Drought-ex: Soil fertility affects tropical dry forest responses to experimental drought

German Vargas G., Plant and Microbial Biology, University of Minnesota, Saint Paul, MN, Daniel Perez-Aviles, Plant and Microbial Biology, University Of Minnesota, St. Paul, MN, David Medvigy, Department of Biological Sciences, University of Notre Dame, Notre Dame, IN, Bonnie G. Waring, Grantham Institute, Imperial College of London, London, United Kingdom and Jennifer S. Powers, Department of Ecology, Evolution, and Behavior, University of Minnesota, St. Paul, MN

Background/Question/Methods
Tropical forest ecosystems account for a great proportion of global net primary productivity. Among tropical regions, the tropical dry forest (TDF) is expected to face a decrease in mean annual rainfall and a change in the timing of precipitation events as a consequence of climate change. At the same time, atmospheric nutrient deposition and the soil heterogeneity present in this biome could create an array of responses to drought as a function of soil nutrient availability. These characteristics challenge our understanding and ability to predict how this ecosystem will respond to novel, drier conditions. To test whether soil fertility affects ecosystem net primary productivity responses to drought, we established the first large-scale through-fall exclusion (50% of incident rainfall) crossed with a fertilization experiment in the TDF biome. We used a nested design with four replicates per level: control, fertilization, drought, and drought+fertilization. We established the experiment in mixed plantations of tropical hardwood species that have not received management for the last 25 years. This set up allowed us to replicate responses to experimental treatments in six focal tree species that represent four leaf habit functional types (deciduous stem-succulent, deciduous, brevi-deciduous and evergreen) commonly used in modeling vegetation dynamics.
Results/Conclusions
We present four years of data on leaf-litter productivity (PPL), leaf area index (LAI), fine roots production (PPR) and woody biomass growth (PPW), as metrics of primary productivity. The treatments did not significantly change the total annual fluxes of neither PPL ,PPW and PPR. However, PPW was on average 40% higher for the control over the the drought treatment. Moreover, we found a shift in the phenological cycles of LAI, PPL and PPR. Despite the annual biomass fluxes being relatively similar among treatments, changes in plant phenology with global change may have consequences for higher trophic levels. In general, nutrient additions seemed to alleviate the effect of drought on primary productivity as the drought+fertilization treatment always showed primary productivity values similar to those of the control or the fertilized plots. These findings suggest that TDF responses to drought may differ as a function of soil fertility. Ultimately, the results of this work will fill a gap in current empirical knowledge of responses to drought in the TDF and improve the accuracy of dynamic vegetation modeling of the TDF.