Collaborative Research: Rainfall variability and the axes of tree-grass niche differentiation

Kansas State University, Manhattan KS

Investigators

Abstract

The savanna biome, in which a discontinuous tree layer coexists with a continuous grass layer, has a worldwide distribution. Increasingly, research suggests that the relative proportion of tree and grass biomass in savannas is strongly influenced by rainfall patterns, with higher rainfall leading to greater woody biomass. Despite mounting evidence for rainfall effects, however, it is unclear if more rainfall leads to greater tree biomass in savannas because of deeper water infiltration, changes in rainfall intensity, or because the length of the growing season increases with the amount of annual rainfall. This question is important because we cannot accurately predict future changes to the savanna biome without a detailed understanding of how tree and grass species respond to rainfall patterns. This project will seek answers to this and other questions using a South African savanna as a model system. North American savannas have been largely replaced by forests, and the reasons for this are not fully understood. Results from this project can be used to inform the restoration and maintenance of savannas in North America and elsewhere, as well as to predict with broad generality how the savanna biome might change in the future. The project will also introduce undergraduate students to the ecology of savannas through a multi-year field course and educate local communities about the role of rainfall variability on natural resource availability. The project will combine experimental, observational and modeling approaches to test alternative hypotheses about the relationship between rainfall and tree to grass ratios in savannas. The research has three specific aims: (1) identify the rainfall regimes that favor trees over grasses, (2) identify the functional traits and tradeoffs that differentiate savanna trees and grasses, and (3) develop and test a mechanistic model of tree and grass dynamics as a function of rainfall. The core element of the study will be a field experiment that will manipulate the timing, seasonality and depth distribution of rainfall, and monitor the response of a multi-species assemblage of known-age trees embedded in a grass matrix. In addition, the project will examine functional differences (e.g., drought tolerance, water use efficiency and functional rooting depth) between six tree and six grass species using greenhouse, growth chamber and field experiments. Finally, the project will use parameter estimates from this second component to develop and test a mechanistic model of tree growth and tree to grass ratios based on whole-plant traits, and validate the model under field conditions by comparing modeled and observed changes in tree to grass ratios to rainfall manipulations in the field experiment. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.