Invasions by African grasses and Cerrado restoration: an approach by state-and-transition models

Abstract

Alternative stable states (ASS) are the possible stable configurations in which an ecological system can exist due to its abiotic and biotic characteristics. Furthermore, the same ASS can handle distinct system configurations caused by minor environmental alterations. The presence of invasive species may shift system configurations and the accumulation of such modifications can cause the transition between ASSs. Ecological restoration of invaded areas is more viable when it intends to reverse a transition between configurations within a unique ASS than when the transition of an ecological threshold have already occurred. In this case, restorations could promote a third and distinct ASS. Therefore, it is fundamental to understand the mechanisms and biotic and abiotic characteristics that create the degradation/restoration dynamics in invaded ecosystems. This project aims to construct a conceptual framework based on state-and-transition models for Cerrado invasions by African grasses, using Urochloa brizantha as the focal studying species. The model will be constructed over primary data acquired across the levels of microhabitat (light incidence, humidity and temperature), individuals (specific leaf area - SLA), community (structure and species richness and diversity), and ecosystem (primary productivity, fluxes of carbon and decomposition rates). Experiments will be of BACIP type (before-after control-impacted paired) and implemented in central and southern portion of the Cerrado. In each area, we will establish three blocks of paired plots (each plot of 25m2), each pair being composed by a non-manipulated (control) and manipulated plot. Each block will cover a gradient in the abundance of U. brizantha (0, 25, 50, 75 and 100 percentage of vegetation cover by this species). We will conduct three prescribed annual burns at the transition between the dry and rainy seasons in order to investigate the relationship between the invasive abundance and the state of the ecological system after the manipulation. We will use a sort of statistical analysis (pathway analysis, generalized additive and linear mixed models, ordination and permutation analysis) and also a heurist approach to the model creation. Finally, this proposal also comprehends an international scholarship into the University of Western Australia under the advisor of Prof. Samantha Setterfield, which will collaborate with the development of the conceptual model.