Spatial variation in life-history traits of the marbled parrotfish, Leptoscarus vaigiensis, in coastal Kenya

Abstract

Phenotypic plasticity in life-history attributes of fishes is poorly studied in Western Indian Ocean (WIO) region but may be important in maintaining resilience of populations to disturbance events. This study examined the existence of phenotypic divergence in biological attributes of the commercially important marbled parrotfish (Leptoscarus vaigiensis) among reefs in coastal Kenya in an attempt to bridge the data gaps and understand ecosystem function. Phenotypic divergence in the reproductive, feeding, growth and mortality attributes of L. vaigiensis was studied during May 2011-April 2013 at six reef sites exposed to varying levels of fishing pressure in coastal Kenya. Baited fish traps were used to sample specimens of L. vaigiensis, an economically important species, within two no-take marine parks (Malindi and Watamu) and their adjacent Reserves in which regulated fishing is allowed. Monthly samples were also obtained from fishers operating at two unprotected sites (Kanamai and Vipingo). Biological attributes of the species were then compared between sites using a combination of univariate and multivariate statistical analysis. Results showed reproductive attributes of fish in the Parks and Reserves had higher fecundities than those on unprotected reefs indicating spatial differences in the reproductive potential of the species. However, fish at unprotected sites had higher fecundities at smaller sizes relative to larger fish at protected sites. Females at unprotected sites matured at lower lengths (13.4 cm, range, 11.8-14.8 cm, 95% CI) than those in marine parks (17.8 cm, range 17.5-18.1 cm, 95% CI) and reserves (17.4 cm, range, 16.2-18.4 cm, 95% CI) suggesting spatial difference in the trade-off between gonadal development and somatic growth likely influenced by differential fishing pressure. Feeding data showed higher niche breadths for fish from protected sites during the northeast monsoon (NEM) season and at fished sites during the southeast monsoon (SEM) season suggesting the influence of sites and seasonality on feeding activity. Results on growth showed higher asymptotic length (L∞) estimate (~ 31 cm) for the species from a reserve site and lowest estimates (26-27 cm) from a marine park site. As expected, total mortality estimates (Z yr-1) were lower for protected areas (1.69-1.96) and higher for reserve (2.87-9.15) and non-protected (4.39-5.30) sites. It is concluded that the potential for growth is influenced by fishing and other factors not covered in this study including temperature and food availability. Overall, the results indicated variable levels of phenotypic divergence in attributes of the species between sites. The real causes of this divergence cannot be partitioned between fishing mortality, genetic variability or habitat-induced variation and will require further investigations. The main thesis of the research is that populations of this commercially important species are likely sustained by resilience to exploitation caused by plasticity in life-history traits. The existence of spatial variability in some parameters (e.g. fecundity and size at maturity) and functions (e.g. niche breadth and growth) requires that scientific management of stocks be based on spatially explicit models that use these parameters.