Larval fish assemblages of selected estuarine and coastal systems in Kwazulu-Natal, South Africa

Abstract

This study focuses describing assemblages of fish larvae occurring in potential nursery habitats and elucidating the recruitment processes of fishes in the KwaZulu-Natal region. The composition, the degree of estuarine-association of all taxa, temporal and spatial abundance patterns, developmental stages and were examined in three large estuarine systems, a surf zone habitat and in the nearshore marine environment. In addition, two 24-h inlet studies were undertaken to ascertain diel patterns. Relationships between larval densities with environmental variables were also examined to gain insight into the possible causal mechanisms for the observed abundance patterns. Larval fish samples from Durban Harbour and Richards Bay Harbour were collected at top, mid and bottom depths in the dredged channel 100 to 200 m from the harbour entrance, over 13 and 12 month study periods, respectively. Larval fish from the St Lucia Estuary were collected over 12 months at a foxed station 4 km from the mouth. A total of 8 797 fish larvae, representing 64 families and 144 taxa, was collected in Durban Harbour. From Richards Bay Harbour, 105 taxa representing 53 families were collected from a total of 7 163 larvae. Larvae of the thorny anchovy, Stolephorus holodon, were very abundant in both harbours (32% and 10% of the total catch, respectively), with the blueline herring Herklotsichthys quadrimaculatus (30%) and an unidentified goby Gobiid 12 (9%) being dominant in Durban Harbour and in Richards Bay Harbour, respectively. Conversely, in the St Lucia Estuary, 51 590 larvae in total were collected represented by 85 taxa from 44 families. The river goby Glossogobius callidus (67%) and the estuarine roundherring Gilchristella aestuaria (19%) were the most dominant species. Larval fish in the surf zone were collected monthly, using a pushnet, at six stations up to 3 km north of the St Lucia Estuary mouth. To determine any diel patterns a 24-h study was also undertaken in the surf zone when the estuary mouth was closed. Far more larvae were collected during the 24-h study compared to the 12-month study: 13 731 larvae and 2 931 larvae, respectively. The larval assemblage was characterised by taxa in the families Sparidae, Haemulidae, Ambassidae, Tripterygiidae and Chanidae. The most abundant species were Pomadasys olivaceum, Ambassis sp., Rhabdosargus holnbi, Croilia mossambica and Chanos chanos. Fish larvae were also sampled in the adjacent nearshore coastal zone, from five stations in a transect up to 2.5 km offshore, and from two stations north and south of the St Lucia Estuary mouth. The assemblage consisted of 246 taxa representing 98 families from a total of 6 069 larvae. Larvae in the families Myctophidae and Tripterygiidae comprised 21% and 16% of the total catch, with the dominant species being an unidentified triplefin, Tripterygiid 1, and the lanternfish Benthosema fibulatum. The percentage contribution to total number of larvae of estuarine-associated fish species increased from "estuarine-habitat" (St Lucia Estuary) to "semi-estuarine habitat" (Richards Bay Harbour and the surf zone) to "marine-habitat" (Durban Harbour and the nearshore coastal zone). Species representative of each estuarine-association group were, for example: G.aestuaria, Thryssa vitrirostris and Rhabdosargus spp.-estuarine-dependent; S.holodon, C.mossambica smdAmbassis spp - partially estuarine-dependent; unidentified gobies (Gobiid 12 and Gobiid 27), unidentified tripleflns (Tripterygiid 1), Lampanyctus alatnst H.quadrimaculatus and Umbrina ronchus -estuarine-independent. In all study areas, different recruitment periods were evident for each estuarine-association group with peaks in abundance occurring in all four seasons. This was dependent upon the seasonality of the dominant species present in a particular study area. In the estuarine environments, young larvae (preflexion and flexion developmental stages) of estuarine-dependent species were moderately abundant whereas, old larvae (postflexion) and early juveniles of partially estuarine-dependent species predominated. The surf zone habitat was dominated by postflexion larvae (68% of total catch) of estuarine-associated species, with early juveniles of estuarine-independent species being prevalent in the 24-h study when the estuary mouth was closed (50%). In the nearshore coastal environment larvae of marine species not dependent of estuaries were mainly at the preflexion and flexion developmental stages (67% of the total catch). Stepwise regression statistics and multivariate analysis (classification and ordination) clearly indicated the importance of environmental factors in determining the structured of the larval fish assemblages. Furthermore, the correlations of larval densities with environmental variables was found to be species-specific. The most significant variable accounting for the observed distributional patterns of the fish larvae was turbidity (pw = 0.55 - weighted Spearmans rank correlation). However, the intercorrelations between all the environmental variables measured were also important in determining abundance patterns (e.g. pw = 0.48 for salinity + turbidity; pw = 0.41 for salinity + temperature + turbidity). Analyses of tidal exchange of fish larvae on flood and ebb tides, in Durban Harbour, Richards Bay Harbour and in the St Lucia Estuary mouth, indicated that selective tidal stream transport is a recruitment mechanism employed by certain species (e.g. Argyrosomus sp., Pomadasys commersonnii and Solea bleekeri) to enable retention in the estuarine habitat i.e. larval densities of these species were significantly higher on flood tides and in bottom waters (P < 0.05). In conclusion, results of the present study have indicated that diverse and dynamic larval fish assemblages are present in the estuarine and coastal systems on the KwaZulu-Natal coast. Larvae (mainly postflexion) of estuarine-associated fish species were abundant in all three estuarine systems and the surf zone suggesting that all these habitats function as nursery sites for particular species. Additional species-specific studies on larval stages of recreational and commercially important linefish species are needed in order to make management decisions regarding the conservation and exploitation of these species.