The biogeography and community structure of the fishes of South African estuaries was investigated. In all, 109 systems were examined representing two broad types: temporarily closed and permanently open estuaries. Multivariate analyses of the fish communities identified three biogeographic regions. A cool-temperate region extended along the west and southwest coasts; a warmtemperate zone stretched along the south, southeast and east coasts and a subtropical region occurred along the east coast. The boundaries of these biogeographic regions were also delineated. The general physico-chemical characteristics of the estuaries within the three biogeographic regions also reflected regional differences in climate, rainfall and ocean conditions. Estuarine temperatures followed the trend for marine coastal waters, decreasing from subtropical estuaries toward cool-temperate systems. The low rainfall and runoff in the warm-temperate region together with high evaporation rates and strong seawater input resulted in higher salinities in these estuaries. These factors also accounted for the predominantly clearer waters in warm-temperate estuaries. The estuaries in the three biogeographic regions were also shown to contain somewhat distinctive fish assemblages. Temperature and salinity appeared to be the two main factors affecting the distribution and abundance of fishes in South African estuaries. Subtropical systems were characterised by fishes mostly of tropical origin as well as certain south coast endemic species. Warm-temperate estuaries were dominated by endemic taxa with some tropical species also present. The fish fauna of cooltemperate estuaries mostly comprised south coast endemic species with cosmopolitan and temperate taxa also present. Certain functional components of the ichthyofauna also exhibited slight differences between regions. Freshwater fishes were a major component of closed subtropical estuaries while estuarine resident species were more abundant in warm-temperate estuaries. Overall, estuarine-dependent marine species dominated the fish fauna of the estuaries in all biogeographic regions, signifying that South African estuaries perform a vital nursery function for this group of fishes. Slight differences were also apparent in the trophic structure of the fishes; these were related to environmental differences between regions. Zooplanktivores and fishes that feed on aquatic macrophytes/invertebrates assumed a relatively higher importance in warm-temperate systems. Overall, detritivores dominated the estuarine fish fauna in all regions, indicating that detritus forms the main energy source in South African estuaries.

Goatfishes inhabit inshore reefs and corals and are commercially important across their distribution in the Western Indian Ocean (WIO). The biogeography of these species in the WIO has not been explored with regards to their levels of diversity and relationships among regions. The genetic connectivity and differentiation of three goatfishes of the genus Parupeneus (P. barberinus, P. macronemus and P. rubescens) was studied using two mitochondrial genes (ND2 and 16S rRNA) and one nuclear gene (RAG1) using specimens from East and southern Africa, islands around the Mascarene plateau, Oman, Maldives and the Red Sea. Haplotype diversities, networks and AMOVA were used to measure genetic variance among localities and defined regional groups. There were high haplotype (HD > 0.9) and low nucleotide diversities (< 0.006) among all species for all gene regions, suggesting high levels of genetic differentiation among different areas, except for the mtDNA 16S data for P. macronemus and P. rubescens. For all three species, the FST population pairwise values revealed significant differentiation in all datasets for most population pairwise comparisons with the Maldives and genetic connectivity with haplotypes being shared among other localities. The 16S and RAG1, AMOVA for P. barberinus revealed a significant (P < 0.05) strong genetic structure among groups, for example P = 0.00 was estimated in the 16S data for four groups (the Maldives, WIO islands, Kenya and eastern mainland). This study found evidence for regional differentiation within the WIO for these three species supporting the presence of genetic breaks among areas. This differentiation could be either due to the historical isolation among areas or due to geographic and oceanic barriers such as the Mascarene Plateau and the Agulhas Current eddies in the Mozambique Channel. The effects of oceanographic features and physical barriers in the species distribution range and the dispersal potential based on the life history features of the species can have an influence on the genetic structuring of a population. It is also important to note that the length of the pelagic larval phase is just one factor affecting dispersal in marine organisms that can also explain the difference in genetic population structure. Unfortunately there is no specific information on the larval dispersal of these three goatfish. Therefore, studies are needed to be conducted on the specific biology and life history strategies of each Parupeneus species. These results suggest the importance of other factors, such as currents, and larval retention that may cause strong differentiation. These factors should also be considered when observing larval dispersal and its effect on population genetic structure. This study support the hypotheses that physical factors, processes (geographic barriers and oceanographic characteristics) and life history parameters need to be studied to understand the genetic differentiation of these Parupeneus reef fishes.

Dissertation (PhD)--University of Stellenbosch, 2006. / ENGLISH ABSTRACT: The central goal in macroecology is to determine species diversity patterns across ecological gradients. Altitudinal and latitudinal patterns in species richness are often assumed to be analogous. Furthermore, the primary mechanisms underlying richness patterns along these two gradients might be similar. To date, few studies have tested whether the hypotheses proposed to explain latitudinal richness variation apply to patterns across altitude. This study therefore tests several hypotheses proposed to explain patterns in species diversity (i.e. ambient energy, productivity, area and geometric constraints) and their underlying mechanisms using altitudinal gradients in epigaeic ant and beetle species richness in the Greater Cederberg Biodiversity Corridor (GCBC) (Western Cape, South Africa). The study was conducted across an altitudinal gradient that was laid out from sea level to the top of a mountain (approximately 2000 m above sea level) and down the other side thereof. First, it was determined how the ant and beetle assemblages differ between the main vegetation types included in the transect and which environmental variables might underlie these differences. Thereafter, the variation in species richness and range size patterns of the two groups was investigated across the full altitudinal gradient. This is the first study that tests the applicability of two mid-domain models across such an altitudinal gradient using both complete and partial assessments. The models explained large proportions of the variance in range sizes across three domains but the ranges could have been constrained to show peaks in the middle of the domains due to the way in which the boundaries of the domains were selected. By contrast, the mid-domain models were not important in explaining species richness patterns, which suggests that they cannot explain diversity across the gradient. The species richness patterns of the two groups did not show the predicted mid-altitudinal peak. Moreover, it was demonstrated that different processes structure ant and tenebrionid assemblages across the same altitudinal transect. Ant species diversity was highly correlated to contemporary climatic variables, while historical factors appear to play a more important role in structuring tenebrionid beetle assemblages. Furthermore, support was found for the species energy theory in the ant assemblages, as well as for two of its underlying mechanisms, namely the more individuals hypothesis and the niche position mechanism. These results suggest that there are likely to be substantial and complex changes to ant assemblages under the predicted climate change scenarios for the region. Given the crucial role played by this group in ecosystem functioning (e.g. myrmecochory) it is suggested that these responses are not likely to be a response solely to vegetation changes, but might also precipitate vegetation changes. This study also forms the basis of a long-term monitoring programme to establish baseline data for the epigaeic ants and tenebrionids and to monitor changes in these communities due to climate change. / AFRIKAANSE OPSOMMING: Een van die sentrale idees in makro-ekologie is om die patrone in spesies diversiteit oor ekologiese gradiënte te ondersoek. Verder word daar aangeneem dat spesie rykheidspatrone oor hoogte- en breedtegradiënte analoog is aan mekaar en dat die primêre onderliggende meganismes van die patrone dieselfde kan wees oor hierdie twee gradiënte. Tot dusver het min studies getoets of die voorgestelde hipoteses wat breedtegradiënte in spesie rykheid verduidelik van toepassing is op hoogtegradiënte. Hierdie studie toets dus verskeie van hierdie hipoteses (aanvoelbare temperatuur, produktiwiteit, area en geometriese beperkinge) en hulle onderliggende meganismes in mier en kewer spesie rykheid in die Groter Cederberg Biodiversiteits Korridor (GCBK) (Wes Kaap, Suid Afrika). Die studie is uitgevoer oor ‘n hoogtegradiënt wat vanaf see vlak tot ongeveer 2000 meter bo seevlak en weer aan die ander kant van die berg af uitgelê is. Eerstens is daar bepaal hoe die mier en kewer diversiteit verskil tussen die hoof planttipes wat oor die hoogtegradiënt voorgekom het en watter omgewingsveranderlikes daarvoor verantwoordelik is. Daarna is die variasie in spesie rykheid en area van verspreiding van die twee groepe ondersoek oor die hele hoogtegradiënt. Hierdie is die eerste studie wat die toepaslikheid van twee mid-domein modelle oor so ‘n hoogtegradiënt toets met behulp van volledige en gedeeltelike ondersoeke. Die modelle het baie van die variasie in area van verspreiding verduidelik oor drie domeine maar die areas van verspreiding kon beperk gewees het om pieke in die middel van die domeine te vorm as gevolg van die manier waarop die grense van die domeine gekies is. In teenstelling, het die modelle nie spesie rykheid verduidelik nie en dus kan hulle nie spesie diversiteit oor hierdie gradiënt verduidelik nie. Die spesie rykheidspatrone van die twee groepe het nie die verwagte piek by midhoogte gewys nie. Verder het verskillende prosesse mier en kewer groeperings oor die hoogtegradiënt gestruktureer. Mier diversiteit was hoogs gekorroleer met kontemporêre klimaatsveranderlikes, terwyl historiese faktore belangriker was vir die kewers. Die spesie-energie teorie was ondersteun deur die data, asook die meer individue hipotese en die nis posisie meganisme. Hierdie resultaat dui daarop dat daar moontlik komplekse veranderinge in mier groeperings gaan plaasvind soos die klimaat verander. Miere vervul belangrike ekologiese prosesse in ekosisteme, wat beteken dat die laasgenoemde verandering nie bloot net gaan plaasvind as gevolg van veranderinge in die plantegroei nie, maar dat hulle self ook veranderinge kan veroorsaak. Hierdie studie vorm ook die basis van ‘n langtermyn moniteringsprogram om basislyn data vir hierdie twee ekologies belangrike groepe vas te stel en om veranderinge wat in hierdie gemeenskappe plaasvind, as gevolg van klimaatsverandering, te monitor.

Five techniques were used to predict the potential biogeography of the four alien plant species, Acacia longifolia, Acacia mearnsii, Opuntia ficus-indica and Solanum sisymbrifolium. Prediction was based on five environmental factors, median annual rainfall, co-efficient of variation for rainfall, mean monthly maximum temperature for January, mean monthly minimum temperature for July and elevation. A geographical information system was used to manage the data and produce the predictive maps. The models were constructed with presence and absence data and then validated by means of an independent data set and chisquared tests. Of the five models used, three (the range, principal components analysis and discriminant function analysis) were linear while the other two (artificial neural networks and fuzzy logic) were non-linear. The two non-linear techniques were chosen as a plant's response to its environment is commonly assumed to be non-linear. However, these two techniques did not offer significant advantages over the linear methods. The principal components analysis was particularly useful in ascertaining the variables that were important in determining the distribution of each species. Artifacts on the predictive maps were also proved useful for this purpose. The techniques that produced the most statistically accurate validation results were the artificial neural networks (77% correct median prediction rate) and the discriminant function analysis (71% correct median prediction rate) while the techniques that performed the worst were the range and the fuzzy classification. The artificial neural network, discriminant function analysis and principal component analysis techniques all show great potential as predictive distribution models.

The invasion of Acacia mearnsii in the Kouga catchment, Eastern Cape, South Africa, has various negative impacts on the ecosystem. These impacts include: reduced species richness, increased water use, increased nutrients and increased N cycling rates. The native shrubby fynbos vegetation has adapted to the acidic nutrient poor soils and Mediterranean climate of the Kouga Mountains. Fynbos, however, is currently being out competed by the much taller Acacia mearnsii trees, due to their competitive nature and ability to fix nitrogen, thereby enriching the soil. The invaded sections of the valley bottoms and lower hill slopes are characterised by an almost complete monoculture of Acacia mearnsii, with very few fynbos species still present. The Department of Water and Environmental Affairs sponsored Working for Water programme started clearing Acacia mearnsii in 1996 in the Kouga Mountains. Cleared sites have remained bare for long periods, indicating that soil properties are not favourable for indigenous propagule re-establishment. The aim of this research was to assess how A. mearnsii invasion and clearing affect fynbos recovery through its impact on soils. This was done by characterising vegetation and soil properties on fynbos, infested and cleared slopes. Vegetation cover for various growth forms was determined and a species list was compiled for each plot. The slope angle, surface hardness, litter cover, bare ground cover and soil depth were measured in the field, whereas water repellency, particle size and the chemical composition were measured in the laboratory. Furthermore, the plant establishment capacity of soils from fynbos, infested and cleared slopes was calculated. This was done by germinating fynbos seeds and growing fynbos plants in soils from the various slopes. The effect that invasion and clearing has on soil erosion was quantified using erosion plots on fynbos, infested and cleared slopes. The invasion and clearing of Acacia mearnsii led to an increase in soil nutrients, especially nitrogen, phosphorus, potassium, carbon and manganese. Furthermore, soils became more acidic, with increased water repellency and reduced surface hardness. The vegetation changed to a tree-dominated structure, replacing the native species. Native plant germination was relatively unaffected by invasion and clearing, with an increase in germination just after clearing. Plant growth of a native grass, Themeda triandra, and herb, Helichrysum umbraculigerum, has increased on soils from cleared slopes. This study showed that soil movement increased on slopes which are invaded and cleared of Acacia mearnsii, with erosion rates doubling on invaded slopes