4. Discussion
Urban areas are thought to promote the homogenisation of biodiversity by simplifying habitats and creating sterile landscapes which are either unsuitable for most species or which only favour a handful of dominant, urban-exploiting species (Melles, Glenn, & Martin, 2003). Instead, in this study we found that the combination of a river, a heterogeneous urban area and an adjacent national park makes the Liesbeek catchment surprisingly biodiverse for a small and highly disturbed urban river catchment, both in terms of species richness and functional diversity. The contribution of these features to local biodiversity highlights the importance of green space networks and ecological infrastructure in enhancing urban ecosystems. In this case, the river in particular is responsible for the occurrence and persistence of certain species and functional groups.
A total of 95 species were observed in this study, but over
120 bird species are known to occur within the 327km2 study
area. In comparison, at least 367 bird species occur within the greater City of Cape Town (Holmes, Rebelo, Dorse, & Wood, 2012). Thus, in an area of less
than a sixteenth of the city’s size, the Liesbeek catchment represents almost a third of all the species which occur within the city. While the connectivity and proximity between natural patches in the catchment help to increase species richness, increased habitat heterogeneity also provides more niches for specialist species (Fernandez-Juricic & Jokimaki, 2001; Shanahan, Miller, Possingham, & Fuller, 2011). The catchment was also revealed to be rich in terms of functional diversity, with all nine groups represented. These groups were present in almost the same proportions as in the bird community of the whole of southern Africa. Moreover, while urban areas are expected to experience a relative decline in the number of insectivores, pollinators, raptors and scavengers (Child et al., 2009; Faeth, Bang, & Saari, 2011), all of these groups were well-represented within the catchment. Given that invertebrate abundance can be severely affected by human disturbance, the presence of several uncommon smallbodied insectivores is a particularly good sign of ecosystem health (Larsen, Sorace, & Mancini, 2010). Raptors are also significant for ecosystem functioning because each species, though superficially similar, may occupy distinct niches and can also act as umbrella species (Palomino & Carrascal, 2007; Sekercioglu, 2006). In this study, almost every raptor species known to occur locally was recorded.
Our results might be interpreted as indicating a surprising degree of resilience in the functional composition of the bird community. If we consider the mechanisms underlying community structure, much of the observed diversity in this study can be directly attributed to the river. Of the 95 species recorded in the catchment, 64 species were recorded on the river itself, even though less than a third of the sites were located on the river. Although surrounded by an industrial area and several major highways, species richness was highest at the lower reaches of the river. Bird communities in these sections of the catchment were distinct in that they were strongly shaped by the presence of the two rivers and associated wetlands. The RLQ and fourth-corner analyses pointed to certain functional groups favouring this area. For example, nutrient movers and scavengers appeared to respond positively to areas of low elevation, high visibility, close to the river and far from the river’s source – all characteristics of this particular section of the river. Despite covering a small area in the catchment, this portion of the river and catchment is clearly a very important feature in the landscape for birds. It contains a bird sanctuary and open stretches of wetland where the Liesbeek and Black Rivers meet, and an abundance of waterfowl and aquatic species. This supports previous studies showing that richness of aquatic species increases with higher river discharge (Xenopoulous & Lodge, 2006) and that wetlands support greater bird richness than adjacent uplands (Robinson, Tockner, & Ward, 2002; McKinney, Raposa, & Cournoyer, 2011).