OOS 1-1
Interspecific variation in savanna tree rooting profiles and response to competition from grasses: Evidence from modeling and empirical studies

Monday, August 5, 2013: 1:30 PM
101A, Minneapolis Convention Center
Ricardo M. Holdo, Division of Biological Sciences, University of Missouri, Columbia, MO
Background/Question/Methods

The Walter model of tree-grass coexistence posits that trees and grasses are able to coexist in stable equilibrium as a result of rooting differences. Grasses are assumed to be competitively dominant in shallow soil layers while trees have access to deep moisture. A key premise of this two-layer model is that exclusive access to deep water is required by one of the two functional groups. A theoretical model that couples soil moisture and tree uptake and biomass dynamics suggests that exclusive access is not necessary for coexistence, as long as realistic (stochastic) rainfall dynamics are assumed. Subtle differences in rooting profiles allow coexistence of two or more species or functional groups with identical maximum rooting depths. This initial work is expanded here to identify the extent of limiting similarity (defined by two key traits: root shape and water use efficiency) that precludes coexistence under alternative water supply regimes. The theoretical results are then used as a backdrop for a greenhouse experiment designed to quantify the strength of tree-grass competition across three distinct tree rooting strategies.

Results/Conclusions

The model results showed strong effects of root shape, water use efficiency and the timing of rainfall events on coexistence and the strength of competition, and identified threshold root shape differences that permit coexistence after controlling for differences in plant physiological traits. The results of the greenhouse experiment suggested that the strength of tree-grass competition varies markedly among tree species, but the inter-specific differences were counterintuitive. The species with the shallowest rooting system (Dichrostachys cinerea), i.e., the most ‘grass-like’, performed better against grasses than the deeper-rooted Schlerocarya birrea. A possible explanation for this result is that soil moisture is depleted from shallow layers before deep infiltration can occur (i.e., exploitative competition occurs between the shallow-rooted grass and the deep-rooted tree despite the fact that little rooting overlap occurs). This is consistent with the results of the theoretical model, which suggest that the strength of competition between alternative rooting strategies is strongly modulated by the timing and size of rainfall events.