Structure, function and cost-effective revegetation for soil stabilisation in a Mediterranean ecosystem

[Truncated abstract] Southwest Western Australia is one of the most biologically diverse parts of the earth. However, with only 11% of the original primary vegetation remaining, there is a need to preserve it and restore disturbed areas. This is especially important in declining Banksia woodlands on the Swan Coastal Plain, where soils are highly infertile, and weed invasion poses a serious threat. I evaluated revegetation techniques with respect to ecosystem structure, function, biotic and abiotic factors and thresholds, as well as timeliness and cost-effectiveness. I studied a diverse range of projects in areas recently disturbed by construction activities or fire, or restored as offsets for construction in other areas. From the specific details of each project, I aimed to extract general restoration principles for soil stabilisation in disturbed areas. Most local restoration project completion criteria mandate structural parameters such as minimum native plant stem density, projected foliage cover and species richness, with upper limits on weed foliage cover. While the typical period of 3.5 years for satisfaction of completion criteria may be achievable, it is unrealistically short for complete restoration. Restoring ecological function is also important, but local completion criteria do not generally require its monitoring. In Project 1, an experiment in the construction of a seasonal wetland found composted soil conditioner and mulch beneficial to native plant establishment after direct seeding, limiting the potential for soil erosion. Establishment was greatest with the lowest application rate of mulch. In Project 2, large-scale use of woven plastic weed mat allowed the re-establishment of native plants in a seasonal wetland. Prior efforts, spanning 5 years, using conventional revegetation techniques, had been unsuccessful due to abundance of non-native herbaceous species. In Project 3, anecdotal reports of slow Banksia attenuata growth rates were confirmed by measurements of planted seedlings over 7½ years. In a shorter-term experiment, low first-year Banksia attenuata seeding survival, around 30%, was observed. Low-phosphorus slow release fertiliser tablets promoted growth in height and root development significantly, without apparent adverse effects. Faster shoot growth and more extensive root development contribute to soil stability...