Linking allocation and performance tradeoffs in Central Amazonian saplings

Marcel Caritá Vaz, Latin American Chapter (Vice-chair); Department of Ecology & Evolutionary Biology, University of California, Los Angeles, CA, José Luís C. Camargo, National Institute of Amazonian Research, Manaus, Brazil, Alberto Vicentini, Instituto Nacional de Pesquisas da Amazônia, Brazil, Alexandre A. Oliveira, Departamento de Ecologia, Universidade de São Paulo, São Paulo, Brazil and Nathan Kraft, Department of Ecology and Evolutionary Biology, University of California, Los Angeles, CA

Background/Question/Methods
Vertical stratification of foliage is likely the most important form of niche partitioning among tropical forest trees. A mechanism that has been proposed to explain how species may differentiate in their adult heights (Hmax) states that tall species must have saplings with a growth strategy that maximizes height gain (optimists), while saplings that start branching prematurely (pessimists) get stuck in the shaded understory as adults. In a previous work we refuted this hypothesis for our study site, a hyperdiverse forest in the Central Amazon, where Hmax was a very poor predictor of sapling growth strategies. Although it is understandable that saplings from tall understory species could also have optimistic strategies, we were surprised to learn that many canopy or even emergent species have pessimistic saplings. How can these plants grow their way into the canopy after having already allocated resources into crown expansion? To explore this intriguing problem, we tested the effects of sapling architecture (total height, crown projected area, and crown thickness), stem diameter, and crown illumination index on their relative growth rate (RGR) over a period of about 10 years. To account for the great variability in sampling size across >400 species we used a hierarchical Bayesian approach.

Results/Conclusions
Despite very slow diametric growth in the shade (~0.007 cm yr-1cm-1), we found that RGR was negatively related to sapling height, but positively related to sapling crown size. This shows that pessimistic saplings do better in the shade compared to pessimistic ones. When we make model projections, however, the scenario changes: with improving light conditions, optimistic saplings tend to grow faster than pessimistic ones. This points to a performance tradeoff mediated by variation in light availability, yet our model did not indicate a strong effect of architecture in the way saplings respond to changes in illumination. Hmax, however, was a much better predictor of saplings' growth responses to light, with saplings from tall species respondingmore strongly to increases in light compared to short species. Together, these results show that while saplings with larger crowns are better adapted to shaded conditions, they can only escape the understory if they have adaptations to capitalize on improving light conditions. One such adaptation could be the capacity to increase their maximum photosynthetic capacity (Amax). Although it makes sense that canopy trees that regenerate in the shade must have a large plasticity in Amax, this remains to be tested for this particular forest.