The study was conducted on two tributaries of Zambezi River: Sikunga and Lisikili tributaries located in the Zambezi Region of Namibia (Figure 1). The river system is highly pulsed and spread out in terms of water volume during the flooding season. The topography of the region is flat terrain. The tributaries normally reach their peak flow between March and May, after which the water recedes until the end of September. During the dry months (November–April), the floodplains are dry and covered in terrestrial grasses (Simasiku & Hay 2023).

The waterbirds community was surveyed at two Zambezi River tributaries: Sikunga channel (FPA) and Lisikili channel (NFPA) during the winter of 2020 (July to November) (see Figure 1). A 2 kilometre transect was laid out at each tributary. Within each transect, five intra-sites were laid 500 metres apart and parallel to the water’s edge. At each site, we enumerated and identified all the waterbirds within a radius of 50 metres using binoculars and guide books by Reynolds and Tye (2015).

Ethical clearance to conduct this study was obtained from the University of Namibia Ethical Committee (No. KMC0001).

A total of 40 carnivorous waterbird species belonging to 9 orders were recorded during the survey (Table 1). Of these 40 species, 10 (25%) predominantly feed on fish, 6 (15%) predominantly feed on insects and 24 (60%) species predominantly feed on both fish and insects. At the FPA site, a total of 35 species were recorded, comprising 10 (29%) species that predominantly feed on fish, 4 (11%) species that predominantly feed on insects and 21 (60%) species that feed on both fish and insects. At the NFPA site, a total of 31 species were recorded comprising 8 (26%) species that predominantly feed on fish, 4 (13%) species that predominantly feed on insects and 19 (61%) feed on both fish and insects (Appendix 1). Generally, species composition among the three feeding guilds was averagely similar between the two sites (Morisita index: 63–71 for all pairwise combinations; Table 2). The two sites had an overlap of 25 species, accounting for 62.5% of the total species on record (Appendix 2). Nine species, namely, Leptoptilos crumenifer (Marabou Stork), Ardea melanocephala (Black Headed Heron), Platalea alba (African Spoonbill), Circus ranivorus (African Marsh Harrier), Burhinus vermiculatus (Water Thick-knee), Ardea cinerea (Grey Heron), Chlidonias hybrida (Whiskered Tern), Vanellus senegallus (African Wattled Plover), and Pelecanus rufescens (Pink Backed Pelican) were exclusively recorded at the FPA, while another five species, namely, Corythornis cristatus (Malachite Kingfisher), Vanellus crassirostris (Long-toed Lapwing), Nycticorax nycticorax (Black Crowned Night Heron), Vanellus coronatus (Crowned Lapwing) and Sarkidiornis melanotos (Knob Billed Duck) were exclusively recorded at the NFPA.

Species diversity, species evenness and species dominance did not differ between FPA and NFPA. However, species richness was higher at the FPA tributary than at the NFPA tributary (Table 2).

Generally, when considering multi-trait functional diversity indices, carnivorous waterbirds’ ecological distance differed between FPA and non-FPA sites (Table 3). Carnivorous waterbirds’ ecological distance was higher in FPA sites compared to NFPA sites. Similarity, functional evenness, functional richness and functional divergence also differed between FPA and NFPA sites (Table 3). Functional richness and divergence were higher in FPA sites compared to NFPA sites. Contrary to these results, functional evenness was higher at the NFPA sites compared to FPA sites.

When considering functional regulatory single traits indices, all the nine traits used in this study significantly differed between the FPA and NFPA sites, except for breeding type and clutch size. In terms of feeding guild, FPA sites were dominated by carnivorous waterbirds, while NFPA sites were dominated by species that predominantly feed on both fish and insects. In terms of food items, FPA sites were dominated by piscivore birds and NFPA sites were dominated by water birds that feed both on fish and insects.

When considering community weighed single traits functional diversity, add four traits, namely, food item, weight and movement type were significantly different between FPA and NFPA sites (Table 4). Similar to functional regulatory single trait diversity, FPA sites were dominated by carnivorous waterbirds, while NFPA sites were dominated by waterbirds that feed dominantly on insects. In terms of food items, FPA sites were dominated by piscivore birds and NFPA sites were dominated by water birds that feed both on fish and insects (Table 4).

Piscivorous waterbirds at the study site were dominated by 9 (23%) species out of the possible 40 total number of sampled species. These dominant species were Little Egret (15%), Reed cormorant (15%), White Breasted Cormorant (14%), Great Egret (11%), Grey Headed Gull (10%), Squacco Heron (7%), African Open bill stock (5%), Pied Kingfisher (5%), and Blacksmith Plover (3%) (Figure 2a). Of the nine dominant species, six (67%) feed predominantly on fish, two (22%) species feed on both fish and insects, while only one (11%) feeds on insects. Further analysis showed that, at the FPA, carnivorous waterbirds were dominated by 8 (23%) species out of the possible 40 total number of sampled species. The dominant species on FPA sites were White-breasted Cormorant (17%), Reed Cormorant (16%), Great Egret (14%), Little Egret (10%), African Sacred Ibis (6%), White faced Duck (6%), Squacco Heron (5%), and Grey-headed Gull (5%). Of the eight species, three (37.5%) feed predominantly on fish and five (62.5%) predominantly on fish and insects. In contrast, at the NFPA sites, carnivorous water birds were dominated by 6 (15%) species out of the possible 40 total number of sampled species. Of the six species, four (67%) feed predominantly on fish, one species (17.5%) feeds on fish and insects and one species (17.5%) predominantly feeds on insects. Three (37.5%) out of the above eight species (i.e. African Open bill Stork, Great Egret and White Breasted Cormorant) decreased in abundance at the NFPA sites. Of these three decreasing species, only the African Open bill Stork, was substituted by functionally similar and increase in abundance but minor species (Black Smith Plover) (Fisher’s exact test, p > 0.05).

On NFPA sites, the functional similarity (in terms of predominant food items) of dominant species was similar to the functional similarity of dominant species of the whole study area (Fisher’s exact test = 1.277, p = 0.557). On FPA sites, functional similarity of dominant species was dissimilar to the functional similarity of the dominant species of the whole study area (Fisher’s exact test = 6.685, p = 0.034).

This study aimed to explore the abundance and distribution of carnivorous waterbirds between FPAs and NFPAs on the Zambezi River tributaries. A total of 40 different species, representing 35 species at the FPA and 31 species at the NFPA were recorded from July 2021 to October 2021. Among these species, only a few dominant species were documented at both sampling sites, supporting our prediction that at both sites, carnivorous waterbirds community abundance was dominated by few species and have many minor species. Throughout the study period, only a few dominant species were spotted at the two sampling sites, but dominancy was not equally distributed among the feeding guilds. Waterbirds species that feed on both fish and insects dominated the community assemblages of waterbirds at both the FPA (60%) and NFPA (61%) sites. This suggests that piscivorous waterbirds that had a wider niche (feeding on fish and insects) were resilient to fishing pressure, and piscivorous waterbirds that mainly feed on fish were affected by fishing pressure. It predicts that piscivorous waterbirds with wider niches can shift among available diet to sustain their food requirement. Similar distributions have been reported in other wetlands (De Arruda Almeida et al. 2018; Lorenzón, Ronchi-Virgolini & Blake 2020), and terrestrial ecosystems (Rutina & Moe 2014; Walker et al. 1999). Henry and Cumming (2017) reported that wetland birds were dominated by few species whenever water levels of wetlands were high or low, while De Arruda Almeida et al. (2017) reported that waterbirds were dominated by few species both at artificial and natural wetlands. Walker et al. (1999) and Rutina and Moe (2014) reported that graminoid and tree species were dominated by fewer species across grazing and browsing gradients, respectively. However, opposing results have been reported where no signs of redundancy in species dominance were observed (Aarif et al. 2017; Petchey et al. 2007). In line with this study, Petchey et al. (2007) also reported that there was no redundancy in British avian assemblages.

Species composition between the two sampling sites was averagely similar (Morisita’s index: 0.63 to 0.71). The two sampling sites had an overlap of 25 species, accounting for 62.5% of the total species on record. These observations suggest that the external effect of fishing did not influence the species richness and evenness distribution of carnivorous waterbirds between the two sampling sites. However, the influence of fishing pressure on other traits that are associated with growth forms and resources use strategies cannot be ruled out. Connectivity between the two sampling sites could equally explain similarity in catch composition as there is flow and exchange of similar species between two geographically connected sites in the absence of external factors such as fish pressure.

The distribution of minor species at the FPA and NFPA had functionally similar dominant species; except for waterbirds species that fed on insects (e.g. Black Smith Plover). It has been hypothesised that minor species have to be functionally similar to dominant species in order to stabilise the functions and processes of an ecosystem (Rutina & Moe 2014; Walker 1999; Walker et al. 1999). However, a lack of the distributions of insectivorous species at both sites would suggest that fish is the most predominant food item governing the distribution of carnivorous waterbirds at the two sampling sites and that competing activities such as fishing have not affected prey abundance for carnivorous waterbirds (Anthal & Sahi 2017; Wenny et al. 2011).

Further analysis on species diversity and richness showed that species richness was higher at the FPA compared to the NFPA sites. Similarly, the functional diversity indices were equally higher at the FPA than at the NFPA sites, while the functional evenness was higher at the NFPA sites than FPA sites. Low disturbances have been associated with high diversities (Schellenberger Costa et al. 2017) and species dominancy (Walker et al. 2013). If the concept of functional traits-ecosystem functions relationship holds, the FPA sites should act as an important refuge for a flock of species when water availability is low and enhance resource availability and accessibility to waterbirds (Ma et al. 2010). The current results showed that large body piscivorous waterbirds (e.g. Pink Backed Pelican) were dominantly sighted at the FPA sites than at the NFPA sites. This could be linked to high abundances of prey food resources such as fish and insects at the FPAs (Simasiku & Hay 2023).

Our third prediction that a decline in abundance of dominant waterbird species at the FPAs is counterbalanced by an increase in abundance of taxonomically and functionally similar dominant species at the NFPA sites was partly supported by this study as only some species that declined in abundance under changes in fishing pressure were replaced by increasing functionally similar species at the FPAs. The distributions of minor species at the FPA and NFPA sites had functionally similar dominant species; except for waterbird species that fed on insects (e.g. Black Smith Plover). It has been hypothesised that minor species have to be functionally similar to dominant species in order to stabilise the functions and process of an ecosystem (Rutina & Moe 2014; Walker 1999; Walker et al. 1999). However, a lack of the distribution of insectivorous species at both sites would suggest that fish is the most predominant food item governing the distribution of carnivorous waterbirds at the two sampling sites and that competing activities such as fishing have not affected prey abundance for carnivorous waterbirds (Anthal & Sahi 2017; Wenny et al. 2011). Species that predominantly feed on fish were generally reduced in their contributions to the overall abundance of species at the NFPA sites compared to FPA sites, suggesting that fishing pressure might have negatively affected the distribution and abundance of these species at the NFPA sites. On the contrary, insectivorous waterbirds seemed to be favoured by fishing activities.