Different drivers create spatial vegetation cover and vertical structure in semi-arid African savannas

Important drivers of woody cover in African savannas are rainfall, soil and nutrients, and disturbance factors, such as fire and herbivory. However, very little is known about how these drivers influence woody cover at specific height classes. The main aim of this study was to identify which of these drivers are the best predictors of woody cover at three height classes: shrub (0.75–2.5 m), brush (2.5–5.5 m) and tree (>5.5 m). Percent woody cover estimates were collected using a rapid monitoring technique over 1 700 sites in Kruger National Park, South Africa. Geology (basalt and granite), mean annual rainfall, fire frequency and elephant density were analysed as potential drivers of woody cover. Results indicate that mean annual rainfall was negatively associated with shrub cover, fire frequency was negatively associated with brush cover, and elephant density was negatively associated with tree cover. Patterns of woody cover show that while geology influences the spatial distribution of horizontal woody cover, variation in vertical vegetation structure is created and maintained by top-down disturbance. This provides the first documentation of drivers associated with structurally differentiated woody cover at regional scales. Future studies on woody cover in disturbance-driven environments should not ignore vegetation structure.     

The nutrient status of the agriculturally impacted Gamtoos Estuary,South Africa,with special reference to the river‐estuarine interface region (REI)

• 1. Ecological water reserves are part of the new water management philosophy in South Africa. The estuaries are threatened by excessive water abstraction from their rivers and the consequent reduction in freshwater input. This study was conducted to further improve our understanding of freshwater requirements in terms of nutrients for South African estuaries.
• 2. The Gamtoos Estuary was sampled between November 1996 and November 1998 to determine its nutrient status and freshwater input rates. Furthermore, the possible existence of a distinct river‐estuarine interface (REI) in terms of nutrients has been investigated. Nutrients analysed included phosphate, nitrate, nitrite, ammonia, dissolved organic nitrogen (DON), dissolved organic carbon (DOC), particulate organic nitrogen (PON) as Kjieldahl nitrogen, total particulate phosphorus and total particulate carbon.
• 3. The Gamtoos River was found not to be the ultimate source of nutrients to the estuary, but non‐point sources such as the seepage of fertilisers from adjacent agricultural fields determined to a great extent the nutrient status of the estuary.
• 4. Results showed that it was not possible to identify a defined river‐estuarine interface region for all nutrients. Most were measured in higher concentrations in the lower salinity reaches (<17‰) where we observed the combined effect of inputs from the Gamtoos River, an agricultural drainage pipe, and fertilizer seepage from non‐point sources on nutrient concentration.
• 5. Prolonged water withdrawal for human use impacts the continuous renewal of the nutrient pool so important to other South African estuaries with relative pristine catchment areas. However, reduced fresh water inputs could enhance eutrophication in polluted estuaries due to a decreased flushing potential of the estuary. In this case the Gamtoos Estuary has a freshwater requirement for freshets and floods to “clean” the estuary of accumulated nutrients and other organic material.
• 6. This should be a consideration in further freshwater abstraction policies. Conservation issues arise since the Gamtoos Estuary is one of the few permanently open estuaries along the South African coast and serve in this study as an example and as a first attempt to relate freshwater requirements to nutrient dynamics in a South African context.

Copyright © 2002 John Wiley & Sons, Ltd.