Suppression of Knapweed Invasion by Crested Wheatgrass in the Dry Interior of British Columbia

Abstract

We resampled an experimental plot established 11 years previously in the dry interior of British Columbia to test the ability of crested wheatgrass and Russian wild rye to suppress the invasion of diffuse knapweed. Knapweed density was high in non-seeded plots, moderate in Russian wild rye plots, and very low in crested wheatgrass plots. Watering experiments indicated that lack of soil moisture resulted in high seedling mortality and prevented knapweed invasion into crested wheatgrass plots. Diffuse knapweed reinvaded a similar experimental area in a higher rainfall region of B.C., which shows that the same cultural practices will have different effects on knapweed suppression under different climatic regimes. Diffuse knapweed (Centaurea diffusa) has become an important rangeland weed in the dry interior of British Columbia (Harris and Cranston 1979). The rapid invasion of rangeland by this weed shows it to be a vigorous competitor with the native vegetation. But an unanswered question is whether knapweed can invade range in good condition, or only range that has been disturbed or overgrazed. As part of a study to investigate this question, we examined the status of experimental areas set up by the British Columbia Ministry of Agriculture and the Agriculture Canada Research Station, Kamloops, in 1967, 11 years prior to our visit. These areas were treated with picloram (4 amino-3,5,6, trichloropiclinic acid) and planted in crested wheatgrass (CWG), Agropyron cristatum, and Russian wild rye (RW R), Elymusjunceus. One experimental area is near Cache Creek, B.C., where the precipitation is only 20 cm a year, and another near Pritchard, B.C., which is somewhat moister with an annual precipitation of approximately 33 cm a year. Both areas are in the brown chernozemic soil zone and the Ponderosa pinebunchgrass biogeoclimatic zone (Farley 1979) and the lower grassland (Agropyron-Artemisia) vegetation zone (van Ryswyk et al. 1966). These experiments allowed us to investigate the ability of diffuse knapweed to invade rangeland planted in grass and protected from grazing. The difference in moisture regimes between the areas provided an additional dimension to the analysis. Specifically we wanted to know if diffuse knapweed would invade grassland in the absence of grazing. Authors are at 115 Connaught Crescent, Regina, Canada S4T6M8; and Institute of Animal Resource Ecology and Department of Plant Science, University of British Columbia, Vancouver, Canada V6T IW5. Authors wish to thank Dr. Alastair McLean of the Range Research Station, Agriculture Canada, Kamloops, B.C., for providing facilities and advice to make our work on on knapweed possible. Kate Lindsay and Peter Morrison provided field assistance and Mr. Alf Bawtry, B.C. Ministry of Agriculture, first told us about this location and provided crucial information about the original experimental design for which we are most grateful. This wvork was supported by an Agriculture Canada EMR grant and research contract. Manuscript received June 9, 1980. Study Plots The study sites are described by Hubbard (1975) and additional valuable details were provided by Mr. Alf Bawtry (B.C.M.A., Kamloops). The Cache Creek site is adjacent to the Bonaparte Indian Reserve, approximately 6.5 km N of Cache Creek. In spring 1967, a knapweed infested area was treated with picloram at 0. 14, 0.28, 0.42 and 0.56 kg/ha and planted with Russian wild rye and crested wheatgrass using a rangeland drill. Similar treatments were applied in 1969 at Pritchard. We first visited the Cache Creek site in early May 1978 and the nonuniform pattern of knapweed invasion was so striking that we decided to quantify the pattern and set up seeding experiments in an attempt to determine what was limiting the spread of diffuse knapweed at the site. The Pritchard plot was completely reinvaded by knapweed and we only record that observation and discuss the significance here.