In Mediterranean annual grasslands, vegetation communities are sensitive to environmental variation and historical contingency (priority effects). Keen observers in arid and semi-arid systems have long recognized the importance of these drivers, employing qualitative state and transition models (STMs) to guide restoration and management practice. However, there are few empirical tests of assumptions made by these models, related to the number and composition of vegetation “states” and frequency of transitions between them. In light of predicted climate change effects across Mediterranean grassland systems, quantitative approaches to generating STMs may better predict long-term vegetation patterns used to identify key thresholds and windows of management action. Using experimental plantings of three classically-defined groups of grassland taxa – native perennial grasses, naturalized annual forage grasses, and noxious invasive species – we examined transitions between vegetation types over a period encompassing typical precipitation patterns (2007-2010), drought (2011-2015), and post-drought recovery (2016-2019). Through unsupervised classification and multi-state modelling, we assessed the value of these group definitions and quantified transition probability as a function of historical contingency and climatic variation. We found that classic California grassland STMs do not account for a distinct 4th group of species that respond strongly to drought. We also found that vegetation states differed significantly in their transition frequencies; native species, in particular, were resilient to state change once established, but rarely invaded other communities. These results contribute to the growing evidence of contingency and environmental effects in annual grasslands, in addition to development of quantitative approaches that complement existing management tools.
Audio/Video, Conference Presentation, SER2019
Pre-approved for CECs under SER's CERP program
Society for Ecological Restoration