WWETAC Projects

Project Title:  Stand and landscape-level determinants of invasive species abundance on National Forests in northeast Oregon

Principal Investigator:  Andrew Gray, USFS – Pacific Northwest Research Station, Corvallis,OR


Collaborators:   Thom Whittier, Oregon State University; participants in the Intermountain West Invasive Species Model Workshop

Key Issues/Problem Addressed:                              

Invasive plants are having a growing impact on range and forestland ecosystems, threatening biodiversity and making management more difficult and expensive. Better understanding of where on the landscape these plants are successful and the conditions that pre-dispose sites to being invaded will improve our ability to anticipate, mitigate, and eradicate invasives. The objective of this study is to examine the association between vegetation type, climate, site quality, and management history and the distribution and abundance of selected invasive species at multiple scales on National Forest lands in eastern Oregon.

Setting and Approach:  

This study will rely on plant measurements made on an extensive network of permanent inventory plots initiated by Region 6 and continued since 2001 with standardized regional Forest Inventory and Analysis (FIA) protocols. In addition to estimates of the distribution and abundance of nonnative invasive plants, a rich set of information on native plant communities, stand density and productivity, management and disturbance history, and topography is available to develop predictive models of plant invasion. Intersection of plot locations with climate models and remotely sensed classifications provide additional information on attributes important to the invasion and spread of invasive plants (and for predicting native plant communities). Stand, landscape, and climate variables will be analyzed in logistic models of species presence and mixed models of species cover to determine predictors of abundance of individual species, as done in Gray (2005). The strength of the data lies in the representative sample that crosses broad gradients in site conditions to inform statistical models.

The sample grid density is one plot per 1,800 ac outside Wilderness and one plot per 6,000 ac within. On National Forests, all lands (forest and non-forest) are sampled; on other ownerships only forested lands are sampled at the FIA base grid density of one plot per 6,000 ac. We propose to conduct this study on the Malheur, Umatilla, and Wallowa-Whitman National Forests, with 1,975 plots with vegetation data collected to date. On each plot, all plant species present at ≥3% cover, and selected invasives, are recorded on four 24 foot radius subplots. Data will be analyzed at multiple scales, from plots to watersheds to ranger districts to ecoregions, to assess patterns of invasion and strength of different explanatory variables.

Timing of analyses depends in part on the direction and nature of the collaboration developed in the Intermountain West Invasive Species Model Workshop, but the compilation and QA of the inventory data and overlay of ancillary variables should be completed by October 2012, with modeling, analysis, and writing of results completed by summer 2013. The basic deliverable will be a report to be published in either a GTR or journal, but collaborators may be able to develop additional tools or information delivery options in concert with WWETAC that would enhance the utility of the results.

Progress to Date:      

Work has expanded to include the Ochoco National Forest and surrounding non-NFS lands. Database of climate, topography, disturbance, stand density, and vegetation data has been compiled and analyses of 3 selected species (cheat grass, bull thistle, and timothy) are in progress. A report on the analyses and findings will be completed by summer 2013.


While local managers have a detailed understanding of invasive plant problems from first-hand experience, this analysis will provide a comprehensive, standardized assessment of conditions across different watersheds, vegetation types, and stand conditions to better quantify and communicate the problem, serve as a baseline for future monitoring,  improve understanding of where future invasions are likely to occur, and potentially point to new geographic or topical areas for research.

WWETAC ID:      FY12NG107