Three-way ANOVA showed highly significant differences in leaf- and stem traits among species (p < 0.01), except stem dry matter content (p = 0.209). Leaf thickness, leaf dry matter content and specific leaf area did also significantly vary between forest types (p < 0.05) but not between trenching treatments and interactions between species, forest type and trenching treatments. The mean stem specific density, leaf dry matter content and leaf density were lower while specific leaf area was higher for D. retusa than other species. H. courbaril had higher leaf thickness than the rest of the species. For A. graveolens and T. decandrum, stem specific density, specific leaf area, leaf dry matter content and leaf density did not differ significantly. Among the traits analysed, leaf thickness and leaf dry matter content were significantly higher for seedlings planted in young- than old forest; whereas specific leaf area was significantly lower for seedlings planted in young- than old forest. The trenching treatment (with versus without) had no effect on inter-specific traits. The correlation between inter-specific traits and seedling growth was significant for most of the traits analysed. Seedling height and root collar diameter during the first two years post-planting significantly and positively correlated with leaf thickness and negatively with specific leaf area. The current annual increment in height (CAIH) and root collar diameter (CAID) were positively correlated with leaf density and leaf dry matter content but negatively correlated with specific leaf area. Correlation analysis between intra-specific traits and seedling growth was also performed; and significant correlation was found between leaf thickness and stem dry matter content with seedling growth in three out of four species; especially leaf thickness was correlated with both diameter and height growth of two out of four species. In conclusion, this study highlights the importance of functional traits when looking on survival and growth of different species in different environmental conditions. This new trait-based approach is shown to predict survival and growth and thereby can be used as a possible guide for future restoration projects in similar climatic- and environmental conditions. As sample sizes were few in the present study, I recommend further research using larger sample size for each treatment (sub-plot) and more species to be able to identify the best predictive traits.