Towards ecological restoration strategies for Peninsula Shale Renosterveld: testing the effects of disturbance-intervention treatments on seed germination on Devil's Peak, Cape Town

• Main Author: Waller, Penelope Anne
• Other Authors: Anderson, Pippin
• Format: Dissertation
• Language: English
• Published: University of Cape Town 2014
• Online Access: http://hdl.handle.net/11427/6606

The ecological restoration of Peninsula Shale Renosterveld is essential to redress its conservation target shortfall. The ecosystem is Critically Endangered and, along with all other renosterveld types in the Cape lowlands, declared 'totally irreplaceable'. Further to conserving all extant remnants, ecological restoration is required to play a critical part in securing biodiversity and to meeting conservation targets. Remnants of Peninsula Shale Renosterveld are situated either side of the Cape Town city bowl and, despite formal protection, areas of the ecosystem are degraded and require restoration intervention. The body of research addressing restoration in renosterveld is small, yet growing, and seed-based restoration efforts have achieved limited success. This study primarily set out to further understand community recovery through determining the effects of interventions, implemented to mimic ecological drivers, on several seed germination criteria. Thirty-two interventions (comprised of five, crossed factors: burning, tillage, herbicide-application, rodent-exclusion and seeding) were incorporated into a field experiment situated in an area that was most likely ploughed over a century ago and is currently dominated by alien, annual grasses. Additional aspects of the study included determining the physiological status of the seed from 31 harvested species (through viability and germinability tests) towards identifying key restoration species, and, assessing intervention costeffectiveness as a measure of intervention feasibility. The majority of the harvested species exhibited moderate to high levels of viability and germinability and occurred in the middle or upper key-restoration-species index range, indicating their potential for use in future restoration efforts. Seeding contributed considerably to overall community attributes, significantly increasing indigenous seedling density, species richness and canopy cover. Due to the dominance of alien, annual grasses, seeding alone was relatively ineffectual and recruitment was considerably improved when seeding was implemented in conjunction with one or more of the other interventions. These findings indicate that a lack of available seed is not the only barrier to community recovery and that competition exerted by the alien grass component plays a large role in inhibiting seedling recruitment of desirable species. Intervention effectiveness increased with the number of factors per intervention yet, fortuitously, the most effective interventions were not necessarily the most costly. Some interventions resulted in good performances and have the potential to restore self-perpetuating communities with a semblance of ecosystem composition, structure and function. There is a clear and promising way forward incorporating these findings into feasible, implementable, landscape-scale, ecological restoration strategies for Peninsula Shale Renosterveld and potentially other Critically Endangered and 'totally irreplaceable' renosterveld ecosystems.