Feedbacks between vegetation and disturbance processes promote long-term persistence of forest–grassland mosaics in south Brazil

• Published in: Ecological modelling 2014-11, Vol.291, p.224-232
• Main Authors: Blanco, Carolina Casagrande, Scheiter, Simon, Sosinski, Enio, Fidelis, Alessandra, Anand, Madhur, Pillar, Valério D.
• Format: Article
• Language: English
• Subjects: Alternative stable states; Animal and plant ecology; Animal, plant and microbial ecology; Biological and medical sciences; DGVM; Disturbances; Ecology; Feedback; Fires; Forest expansion; Forestry; Forests; Fundamental and applied biological sciences. Psychology; General aspects. Techniques; General forest ecology; Generalities. Production, biomass. Quality of wood and forest products. General forest ecology; Grasslands; Heterogeneity; Methods and techniques (sampling, tagging, trapping, modelling...); Seed dispersal; Spatial heterogeneity; Subtropical grasslands; Synecology; Terrestrial ecosystems; Vegetation
• ISSN: 0304-3800; 1872-7026
• DOI: 10.1016/j.ecolmodel.2014.07.024

•We model dynamics of forest–grassland mosaics and fire in south Brazil.•We developed a spatially explicit dynamic global vegetation model.•Under frequent fires but implicit space, the potential vegetation is forest.•Seed dispersal delayed spatial dynamics even in the absence of fire.•Spatial patterns of vegetation patches determined long-term stability of alternative states. Vegetation changes, such as shrub encroachment and forest expansion over grasslands, prairies and savannas have been related to changes in climatic (mainly rainfall and temperature) and atmospheric conditions (CO2 concentration). However, a longstanding question in ecology is how mosaics of forests and open-canopy ecosystems could persist over millennia in sites where climatic conditions favor forests. Here we tested the influence of interactions between grass-tree competition, environmental heterogeneity (topography), seed dispersal, initial density and spatial aggregation of vegetation patches and disturbance behavior (fire) on the long-term coexistence of forests and grasslands in South Brazil. For this, we incorporated the adaptive dynamic global vegetation model (aDGVM) into a spatially explicit modeling approach (2D-aDGVM). Our results showed that recurrent disturbance related to grasses such as fires plays a key role in maintaining the long-term coexistence of forests and grasslands, mainly through feedbacks between disturbance frequency and grass biomass. Topographic heterogeneity affected the rate of forest expansion by adding spatio-temporal variability in vegetation-fire feedbacks. However, the spatial pattern and connectivity of fire-prone (grasslands) and fire-sensitive (forest) vegetation patches were more important to maintain the long-term coexistence of both alternative vegetation states than the initial proportion of forest and grasslands patches. The model is the first individual-based DGVM to consider the combined effects of topography, seed dispersal and fire spread behavior in a spatially explicit approach.