9615226 Hubbell Unexpectedly large differences in tree diversity that correlate with topography and dry season soil moisture have been found on small spatial scales in tropical forests. This research is an experimental test of the simple hypothesis that tropical tree diversity is controlled largely by drought stress in the seedling and small sapling stages. In drier sites, a smaller fraction of tree species may have sufficiently deep root systems as seedlings to survive prolonged periods of drought than is the case in wetter sites. The hypothesis will be tested by a watering experiment on seedlings transplanted to sites with greater or lesser dry season soil moisture and drought stress within a single tropical forest. The experiment will demonstrate whether dry season watering increases seedling survival and growth compared with seedlings in non-watered control plots. The experiment will also show whether the effect is sufficiently large to explain the large observed differences in species richness between sites. World-wide patterns of tree species richness in tropical forests show a strong, unexplained pattern of increasing richness with mean annual rainfall. This research will help us to understand the potential causes of large geographical variation in tropical forest diversity and whether it is controlled by events in the seedling stages. The study will improve predictions on the impact of global climate change on loss of tropical forest diversity. Major changes are forecast in the next century by Global Climate Models in the worldwide distribution of annual and seasonal rainfall, and much of the tropics is expected to become drier.
