Resource efficient control of Elymus repens

Elymus repens is a perennial grass weed that causes great yield losses in a variety of \ncrops in the southern and northern temperate zones. Primary control methods for E. \nrepens are herbicides or intensive tillage, both of which have a number of negative \nside-effects, e.g. herbicides can contaminate groundwater, and tillage can cause \nincreased nitrogen leaching. The aim of this thesis was to investigate how to make non- \nherbicide control of Elymus repens more resource efficient in terms of less energy \ndemanding soil cultivation and reduced nitrogen leaching. Three field experiments \nwere used to test cover crop competition, mowing and different types of optimised \ntillage techniques and timing, as well as the combination of under-sown cover crops \nand mowing or row hoeing. The growth, biomass allocation and morphological \nresponses of E. repens to competition were studied in a greenhouse experiment. \n \nThe effect of competition from under-sown cover crops on E. repens seems to \ndepend greatly on the cover crop biomass achieved. At high biomass levels, the cover \ncrop can be highly suppressive (Paper IV) and reduce nitrogen leaching (Paper III), \nwhile at low levels they can still provide benefits such as reduced E. repens shoot \nbiomass and increased subsequent cereal yield (Paper I). However, a low-yielding red \nclover cover crop increased E. repens rhizome production by 20-30%. Under-sown \ncover crops were successfully combined with both mowing and row hoeing (Paper I & \nIII), but while repeated mowing reduced E. repens rhizome production by 35% it could \nnot be shown to give a competitive advantage to the cover crops over E. repens (Paper \nI). However, the low nitrogen leaching and reduced downward transport of nitrogen \nwhen mowing or row hoeing was combined with under-sown cover crops make them \ninteresting control methods for future research. Delaying tine cultivation by a few days \nafter harvest did not reduce E. repens control, but a delay by 20 days tended to result in \nhigher E. repens rhizome biomass and shoot densities, compared to performing it \nwithin a few days of harvest. Repeated tine cultivation did not improve control of E. \nrepens or increase subsequent cereal yield, compared to a single cultivation directly \nafter harvest. Repeated cultivation during autumn should therefore not be used \ncategorically, but only when there is reason to believe the shoots will pass the \ncompensation point due to the autumn conditions. We conclude that a site specific \napproach is necessary to achieve resource efficient control of E. repens.