Skip to content Skip to navigation
Dahlin KM. 2012. Ecosystem assembly in California: Linking pattern to process at the landscape scale. PhD Dissertation, Department of Biology, Stanford University.
Year Published: 2012

The processes controlling ecosystem assembly are diverse and interacting. While landscape ecologists and geographers tend to focus on the relationships between plants and the environment in space ('environmental filtering'), community ecologists often emphasize species interactions over time ('competition', 'migration', 'facilitation'). Here I integrate concepts from community and landscape ecology with geostatistics and data from the Carnegie Airborne Observatory to better understand what observed patterns in vegetation can tell us about ecosystem assembly processes. This work took place in two California ecosystems, where the mediterranean-type climate has helped to generate a distinct mosaic of grassland, shrubland, and forest. Chapter 1 focuses on understanding controls on aboveground biomass (AGB) in a small protected area (Jasper Ridge Biological Preserve, Woodside, CA). One conclusion from this work is that mapped environmental gradients were not the clear dominant drivers of AGB patterns, suggesting that dynamic processes like disturbance, dispersal limitation, and limits to establishment may be critical drivers of patterns in this system. Chapter 2 explores the controls on greenness and vegetation structure on a much larger scale (Santa Cruz Island, Santa Barbara County, CA). I show that environmental gradients do explain some of the variation in this recently disturbed landscape, but that grazing intensity and spatial autocorrelation are also important predictors. Chapter 3 returns to Jasper Ridge to consider the links between plant functional types, foliar chemistry, and environmental gradients. I conclude that at the landscape scale environmental filtering can explain some of the observed patterns in chemical traits, but that researchers' efforts to confirm or eliminate possible explanations for observed trait patterns are likely closely tied to the specific ecosystem or community in which they are tested, and results can vary even within a single landscape. link to publication