NCEAS Project 3540

The ecology of steppe and grassland vegetation alliances of the Columbia River Basin (Hosted by NCEAS)

  • Michael D. Jennings

ActivityDatesFurther Information
Graduate Student1st October 1999—30th June 2001Participant List  

There is broad agreement in the U.S. about the need to manage the nation's biological resources with an understanding of the biogeographic and ecological context of a given management decision as well as the context of any particular biotic element in question (e.g., National Environmental Policy Act of 1969, National Research Council 1993, 2000). To do so, however, requires spatially comprehensive as well as time series information of species distributions. In this regard, the importance of detailed vegetation maps that cover large regions is substantial (Scott et. al 1993, Margules and Pressey 2000, Jennings 2000). Since such maps use remote sensing of the vegetation canopy as a primary input, it is probable that direct characterization of the finest thematic unit will continue to rely on identification of dominant species. Most maps expressing dominant vegetation types label and define them as "alliances" under the National Vegetation Classification (FGDC 1997). Given the importance of vegetation alliances as units to model and manage biodiversity, there has been surprisingly little work on understanding their behavior and rules of assembly. This study focuses on opportunities for developing a better basic understanding of alliances as a synthetic construct, as well as their utility for conservation and as a vehicle for communicating ecological complexity. Are there additional ways to usefully categorize the behaviors or structures of individual alliance types? Are there general differences between the environmental response patterns of the plant species used to characterize an alliance and the other member species? Can individual alliances be characterized by a suite of functional morphological traits, the physical adaptations of their constituent species to past environments, linking species composition directly to environmental forcings? Could understanding and predicting the effects of climate change on biodiversity be enhanced by knowing the explicit relationship between certain plant functional traits or plant functional types (Smith et al. 1997, McIntyre et al. 1999) and alliances? How does the alliance concept fit with contemporary species assemblage theories, for example, the core and satellite hypothesis (Hanski 1982), the concept of nested hierarchies (Kolasa 1989), or of nestedness (Wright and Reeves 1992)? Might additional metrics of alliances and their constituent parts offer new models for explaining patterns of species distributions along gradients? This study attempts to answer these and other related questions by assembling a large database of vegetation plots, climate, soils, and morphological traits in the steppe and grassland ecosystem in the Columbia River Basin of North America, covering about 30 million hectares. Gradient analysis, ordination, generalized additive models, and fuzzy set theory, will be applied to about 20,000 vegetation plot records, a 500m grid surface of multiple daily meteorological parameters, soils, and importantly, detailed morphological traits for each plant species. Results will be used to establish conceptual models for the conservation of these habitats and to show how this type of data-driven approach to vegetation can lead to a robust predictive capability for managing biodiversity.

TypeProduct of NCEAS Research
Data Set Jennings, Michael D. 2006. Multisource field plot data for studies of vegetation alliances: Northwestern USA. (Online version)