Performance of native succulents, forbs, and grasses on an extensive green roof over four years in subtropical Australia

To achieve enduring low-maintenance green roofs that provide maximum benefit in the subtropics, plants need to be well adapted to the local climate with extensive green coverage. The use of native plants contributes to ecological restoration, yet there is very limited information available on their performance in the subtropics. This study aimed to determine the performance of nine species of Australian native forbs, succulents, and grasses in monofunctional and functionally-mixed plant communities on an extensive green roof in Brisbane over four years. Green coverage and survival were monitored via image analysis and observational visits. Forbs established slowly, but successfully (82 % green coverage, 67 % survival after four years) with species Hibbertia scandens (100 % survival) and Dianella brevipedunculata (58 % survival and extensive recruitment) performing the best. Grasses, including Themeda triandra, Poa labillardieri and Cymbopogon refractus, achieved a fast initial green coverage (47 % after 10 months), but showed high seasonal variability and declined to 2 %. Contrary to expectations, succulents performed poorly (<5 % green coverage, 14 % survival). Green coverage had an inverse trend to plant survival in successful plots (forbs, functionally-mixed) indicating a gap-filling effect by high performing species. Survival was more strongly linked to individual species than to functional type or plant community arrangement. Based on the results of this study, we recommend consideration of functionally-mixed green roof plantings with a variety of indigenous plants including forbs for low maintenance and resilient green coverage in the subtropics. Habitats that provided suitable plant species for this study were coastal headlands, dunes, montane areas, and open eucalypt forests. Analogous habitats in other regions can be explored to find new native plant species to increase the plant palette of potentially suitable species. It would be beneficial to trial more plants from these habitats over multiple years to increase the knowledge of suitable species available to green roof designers.