The conservation of South African white sharks : population number, genetic distinctiveness and global connections

Thesis (D.Phil)--Stellenbosch University, 2015.

ENGLISH SUMMARY: The white shark (Carcharodon carcharias) is IUCN red listed as vulnerable, but the lack of basic biological information is arguably the biggest obstacle facing the conservation of the species. The aim of this project was to aid in the conservation of white sharks by producing various data sets that can be used in adaptive management. In doing so we estimated the South African white shark population number, their genetic connectivity along the coastline and investigate various behavioural aspects. We developed a categorization system to manage large photographic databases for individual identifications of white sharks by making use of dorsal fin images. The novel categorization system was developed by making use of 4398 photos taken over a 27 month period. A notches code method was produced and this proved to significantly reduce the search time associated to accurately identify individuals. From the photos we identified 426 individuals in the Gansbaai region of South Africa. By using a mark-recapture technique and the open population model POPAN, we estimated a range between 353 - 522 individuals (95% confidence). These data were confirmed by analyses of 14 polymorphic microsatellite markers for C. carcharias that revealed a contemporary effective population size (CNe) of 338 individuals (95% confidence, Pcrit = 0.01). Both estimates are in the same range but considerably less than a previously published estimate (e.g. N = 808 to 1008) that relyied on fin matching software (DARWIN) to automatically match the sharks’ dorsal fin. Through software validation, we provided evidence that DARWIN failed to produce accurate estimates and the discrepancy in population numbers are most likely due to the inclusion of false negatives in the published literature. To determine whether the Gansbaai population forms a unique evolutionary unit, the phylogeography of white sharks along the South African coastline was investigated by making use of mtDNA and microsatellite markers. A total of 238 unique individuals were sampled originating from five aggregation sites. Four mtDNA haplotypes were found for the entire range. One common mtDNA haplotype was shared by 89% of the individuals sampled, and a second haplotype (13 bp different) was present in 10% of the remaining sharks. No phylogeographic structure was found among aggregation sites. This finding was supported by microsatellite analyses and both data sets show a remarkably low level of genetic diversity (h = 0.02, π = 0.0027; Na = 7.6, Ho = 0.675). The genetic results suggest that the South African population is the result of a founder event or a severe bottleneck in the recent past. These data were combined with published mtDNA data at the global level and results suggest that at the continental scale three distinct mtDNA clades occur. These are confined to the Mediterranean and Indo-Pacific Oceans (Clade 1), the Atlantic and Indian Oceans (Clade 2) and a single haplotype restricted to the waters of South Africa (Clade 3). These clades are probably the result of allopatric speciation associated with the closure of the Isthmus of Panama as confirmed by dating analyses. By combining the mark-recapture analyses and genetic techniques, permutations test revealed that during scavenging situations, sharks associate with conspecifics in a non-random structure (Mean of pairwise associations = 728, mean of permutation test = 597, P = 0.00). Analyses of the nature of such associations were not statistically significant, but provided some insights (e.g. partial sexual segregation and different class size groupings) indicative of a complex social system which may rather mirror that of marine mammals. Based on the results of this study South African white sharks require more protection than previously thought and a long term management plan is needed to secure the future survival of the species, this will need to take into account the low genetic diversity and to include constant assessments of the population numbers.

AFRIKAANS OPSOMMING: Die wit haai (Carcharodon carcharias) is deur die IUCN Rooi Lys gelys as kwesbaar, maar die gebrek aan basiese biologiese inligting is waarskynlik die grootste struikelblok wat die bewaring van die spesies in die gesig staar. Die doel van die projek was om te help met die bewaring van die wit haai deur verskei stelle data te genereer wat vir die aanpasbare bestuur van die spesie gebruik kan word. Deur dit te doen kon ons skat wat die Suid-Afrikaanse bevolkingsgrote van die wit haai is, hulle genetiese verbinding langs die kus is, asook verskeie gedrags aspekte ondersoek. Deur gebruik te maak van fin-beelde het ons ‘n groeperings stelsel ontwikkel wat help om groot fotografiese databasise te orden vir individuele identifikasies van die wit haai. Die unieke kategoriseringstelsel is ontwikkel deur gebruik te maak van 4398 fotos wat geneem is oor ‘n tydperk van 27 maande. ‘n Kerwe kode metode was ontwikkel wat drasties gehelp het om tyd te bespaar tydens die akurate indentifikasie van individuele diere. 426 individuele haaie was geidentifiseer in die Gansbaai streek van Suid Afrika. Deur gebruik te maak van ‘n merk-en hervang tegniek en die oop populasie model POPAN, het ons beraam dat daar tussen 353-522 individue tans is (95% akuraat). Hierdie data is bevestig deur die ontleding van 14 polimorfiese mikrosatelliet merkers vir C. carcharias wat 'n kontemporêre effektiewe bevolkingsgrootte (CNe) van 338 individue voorstel (95% vertroue, Pcrit = 0,01). Beide berekininge is in dieselfde omtrek maar aansienlik minder as wat voorheen beraam is (b.v. N = 808 tot 1008). Die vorige beraming is gedoen deur te vertrou op die sagteware program (DARWIN) wat die haaie se dorsale-finpatroon outomaties soek en identifiseer. Deur die sagteware te toets het ons bewys dat DARWIN nie kon voldoen aan die vereiste om akkurate skattings te produseer nie en die verskil in bevolkingsgetalle is waarskynlik as gevolg van die insluiting van vals positiewes in die gepubliseerde literatuur. Om te bepaal of die Gansbaai bevolking 'n unieke evolusionêre eenheid vorm, was die filogeografie van die wit haaie langs die Suid-Afrikaanse kuslyn ondersoek deur gebruik te maak van mtDNA en mikrosatelliet merkers. 'n Totaal van 238 unieke individue is versamel afkomstig uit vyf verskillende streke. Slegs vier mtDNA haplotipes is gevind vir die hele Suid Afrikaanse verspreiding. Een algemene mtDNA haplotipe is gedeel deur 89% van die individue, en 'n tweede haplotipe (wat 13bp verskillend was) was teenwoordig in net 10% van die oorblywende haaie. Geen filogeografiese struktuur is gevind langs die kuslyn nie. Hierdie bevinding word ondersteun deur die mikrosatelliet ontledings en beide stelle data toon 'n merkwaardig lae vlak van genetiese diversiteit (h = 0,02, π = 0,0027; Na = 7.6, Ho = 0,675). Die genetiese resultate dui daarop dat die Suid-Afrikaanse bevolking moontlik die gevolg is van 'n stigter-gebeurtenis of 'n ernstige bottelnek in die onlangse verlede. Hierdie data is gekombineer met gepubliseerde mtDNA data op die globale vlak en resultate dui daarop dat op kontinentale skaal, drie afsonderlike mtDNA klades voorkom. Dit is beperk tot die Middellandse See en Indo-Stille Oseaan (klade 1), die Atlantiese en Indiese Oseane (klade 2) en 'n enkele haplotipe beperk tot die waters van Suid-Afrika (klade 3). Hierdie klades is waarskynlik die gevolg van allopatriese spesievorming wat verband hou met die sluiting van die Isthmus van Panama soos bevestig deur daterings wat gedoen is. Deur die merk-en hervangmetode ontleding en genetiese tegnieke te kombineer, het permutasie toete getoon dat gedurende aas-voeding situasies, assosieer haaie met mekaar in 'n nie-ewekansige manier (gemiddelde van paarsgewyse verbindings = 728, gemiddeld van permutasie toets = 597, P = 0,00). Ontleding van die aard van sulke verwantskappe was nie statisties beduidend nie, maar verskaf 'n paar insigte (bv. gedeeltelike seksuele segregasie en verskillenede klas grootte groepe assosieer met mekaar). Hierdie toon op 'n aanduiding van 'n komplekse sosiale stelsel wat weerspieël word in ander mariene soogdiere. Gebaseer op die resultate van hierdie studie, dui dit duidelik daarop dat die Suid-Afrikaanse wit haaie vereis meer beskerming as wat voorheen gedink is en 'n langtermyn bestuursplan is nodig om die voortbestaan van die spesie te verseker. Lae genetiese diversiteit moet in ag geneem word en die gereelde assessering van bevolkings-getalle moet in plek gestel word.

Doctoral