A Method for replacing serrated tussock (Nassella trichotoma) with kangaroo grass (Themeda triandra) in lowland native grassland remnants

A 'spray and hay' method was developed which replaces the noxious weed serrated tussock (Nassella trichotoma (Nees) Arech.) with indigenous kangaroo grass (Themeda triandra Forrsk.) in partly disturbed remnants of western (basalt) plains indigenous grassland, a community listed as vulnerable to extinction. Investigations were carried out on Melbourne's north-west urban fringe. Refinements of the 'spray and hay' method resulted in low densities of other weeds amongst the newly established T. triandra swards. First steps in the method involved removal of weeds by slashing and follow-up treatment with either of the herbicides, glyphosate (as Monsanto Roundup®) or atrazine (as Nufarm Nutrazine®). These steps resulted in close to 100% kill of mature N. trichotoma plants. Greater than 98% replacement of N. trichotoma with T. triandra was achieved by thatching over herbicide-treated plots with seed-bearing T. triandra hay harvested from remnant grasslands and then removal of the hay several months later by either burning or physical removal. The type and timing of herbicide application, thatching and removal of thatch were found to be central for successfully establishing competitive densities of T. triandra seedlings and minimizing re-establishment of N. trichotoma and other weed seedlings. A treatment set involving: slashing weed biomass in summer (January), herbicide application in autumn (April) followed by thatching with seed-bearing T. triandra hay in winter (July) and removal of thatch in spring (October) produced the best results. Assessing the seed content of hay and germinability of seed prior to revegetation were also important for calculating the amount of hay laid and subsequent seedling densities established. The seed content and germinability of seed-bearing hay was found to vary markedly in samples harvested in three different years, across discrete remnant grassland sites and even within undisturbed grassland sites. Reasons for the success of the method, and why other variations are less successful are discussed, as is the wider application of the method for weed control and replacement with indigenous grasses in lowland grassland remnants