Gillnetting in nature reserves: a case study from the Kosi Lakes, South Africa
Introduction
The Kosi Lakes, situated on the east coast of South Africa just south of the Mozambique border, are a series of water bodies about 10 km in length (Fig. 1). The lakes run parallel to the Indian Ocean immediately behind high, vegetated, coastal dunes and the mouth has closed only once in living memory, in 1965 for 106 days. The lakes and much of the surrounding wetland were proclaimed a Nature Reserve in 1987 and listed under the RAMSAR Convention as a wetland of international importance in 1991. Various aspects of the lakes and their fisheries have been described (Tinley, 1964, Begg, 1978, Begg, 1980, Blaber, 1978, Blaber and Cyrus, 1981, Kyle, 1986).
For many generations indigenous people have fished in the lakes (Tinley, 1964, Felgate, 1965) using traps and traditional spears, while in the last 30 years recreational angling has become increasingly popular. Traditional fishing is controlled by the local community, whereas recreational angling is controlled by Provincial regulations (Natal Nature Conservation Ordinance, No 15 of 1974). Other fishing methods, including gillnetting, were termed illegal. Harvested fish species are primarily of estuarine or marine origin (Blaber and Cyrus, 1981, Kyle, 1986).
The Nature Reserve is administered by the KwaZulu Department of Nature Conservation (KDNC) whose policy is to promote the wise use of renewable natural resources. The KDNC saw traditional fishing by indigenous people as an inalienable right of the community while recreational angling, as a tourism attraction, was considered desirable. In the 1980s it was suggested (Kyle, 1986) that extant legal fishing methods did not compete greatly with each other and overall exploitation levels were low.
A tagging study (Kyle, 1986) estimated that traps caught c.5% of the adult fish populations and recreational angling <1%. In 1992, the KDNC proposed an experimental gillnetting project to determine if there was scope for further sustainable utilisation of the fish resources of the lakes. Gillnetting had been carried out illegally in the lakes since the early 1950s (Tinley, 1964) but was suppressed by the authorities as it was believed to interfere with fish migrations, catch too many fish and be difficult to control, being principally a night time activity. Kyle (1992)suggested that if effort and area of operation were controlled, gillnetting could be used to sustainably exploit certain targeted fish species to the benefit of reserve neighbours. Certain fish species, such as spotted grunter Pomadasys commersonni, river snapper Lutjanus argentimaculatus, river bream Acanthopagrus berda, Natal stumpnose Rhabdosargus sarba, sea pike Sphyraena spp. and the kingfishes Carangidae were caught in numbers by both recreational and traditional fishermen but freshwater fish in Lake Nhlange such as the sharptooth barble Clarias gariepinus, and the Mozambique tilapia Oreochromis mossambicus, (Fig. 1) were rarely caught in traps or by rod and line. Largescale pursemouth Gerres rappi, a species abundant in the lakes, was not caught by recreational anglers and only in low numbers in traps. Mullets Mugilidae, were abundant and, although caught in numbers by the traps, were thought to be able to sustain greater fishing pressure. For these reasons freshwater fish, G. rappi and Mugilidae were designated target species for the gillnetting. The ethic behind the project was to sustainably benefit as many local residents as possible and involve them in the management of the fish resource of the lakes. The KDNC decided to implement an experimental gillnet fishery in Lake Nhlange and a phased approach was devised.
It was unlikely that fish migrations would be severely impacted if nets were kept away from channels. Experimental gillnet fishing results indicated that netting in the shallows of Lake Nhlange caught a high proportion of target species (G. rappi and Mugilidae). Netting was considered inappropriate north of Lake Nhlange (Fig. 1) as there were many traps in the lower part of the estuary, which provided a legal, sustainable and controlled way of fishing (Kyle, 1986). Conflict between illegal netters and trap owners had arisen several times in the lower estuary, when the netters were believed to have stolen fish or damaged traps. Netting south of Lake Nhlange was not allowed because of a higher density of crocodiles Crocodylus niloticus and hippo Hippopotamus amphibius. Gillnetting intensity and area of operation were decided upon by the KDNC while implementation was carried out through the local community. This paper describes the evolution of the project and summarises results, progress and problems from inception in 1992 until December 1995.
Section snippets
Methods
Thirty metres of netting were allocated per permit following agreement between the potential netters and the KDNC. No restrictions were imposed on mesh sizes or on the material (monofilament or multifilament) from which the net was made. Mesh size restrictions were considered but a large amount of data from nets confiscated by the KDNC showed that stretched meshes used ranged 60–110 mm, with a mode of about 80 mm, and these were considered appropriate for the fishery.
Kyle (in press)showed that
Effort
There was a progressive increase in the number of permits issued throughout the study (Table 1). Initially the recorded net settings rose in a similar manner and then declined after mid 1994 (Fig. 2). The total monthly catch divided by the number of settings (CPUE) showed no clear overall annual increase or decrease (Fig. 3) suggesting that the “catchability” of fish remained similar throughout the study. On an annual basis there was a tendency for lower CPUE in winter. During the four years of
Discussion
Although traditional fishing has been allowed in marine and other reserves elsewhere (Ramos-Espla and Bayle-Sempere, 1989) gillnetting has not been permitted in such reserves in Africa before 1992. Subsequently it has been tried in the St. Lucia lakes but as yet with inconclusive results (Mann, 1996). Since 1992 the method has been used, on a controlled basis, in the Kosi Bay Nature Reserve to selectively exploit certain fish populations (Kyle, 1994). Gillnetting has been the subject of
Acknowledgements
I thank the KwaZulu Department of Nature Conservation for permission to publish this paper, Diane Kyle, Rudy van der Elst, Bruce Mann and Peter Rutsch for support and assistance with the project and Lynnath Beckley and Pete Fielding for assistance in preparing this paper.
References (42)
- Anon, 1990. Integrated development of small scale fisheries in West Africa. Model project. Benin. Project findings and...
- Begg, G., 1978. The estuaries of Natal. Natal Town and Regional Planning Report 41....
- Begg, G., 1980. The Kosi system: aspects of its biology, management and research. In: Bruton, M.N., Cooper, K.H....
- Blaber, S.J.M., 1973. Temperature and salinity tolerance of juvenile Rhabdosargus holubi (Teleostei: Sparidae). J. Fish...
- Blaber, S.J.M., 1978. Fishes of the Kosi system. Lammergeyer 24,...
- Blaber, S.J.M., Cyrus, D.P., 1981. A revised checklist and further notes on the fishes of the Kosi system. Lammergeyer...
- Blaber, S.J.M., Whitfield, A.K., 1976. Large scale mortality of fish at St. Lucia. S. Afr. J. Sci. 72,...
- Bennet, B.A., Hamman, K.C.D., Branch, G.M., Thorne, S.G., 1985. Changes in the fish fauna of the Bot river estuary in...
- Clover, C., 1992. Plunder of our oceans. Reader's Digest, December,...
- Compagno, L.J.V., Smale, M.J., Davis, B., Cliff, G., Dudley, S.F.J., 1989. Sharks and the Natal Sharks Board. In:...
Cited by (21)
A systematic review and risk matrix of plastic litter impacts on aquatic wildlife: A case study of the Mekong and Ganges River Basins
2022, Science of the Total EnvironmentCitation Excerpt :Entanglement, especially in gill nets, is a major and well-known cause of death of cetaceans, and we have scored the impact of entanglements as “Lethal”. Our literature search retrieved one mention of ingestion of derelict fishing gear (Warner et al., 2016) and numerous studies concerning entanglement of gharials and other crocodilians (Australia. Dept. of the Environment Water Heritage the Arts and Threat abatement plan for the impacts of marine debris on vertebrate marine life, 2009; Hussain, 2009; Kyle, 1999; Platt and Van Tri, 2000; Shaney et al., 2019; Singh and Rao, 2017; Thorbjarnarson et al., 2000). Furthermore, a review into plastic ingestion in reptiles found no reported evidence for plastic ingestion in crocodilians (Staffieri et al., 2019), nor any studies showing evidence of harm.
Accumulation of organochlorine pesticides in fat tissue of wild Nile crocodiles (Crocodylus niloticus) from iSimangaliso Wetland Park, South Africa
2018, ChemosphereCitation Excerpt :Lake St Lucia hosts the largest population with a minimum of 684 adult and sub-adult crocodiles, while Lake Sibaya (26), Kosi Bay (<10) and Nsumo pan (<10) are estimated to contain substantially lower numbers (EKZNW unpublished 2015 aerial survey data). Trend analysis based on aerial count data for the last decade indicates a decline in all four populations (EKZNW unpublished data 2015 aerial survey data) and is attributed largely to direct and incidental anthropogenic pressures, including illegal killings, fish-trap and gillnet mortalities, destructions of nesting sites and eggs, alien plant infestation at nesting sites, boat-collision mortalities, and severe droughts (Kyle, 1999; Leslie and Spotila, 2001; Combrink et al., 2011, 2013; Warner et al., 2016a). Crocodile captures took place during 2016 and 2017 under permit from the iSimangaliso Wetland Park Authority and Ezemvelo KZN Wildlife.
Age and growth of Rhabdosargus sarba (Pisces: Sparidae), from KwaZulu-Natal, South Africa
2002, Fisheries ResearchThe changing status of important marine fishery species in selected South African estuaries
2023, African Journal of Marine ScienceEstuarine Fisheries
2022, Fish and Fisheries in Estuaries: A Global PerspectiveStatus of the Nile Crocodile Population in Pongolapoort Dam after River Impoundment
2017, African Zoology