Local and landscape drivers of avian diversity facets in the naturally fragmented Southern Mistbelt forests of KwaZulu-Natal and Eastern Cape, South Africa = Abashayeli bendawo bezinhlobo zezinyoni ezahlukene emahlathini ase-Southern Mistbelt ahlukaniswe ngokwemvelo ezifundazweni zakwa KwaZulu-Natal nase Eastern Cape eNingizimu Afrika.
Forest fragmentation is a process whereby a forest landscape is subdivided into smaller and more isolated fragments embedded within a matrix of anthropogenic land-uses. The effects of increasing anthropogenic practices surrounding forest remnants threaten species persistence as habitat fragments become further isolated by the matrix, which impedes species movement and dispersal, causing local extinctions when conditions become more unfavourable. However, naturally fragmented forest systems harbour resilient species, but the novel challenges presently experienced by these species have unknown consequences. Here, I determined to assess local avian diversity facets in selected naturally fragmented Southern Mistbelt forests of KwaZulu-Natal and Eastern Cape provinces, South Africa. I conducted fixed-radius point-count surveys across 32 (range: 0.03 – 732.42 ha) of these Southern Mistbelt Forests. Data collection was conducted during the breeding (October-February) and nonbreeding (May-August) seasons in 2018 and 2019. Firstly, I used a multifaceted approach to assess the effects of landscape composition (i.e., matrix quality), habitat fragmentation (i.e., isolation distance and fragment-size) and local habitat heterogeneity (i.e., forest-structural complexity) on avian alpha-diversity (taxonomic, functional, phylogenetic and functional-phylogenetic diversity) of the whole community, forest-dependent (i.e. forest specialist) and non-forest-dependent (i.e. forest generalist) assemblages. Secondly, I revealed how key landscape features (i.e. habitat amount, matrix heterogeneity and average isolation distance) affected local bird diversity and determined the importance of each landscape variable using linear mixed-effect models (LMEs). Thirdly, I mapped connectivity of core Southern Mistbelt Forest patches provisioning the highest (≥ 50 %) avian phylogenetic diversity by using a hybrid of least-cost pathway and ecological circuit theory approach to identify landscape features that promote or impede habitat connectivity of a surrogate forest-specialised and dispersal-limited species (Lemon dove - Aplopelia larvata) to guide landscape connectivity for avian forest communities in the fragmented forest mosaic landscape. My findings revealed that: A) 1. matrix quality was a significant positive predictor of functional (FRic, sesFRic) and phylogenetic (sesPD) diversity; 2. habitat fragmentation had significantly negative effects (i.e. increasing isolation distance and decreasing fragment size) on multiple diversity facets; and 3. diversity facets of forest-dependent species unexpectedly declined with increasing local vegetation complexity. B) Habitat amount in the landscape was the main significant positive predictor of local forest bird diversity, and there were no significant influences of the average isolation distance and landscape (matrix) heterogeneity on local avian diversity. C) Landscape connectivity of Southern Mistbelt Forest fragments is predominantly promoted by indigenous forest cover and reduced in regions of unsuitable habitat (i.e. exotic timber plantations, grassland, agriculture, and residential land-cover). I concluded that i) positive effects of surrounding matrix quality in the landscape mediates the negative effects of habitat fragmentation on local forest bird ecological groups; ii) promoting local vegetation complexity could contribute to the loss of forest-dependent species (i.e. forest specialists); iii) habitat amount in the surrounding landscape was the predominantly important predictor of local avian diversity and promoted habitat connectivity among core forest patches of fragmented Southern Mistbelt Forest. Therefore, I recommend preserving and increasing forest cover in the landscape to ensure the long-term survival of forest species in this naturally fragmented ecosystem. Iqoqa Ukuhlukaniswa kwehlathi kuyinqubo lapho indawo yehlathi ihlukaniswa khona ibe yizingcezu ezincane nezizimele kakhulu ezishumekwe ngaphakathi kwe-matrix yokusetshenziswa komhlaba kwe-anthropogenic. Imithelela yokwanda kwemikhuba ye-anthropogenic ezungeze izinsalela zehlathi isongela ukuphikelela kwezinhlobo zezilwane njengoba izingcezu zendawo yokuhlala zihlukaniswa ngokwengeziwe yi-matrix, evimbela ukunyakaza kwezinhlobo zezilwane nokuhlakazeka, okubangela ukushabalala kwendawo lapho izimo ziba zimbi nakakhulu. Nokho, izimiso zamahlathi ahlukene ngokwemvelo zinezinhlobo ezikwazi ukumelana nezimo, kodwa izinselelo ezintsha ezitholwa yilezi zinhlobo manje zinemiphumela engaziwa. Lapha, nginqume ukuhlola ukuhlukahluka kwezinyoni zasendaweni emahlathini akhethiwe ase-Southern Mistbelt ahlukaniswe ngokwendalo KwaZulu-Natali naseMpumalanga Kapa, eNingizimu Afrika. Okokuqala, ngisebenzise indlela enezici eziningi ukuhlola imiphumela yokwakheka kwezwe (okungukuthi, ikhwalithi ye-matrix), ukuhlukaniswa kwendawo yokuhlala (okungukuthi, ibanga lokuhlukaniswa nosayizi wesiqephu) kanye nokuhlukahluka kwendawo yokuhlala (okungukuthi, inkimbinkimbi yesakhiwo sehlathi) ezinhlobonhlobo zezinyoni ze-alpha (i-taxonomic, i-functional, i-phylogenetic kanye ne-functional-phylogenetic diversity) yawo wonke umphakathi, okuhlangene okuncike emahlathini (okungukuthi uchwepheshe wamahlathi) kanye nezingancikile emahlathini (okungukuthi i-forestry generalist). Okwesibili, ngidalule ukuthi izici eziyinhloko zokwakheka kwezwe (okungukuthi inani lendawo yokuhlala, ukuhlukahluka kwe-matrix kanye nesilinganiso sebanga lokuhlukaniswa) kuthinte kanjani ukuhlukahluka kwezinyoni zasendaweni futhi nganquma ukubaluleka kokuhlukahluka kokwakheka kwezwe ngakunye kusetshenziswa amamodeli we-mixed-effect linear (LME). Okwesithathu, ngenze imephu yokuxhumana kweziqephu ezimqoka ze-Southern Mistbelt Forest ezinikeza ukuhlukahluka okuphezulu kakhulu (≥ 50 %) kwezinyoni ngokusebenzisa inhlanganisela yendlela engabizi kakhulu nendlela ye-ecological circuit theory, ukuhlonza izici zezwe ezikhuthaza noma ezithiya ukuxhumana kwendawo yokuhlala kanye nezinhlobo zehlathi ezikhethekile nezinomkhawulo wokuhlakazeka (i-Dove Lemon kanye ne-Aplopelia larvata) ukuze iqondise ukuxhumana kwendawo yemiphakathi yamahlathi ezinyoni endaweni ehlukene ye-mozayikhi yehlathi. Engikutholile kwembula: i) imiphumela emihle yekhwalithi ye-matrix ezungezile endaweni ilamula imiphumela engemihle yokuhlukana kwezindawo zokuhlala emaqenjini endawo ezinyoni zehlathi; ii) ukukhuthaza ubunkimbinkimbi bezitshalo zendawo kungaba nomthelela ekulahlekeni kwezinhlobo ezincike ehlathini (okungukuthi ongoti bamahlathi); iii) inani lendawo yokuhlala endaweni ezungezile laliyisibikezelo esibaluleke kakhulu sokuhlukahluka kwezinyoni zasendaweni kanye nokuthuthukiswa kokuxhumana kwendawo yokuhlala phakathi kweziqephu eziwumgogodla zamahlathi e-Southern Mistbelt Forest ehlukanisiwe. Ngakho-ke, ngincoma ukulondoloza nokwandisa ukumboza kwehlathi endaweni ukuze kuqinisekiswe ukusinda kwesikhathi eside kwezinhlobo zamahlathi kulesi simiso sezinto eziphilayo ezihlukene ngokwendalo.
Doctoral Degree. University of KwaZulu-Natal, Pietermaritzburg.