In tropical forests, lianas are increasing relative to trees, with potential negative impacts for tree diversity, timber production and carbon sequestration by tropical forests. One hypothesis for explaining this increase in lianas states that lianas take advantage of increasingly dry conditions, since they would have better access to below ground water sources and outperform trees during more severe dry seasons. We test this hypothesis by simulating the water and carbon balance and predicting growth of adult lianas and trees from a dry tropical forest (~3200 mm rain/year) and wet tropical forest (~1300 mm rain/year) in Panama, sampling in total 256 terminal branches from 100 individuals, distributed over 8 liana species and 8 tree species in each forest. We measured initial sizes of those branches and monthly growth (i.e. length growth, leaf dynamics, and bud/shoot dynamics) for those branches over 18 months, including 2 dry seasons. We parameterized a mechanistic plant model for each of those branches, i.e. the measured dimensions of branches and individuals and species-specific functional traits that are supposed to control leaf hydration and carbon economy of branches. We used the model to predict growth phenology from underlying bio-physical principles, and compared our prediction to observed growth phenology.
Results/Conclusions
Our results imply that lianas can indeed maintain better hydrated leaves during the dry season, certainly in drier forest. Yet, the resultant better carbon economy of lianas during the dry season, as implied by model simulations, does not translate in stronger growth of branches in the dry season. The model cannot predict absolute growth differences in branch growth between liana and tree branches in each forest, but properly predicts the relative differences in liana and tree growth between the 2 forests. This implies that 1) dry site conditions control for a relatively strong growth and dominance of lianas in dry tropical forests, 2) a carbon surplus of liana branches is used for maintenance of a wider fine root system supplying the water, which would be in line with the our hypothesis. Alternatively, we cannot exclude that the carbon surplus is used for development of lianas branches from lower positions, or for more prolific fruiting. Overall, our results confirm that lianas indeed benefit from the dryer conditions by maintaining a better hydrated status, but not by more growth, in the dry seasons relative to trees, and we can only partially explain the dominance of lianas under increasingly drier conditions.