The management of biological invasions, which pose a growing threat to natural resources and human well-being, is critical for reducing associated negative impacts. As part of the process of developing a strategy for the management of biological invasions in the South African National Biodiversity Institute’s (SANBI) gardens, we collated a list of alien plant species from 13 gardens as part of a situational analysis. We requested lists of alien plant species recorded in each of the SANBI’s gardens. A total of 380 records included 225 alien plant species belonging to 73 families. A significant number of species were intentionally introduced through horticultural trade as ornamentals (49%; n = 225), while 20.9% were consumed as either food or medicine by humans. Plant life forms included woody and herbaceous plants, graminoids, succulents and ferns. Herbaceous (42.7%; n = 225) and woody plants (3.8%) were the dominant life forms. The Walter Sisulu National Botanical Garden had the highest number of alien species (88 species), followed by Kirstenbosch (61 species) and Pretoria (46 species) National Botanical Gardens, with herbaceous species constituting the largest number in all gardens (i.e., 47, 19, and 27 species, respectively). The number of species that we recorded that were listed in the National Environmental Management: Biodiversity Act (NEM: BA) (Act No. 10 of 2004): Alien and Invasive Species Regulations’ categories were not notably different from the number of unlisted species (58.2% vs. 42.8%). The number of species listed in the different categories varied significantly across the different gardens, with a significantly higher number of unlisted species and of Category 1b species in the Walter Sisulu, Kirstenbosch and Pretoria National Botanical Gardens than in other gardens. That a significantly larger number of alien species originated from South America points to the need to improve biosecurity controls on existing relations. The results of this study provided a baseline database to help comparison between successive surveys in future.

Olea europaea subsp. africana (Mill.) P.S. Green (medium‐sized tree species known as “African wild olive”), provides important ecological goods and services for sustaining frugivores in the grassland biome in South Africa. We speculate that O. europaea subsp. africana's population has been declining due to habitat loss and exploitation for domestic benefits suggesting an unrecognized conservation threat. Therefore, the study aimed to investigate the anthropogenic conservation threats for O. europaea subsp. africana in the Free State, South Africa and to determine the potential importance of seed dispersal effectiveness in the restoration of the species in the study area. Overall, the results showed that 39% of the natural habitat range has been transformed by human‐mediated activities. Agricultural activities accounted for 27%, while mining activities and human settlement accounted for 12%, of natural habitat loss. In support of the study predictions, seeds of O. europaea subsp. africana had significantly higher germination and germinated faster after passing through the mammal gut (i.e., 28% and 1.49 per week), compared to other seed treatments (i.e., over 39 weeks). However, there were no statistically significant differences between seed germination of the bird‐ingested seeds, with intact fruits as the experimental control, although both were significantly greater than the de‐pulped seeds. Potential seed dispersal distances by birds were relatively larger, ranging from 9.4 km to 53 km, than those of mammals (1.5 km–4.5 km). We propose that the O. europaea subsp. africana's habitat range may have been declining, and since it is a keystone plant species, we recommend that the complementary seed dispersal services by birds and mammals could be important for its recruitment and restoration in the degraded habitat.

In this study, we documented the diversity of bird species in the Eastern Cape coastal nature reserves (i.e., Hluleka, Dwesa, Silaka and Mkhambati nature reserves), and determined the potential role of each bird species in habitat maintenance using two functional traits (i.e., body mass and feeding mode) as the function's proxy. We applied the timed species count approach during bird observations, coupled with drive‐by surveys to maximise spatial coverage of each nature reserve over four years. To evaluate functional diversity, bird species were classified based on functional traits such as the adult body, and their potential ecological role derived from their feeding mode and habitat associations. Over 864 h, we accumulated 818 bird records containing 178 different bird species that were classified into 58 families with 32 species occurring in all nature reserves. Shannon–Wiener Diversity Indices showed very high overall species diversity across the nature reserves (H > 3.5) with no differences detected across sites. Although no significant correlations between vegetation changes measured through Normalised Difference vegetation Index (NDVI) in each nature reserve and the number of bird records, forest bird species were dominant (42.1%; N = 178) throughout years of observation and diversity remained high (H > 3.5). Bird species abundance only increased significantly across all nature reserves during 2018–2019. All four nature reserves had a similar distribution of bird functional traits with both high functional richness (FRic = 1), and divergence (FDiv = 0.8) and moderate evenness (FEve = 0.4). Multiple Correspondence Analysis (MCA) demonstrated a positive correlation between bird sizes and functions with large birds mainly associated with predators and carrion. Small birds and medium birds had a similar composition of species in terms of functionality being seed dispersers across the nature reserves. A significant effect that insectivores and carrions displayed in MCA plots, suggest the availability of indirect pollination services. Despite extreme drought conditions across the country in 2019, NDVI levels remained largely consistent over time in these four reserves; and thus, they offer important refuge for birds during extreme climatic conditions such as drought.

Vertebrate-mediated seed dispersal plays a crucial role in enabling the spread of invasive alien plants that threaten biodiversity in ecosystems such as the southern African grassland biome. Understanding the dynamics of the seed dispersal process by animals is important for assessing invasion risk and designing management interventions. This study investigated whether the southern African grassland biome is vulnerable to future plant invasions facilitated by frugivorous vertebrates in two known plant invasion hotspots in the Free State province: Boshof, dominated by Opuntia engelmannii (Cactaceae), and Fouriesburg, invaded by Pyracantha angustifolia (Rosaceae). We (1) monitored, documented, and compared the diversity of dispersal vector species in sites invaded by non-native woody and cactus species with non-invaded sites; (2) quantified the seed rain from vertebrates’ faecal samples and classified seeds and disperser species into functional groups to identify drivers of plant invasion of the grassland biome; and (3) assessed the effectiveness of vertebrate-mediated seed dispersal by measuring their post-ingestion viability. We predicted that frugivorous vertebrates are key drivers of fleshy-fruited plant invasions in this biome. We found taxonomically diverse dispersers (i.e. 27 families), with Bovidae and Numididae being dominant. Overall vector species diversity was high (H > 3.0), assuring effective population recruitment, although moderate in invaded sites (H > 2.5), and lower in non-invaded sites (H < 2.5). Over 9 months (March–November 2024), mammals dominated dispersal services (74%; N = 328 visits), followed by birds (26%), supporting our study prediction. We noted a high seed rain (i.e. 107,031 seeds) collected from 11,295 faecal samples associated with 43 vertebrate species. Invaded sites experienced significantly higher seed rain than non-invaded sites (P < 0.001) and were dominantly comprised of non-native species. Many seeds were medium-sized (1–5 mm), typical for successful invaders, and had a hard seed coat. Germination trials showed that ingested P. angustifolia seeds retained viability (46.7 ± 8.61%), similar to de-pulped controls (P = 0.53), while O. engelmannii seeds significantly lost viability (P = 0.002). The predicted seed dispersal distances were greater than 15.7 km for birds and ranged between 1.7 km and 8.6 km for mammals. We concluded that the grassland biome is vulnerable to further invasion by non-native plants dispersed by vertebrates, coupled with their long dispersal distances, and that integrating management of dispersal pathways into control plans is essential to optimise control efforts.