Mangroves in South Africa are prone to anthropogenic pressures such as browsing and trampling by livestock and by natural disturbances such as drought and floods. These mangroves exist at one of the most southerly limits in the world providing a unique opportunity to study recovery and resilience to change. This study used long term data at three South African study sites located in the subtropical (St Lucia Estuary) and warm temperate regions (Nxaxo and Kobonqaba estuaries) to compare mangrove growth rate, population structure and responses to disturbance. The first objective was investigated at St Lucia Estuary where site specific or microhabitat differences were measured to assess the influence of these on mangrove growth and population structure. It was suggested that site - specific variability would ensure mangrove survival and analysis showed that seedlings were present at different sites in different years. There was always some recruitment but often little survival to the next size class. Over time seedling numbers were quite variable and self – thinning of adults was evident. The second objective was to identify the environmental factors influencing population density and growth. Results showed that sediment moisture and salinity influenced seedling and adult density due to fluctuations in estuary water level. Mangrove growth rates for Avicennia marina in terms of height was faster (40 – 75 cm yr-1 ) at Site 1 where conditions were waterlogged and moderately saline and slower (5 – 25 cm yr1 ) in dry and hypersaline conditions at Site 4. Overall mangroves at St Lucia have shown persistence through drought and fluctuations in environmental conditions however resilience may be hindered at sites that are subjected to partially flooding and prolonged waterlogging. The second site was Nxaxo Estuary in the Eastern Cape where cattle exclusion plots (nonbrowsed) and control plots (browsed) were used to assess the recovery of A. marina trees from cattle browsing. Trees in the browsed plots showed no vertical growth while trees in the non-browsed plots grew significantly faster (5.4 ± 0.5 cm yr-1). When cattle were prevented from entering the mangrove area, the trees recovered fairly quickly and within 3 years growth was similar in both the non – browsed (17.9 ± 3.2 cm yr -1) and browsed plots (18 ± 1.6 cm yr -1). Lastly, mangrove recovery was investigated at Kobonqaba Estuary where long-term closure of the mouth to the sea resulted in high water levels, inundation of pneumatophores and die-back of the majority of the mangroves. This study measured the recovery of the mangroves by assessing changes in vegetation cover and sediment characteristics along transects. High initial porewater nutrient concentrations promoted salt marsh growth and plants increased average cover from 0% in 2011 to 18.9% (2013) and 50% in 2015. The total number of A. marina individuals increased from only seven to 27 individuals over four years. Salt marsh competition and facilitation will likely influence mangrove recovery in the future. Overall it was concluded that when a short term pressure (cattle browsing) is alleviated, mangrove forests even at warm temperate sites are able to recovery rapidly by increasing growth and seedling establishment. However, mangroves in the subtropical sites show more resilience and recovery potential to long term pressures such as fluctuating environmental conditions because of faster growth rates.

In South Africa mangrove forests are located in estuaries from Kosi Bay in KwaZulu-Natal (KZN) to Nahoon Estuary in the Eastern Cape. The aims of this study were to determine the present state of mangroves in KwaZulu-Natal, by assessing the current population structure, the changes in cover over time and associated anthropogenic pressures. A second objective of this study was to determine the effect of harvesting on the population structure and sediment characteristics in the Mngazana mangrove forest. To determine if harvesting was sustainable at Mngazana Estuary; the growth and mortality rates and associated growth conditions were measured. Finally by using population modelling sustainable harvesting limits were determined by predicting the change in population structure over time. The study focussed on the KwaZulu-Natal province as a fairly recent study addressed mangrove distribution and status in the Eastern Cape Province. A historical assessment of all mangroves forests in KwaZulu-Natal (KZN) revealed that the potential threats to mangroves in South Africa include; wood harvesting, altered water flow patterns coupled with salinity changes, prolonged closed-mouth conditions and subsequent changes to the intertidal habitat. As a result mangroves were completely lost from eleven estuaries in KZN between 1982 and 1999 and a further two estuaries by 2006. Mangroves only occurred in those estuaries where the mouth was open for more than 56 percent of the time with the exception of St Lucia, where the mouth has been closed for longer but the mangrove communities have persisted because the roots of the trees were not submerged. All mangrove forests in KZN were regenerating in terms of population structure as they had reverse J-shaped population curves as well as high adult: seedling ratios. Kosi Bay and Mhlathuze Estuary were two of the larger forests that showed signs of harvesting (presence of tree or branch stumps), but the greatest threat to smaller estuaries seems to be altered water flow patterns due to freshwater abstraction in the catchments and the change of land use from natural vegetation to sugar-cane plantations. These threats affect the hydrology of estuaries and the sediment characteristics (particle size, redox, pH, salinity, temperature) of the mangrove forests. The environmental conditions under which the mangrove forests currently exist were determined for five species. Lumnitzera racemosa and Ceriops tagal exhibited a narrow range of conditions as these species are only found at Kosi Bay, while Avicennia marina, Bruguiera gymnorrhiza and Rhizophora mucronata were found to exist under a wider range of conditions. The growth rate and response to environmental conditions of the three dominant species were important to determine as these species are impacted by harvesting. Mangrove growth rates were measured at Mngazana Estuary in the Eastern Cape, the third largest mangrove forest in South Africa. Areas of this estuary where mangroves harvesting has occurred, show significant differences in sediment characteristics as well as changes in population structure in harvested compared to non harvested sites. The growth rate (in terms of height) of Avicennia marina individuals increased from seedlings (0.31 cm month-1) to adults (1.2 cm month-1), while the growth of Bruguiera gymnorrhiza stabilised from a height of 150 cm at 0.65 cm month-1. The growth of Rhizophora mucronata peaked at 0.72 cm month-1 (height 151-250 cm) and then decreased to 0.4 cm month-1 for taller individuals. Increases in diameter at breast height (DBH) ranged between 0.7 and 2.3 mm month-1 for all species. Some environmental variables were found to be important drivers of growth and mortality of individuals less then 150 cm. A decrease in sediment pH significantly increased the mortality of Avicennia marina seedlings (0-50 cm) (r = - 0.71, p<0.05) and significantly decreased the growth of Rhizophora mucronata and Bruguiera gymnorrhiza seedlings (r = -0.8, r = 0.52 – p < 0.05 respectively). At Mngazana Estuary, mortality of this species showed a positive correlation with sediment moisture content indicating that this species prefers drier conditions. The density of Rhizophora mucronata was significantly correlated to porewater temperature in Northern KZN as was the growth of adult (>300 cm) Rhizophora trees at Mngazana Estuary. Mortality of Avicennia marina individuals (51-150 cm) was related to tree density indicating intraspecific competition and self thinning. Selective harvesting of particular size classes of Rhizophora mucronata was recorded when comparing length of harvested poles (~301 cm) and the size class distribution of individuals. Taking into account the differences in growth rate for each size class for this species it will take approximately 13 years to attain a height of 390 cm which is the height at which trees are selected for harvesting at this estuary. This is 2.6 times slower than those individuals growing in Kenya. The feasibility of harvesting is dependent on the growth rate of younger size classes to replace harvested trees as well as the rate of natural recruitment feeding into the population. Different harvesting intensity scenarios tested within a matrix model framework showed that limits should be set at 5 percent trees ha-1 year-1 to maintain seedling density at > 5 000 ha-1 for R. mucronata. However harvesting of Bruguiera gymnorrhiza should be stopped due to the low density of this species at Mngazana Estuary. Harvesting of the tallest trees of Avicennia marina can be maintained at levels less than 10 percent ha-1 year-1. Effective management of mangrove forests in South African is important to maintain the current state, function and diversity of these ecosystems. Management recommendations should begin with determining the freshwater requirements of the estuaries to maintain the mouth dynamics and biotic communities and deter the harvesting of (whole) adult trees particularly those species that do not coppice. Further management is needed to ensure that forests are cleared of pollutants (plastic and industrial), and any further developments near the mangroves should be minimized.