Prairie Restoration Research

Regional Strategies for Restoring Native Prairies

A multi-site, long-term, collaborative research project funded by the Priscilla Bullit Collins Trust of The Nature Conservancy


Principal Investigators: Tom Kaye, Institute for Applied Ecology;
Peter Dunwiddie, School of Forest Resources, University of Washington

Want more information? Email Tom

2010  Final Technical Report


Invasive plants, especially non-native perennial grasses, pose one of the most critical threats to protected prairies in the Willamette/Puget/Georgia ecoregion. These prairies are among the most endangered ecosystems in North America, and support many imperiled species. Invasive species reduce native diversity and alter vegetation structure, fire regimes, soil characteristics, and faunal diversity. Our current knowledge regarding the effectiveness of techniques for controlling many herbaceous invasives, especially in sites that retain a significant component of native vegetation, is largely anecdotal or based on results from only a few site-specific studies.

Progress has been hampered by several problems. Funding for prairie restoration has been directed primarily towards on-the-ground actions, with very little devoted to developing the scientific understanding that underpins successful adaptive management. Most studies have been of relatively short duration, and involve only single treatments. Few have employed combinations of treatments, an approach that is likely to produce more effective, synergistic effects, as suggested by results from studies in other systems. We have examined these approaches in a comprehensive manner across a geographic range of sites, and over a time period sufficient to evaluate longer-term system restoration. Our project is designed to address these issues using a comprehensive, multi-site collaboration of scientists and land managers.

back to top

Research Objectives

  • Evaluate and improve strategies for controlling the abundance of invasive non-native herbaceous weeds, while maintaining or enhancing the abundance and diversity of native plant species.
  • Develop an approach to generalize these results so that they can be applied by land managers engaged in prairie stewardship throughout the region.

back to top

Research Approach

At 11 sites in prairie habitats in Oregon, Washington, and British Columbia (see Research Sites, below), we have identical experimental blocks to test different restoration methods.  Experiments began in 2005 and are planned through 2010. Each block consists of twenty 5 x 5 meter plots: 4 replicates of each of 5 treatments combinations.  Each plot is split in half and native seed was added to one half in fall 2006.  In addition to these small replicated experiments, at many of the sites we also have a larger, unreplicated treatment area (typically 100 x 100 meters) to test how one specific treatment combination works at larger scales.

Our treatment combinations were developed at a workshop at the start of the study in 2005 which included managers and scientists.  The treatments include combinations of Poast, a grass-specific herbicide, to reduce dominant exotic grasses; spring or fall mowing, to reduce thatch, and reduce seed production and stored reserves of exotic grasses; burning, to reduce thatch and moss and prepare seed-beds for germination; and post-burn application of glyphosate (a broad-spectrum herbicide), to reduce broad-leaf weeds that typically resprout quickly after fire.  The five treatments are:

Treatment Schedule
A Spring Poast application 2005-2007; Burn followed by glyphosate in fall 2006 and fall 2008
B Initial spring mow 2005; Burn followed by glyphosate in fall 2006 and fall 2008
C Spring and fall mow, all years
D Spring Poast application 2005-2007; fall mow 2005-2008
R Control; no treatment

Native seed was added in the fall/winter of 2006  after the burn and glyphosate treatment to one half of each plot, including controls.  Native species added wereFestuca roemeri, Danthonia spicata or californica, Achillea millefolium, Eriophyllum lanatum, Lomatium utriculatum or nudicaule, Plectritis congesta, and Ranunculus occidentalis. Seeding was repeated in fall 2007

Data on plant communities (percent cover of all vascular plant species) is collected every spring.  Other data collected include soil nutrients, soil moisture, plant biomass, and light interception.

Research Sites

Our research is conducted at 11 sites in Oregon, Washington, and British Columbia. All sites are generally categorized as upland prairie or oak savannah in the Willamette Valley/Puget Trough/Georgia Basin Ecoregion. Our research plots are found at:

  • Pigeon Butte, Finley Wildlife Refuge, OR (USFWS)
  • Bellfountain Road, Finley Wildlife Refuge, OR (USFWS)
  • Ft. Hoskins, OR (Benton County)
  • Scatter creek, WA (WDFW)
  • Mima mounds, WA (WDNR)
  • Glacial Heritage, WA (WDNR)
  • Morgan Property, WA (TNC)
  • South Weir Prairie, Ft. Lewis, WA (US Army)
  • Triangle Prairie, Ft. Lewis, WA (US Army)
  • Smith Priaire, Whidbey Island, WA (AuSable Institute)
  • Cowichan Preserve, BC, Canada (Nature Conservancy of Canada)

Collaborators and Partners

Name Affiliation
Mark Wilson Oregon State University, Associate Professor
Deborah Clark Oregon State University, Faculty Research Associate
Manuela Huso Oregon State University, Consulting Statistician
Elizabeth Borer Oregon State University, Associate Professor
Eric Seabloom Oregon State University, Associate Professor
Stephen Griffith
Machelle Nelson
USDA-Agricultural Research Service, Research Scientist
Andrew MacDougal University of Guelph, Associate Professor
Tim Ennis
Irvin Banman
The Nature Conservancy of Canada
Jock Beall US Fish and Wildlife Service, Refuge Biologist, Finley Wildlife Refuge
Ed Alverson
Jonathan Soll
Eric Delvin
Carrie Marschner
Mason McKinley
The Nature Conservancy
Dave Hays Washington Department of Fish and Wildlife
David Wilderman
Birdie Davenport
Washington Department of Natural Resources
 Al Kitzman Benton County Parks Department
 Rod Gilbert  Fort Lewis
 Karen Reagan  University of Washington