NCEAS Product 5760

Bascompte, Jordi; Rodriguez, Miguel A. 2000. Self-disturbance as a source of spatiotemporal heterogeneity: The case of the tallgrass prairie. Journal of Theoretical Biology. Vol: 204. Pages 153-164. (Abstract) (Online version)

Abstract

Tallgrass prairies are characterized by high levels of litter production, which has a profound effect on live biomass. Litter introduces a delayed inhibition of biomass growth, generating nonlinear dynamics and chaos. In this paper, we study a model of biomass–litter interaction, and focus on the litter persistence rate. The observed dynamics depends largely on this rate of year-to-year persistence. Different scenarios are explored and discussed. A spatially extended counterpart of such a model is later on introduced to account for the effects of space. Temporal chaos introduces spatial heterogeneity in terms of gaps where the current year biomass is almost zero. Such gaps can be colonized by fugitive species. The inhibitory effect of litter on biomass is thus an important source of intrinsic, small-scale heterogeneities that may promote diversity. On the other hand, the huge amounts of litter produced by the competitive dominants in tallgrass prairies enhance the probability of fires. Fires benefit, rather than depress, the superior competitive species. This fact explains why the intermediate disturbance hypothesis (IDH) stating that the highest diversity levels should be observed at intermediate disturbance frequencies, does not work in these communities. We define self-disturbances as small-scale disturbances affecting the growth and survival of the individuals that have generated them (e.g. due to the effects of the litter mass they produce). In the absence of other disturbances, self-disturbances can induce high heterogeneity and diversity levels in tallgrass prairies. We discuss the general implications of self-generated disturbances for landscape heterogeneity and diversity of communities in which the main external perturbations benefit the dominant species.