OOS 62-6 - Drivers of reproductive patterns of woody species in a Puerto Rican wet forest.

Reproductive patterns in tropical forests challenge conventional perspectives. For example, seasonality in the tropics is not driven by periods of low temperatures as in other climatic zones, but by other factors such as rainfall and irradiance levels. Furthermore, seed production in the tropics far exceeds the three-fold increase in net primary productivity from dry boreal forest to wet tropics, suggesting an increased importance of biological interactions at low latitudes. We review reproduction patterns in the Luquillo Forest Dynamics Plot in Puerto Rico, whose studies recently completed 30 years of trap (120) results. The forest type is everwet, never having less than 200 mm of rainfall per month, on average. Seasonality in flowering and fruiting was common among the ~60 species studied after ten years, however, we found relatively little evidence that community flowering and fruiting was highest in summer months when irradiance levels are highest. Thus, seasonal reproduction in this forest are poorly explained by abiotic drivers. We then explored how interannual reproduction in flowering and fruiting (species per month) were driven by El Niño Southern Oscillation (ENSO) and other global climate drivers as well as hurricane disturbance (using 21 years of data representing ~70 species). We found that the effect of ENSO on reproduction was stronger than other geographically closer climate drivers (the North Atlantic Oscillation). Patterns were modified by hurricane disturbance (Hurricane Hugo 1989), which tended to increase reproduction by species (particularly shrubs and lianas); subsequent hurricane damage (Hurricane George 1998) further modified these changes. These combined factors, however, explained only 9 – 40% of the variance in the numbers of reproducing species. We are currently exploring the synchrony in reproduction among species using 30 years of data, comparing species groups categorized by stature (short vs. tall) and the growth rate/mortality trade-off (fast/high vs. slow/low). The hypothesis was that species groups would influence patterns of seasonal and interannual reproduction, e.g., shrubs (short and fast/high) would be seasonally continuous but only reproduce following canopy opening by hurricanes (resilience) while understory treelets (short and slow/low) would exhibit seasonal peaks with relatively little interannual variation (resistance). Other groups would show intermediate characteristics of seasonal and interannual synchrony we predicted. We found limited support for the hypothesis. In sum, these findings suggest we must dig deeper into other biological factors that drive reproductive patterns in tropical trees, giving more weight to issues such as pollination (specialization), predation (masting) and mycorrhizal associations.