Effect of grazing, spraying, and seeding on knapweed in British Columbia.

Abstract

The effects of late fall grazing, application of picloram (4 amino-3,5,6-trichloro-2-pyridinecarboxylic acid), and seeding on the reinfestation by knapweed (Centaurea diffusa Lam.) were investigated on a knapweed-infested grassland range in southern British Columbia. The seeding treatments were an unseeded control, crested wheatgrass (Agropyron cristatum L.), Russian wildrye (Psathyrostachys junceus (Fisch.) Nevski)), 'Drylander' alfalfa (Medicago sativa L.), and a rangeland seed mix. The spraying treatment was applied (0.56 kg a.i. per ha.) to only the unseeded control, Russian wildrye, and crested wheatgrass treatments. Re-establishment of knapweed and establishment of seeded and indigenous species were observed over 4 years. Knapweed cover never exceeded 10% on sprayed plots but ranged from 35% to 60% on unsprayed plots. Knapweed cover was greater on unsprayed grazed plots and re-establishment was more rapid on sprayed grazed plots than on ungrazed ones. Seeding produced little difference in knapweed cover but crested wheatgrass and rangeland mix (which contained crested wheatgrass) had the lowest coverage of knapweed on unsprayed plots. Russian wildrye did not establish and this failure plus the disturbance created by seeding provided ideal conditions for a surge of knapweed growth. No differences in knapweed cover were detected among seeding treatments on sprayed plots. It is concluded that spraying is far more important than seeding for controlling knapweed, and grazing pressure must be carefully controlled to prolong the effects of treatment.