Factors affecting invasion and persistence of broom Cytisus scoparius in Australia.
Abstract
The effective management of invasive plants requires an understanding of their invasiveness in contrasting environments. The effects of disturbance on recruitment of broom C. scoparius, a European shrub invading parts of Australasia, was assessed in a 7-year experiment carried out in two countries, respectively, in its native and exotic ranges. This paper presents the results for Australia and their implications for the biological control of this weed. Disturbance treatments involved removal (cut or cut plus cultivation) and grazing effects on broom seed banks and recruitment, in mature and immature stands and in three habitats, ranging from improved, largely exotic pasture to unimproved native grassland. The time between recruitment and flowering varied between 5 and 7 years depending on habitat, the slowest being in native unimproved grassland. In cultivated plots, regeneration was faster in immature compared with mature broom stands. Following broom removal, whether the ground was cultivated or not had little effect on the speed of regeneration, indicating weak suppression from the ground flora in uncultivated plots. In contrast, grazing tended to increase the rate of regeneration. The timing of seedling establishment had little effect on survival to reproduction, in marked contrast to European populations. The overall rates of seedling mortality were greater in Australia compared with Europe. A simulation model, parameterized from the field data, was used to explore the likely impact of biological control from pre-dispersal seed predation. The simulation studies demonstrated large habitat-specific differences in seed rain requirement for broom persistence and the potential impact of seed predators. Seed rain requirements were fivefold less in improved exotic pasture than in native grassland. Seed loss of 62% was theoretically sufficient to suppress broom in native grassland, whereas >97% was required in exotic pasture. This study suggests that successful broom control in Australia is likely to be site-dependent, because while broom seed has greater opportunities for recruitment in Australia than in Europe, immature plant mortality can be very high, particularly in native grassland.