Part II
Figure II.0
Savanna landscape, Serengeti NP.
The tropical savanna biome, defined most simply by the coexistence of trees and grasses, covers nearly half of Africa’s surface south of the Sahara (Figure II.1). Much attention has been given to explaining why the tree cover doesn’t close up. But while the shade cast by tree canopies can suppress grasses, if sufficiently dense, sunlight is not the main limitation where savanna vegetation formations prevail. It is water in the soil, supplied seasonally and somewhat erratically within seasons by rainfall, and redistributed spatially. Competition among trees and grasses operates primarily underground in the rooting zone and thus out of sight. It takes place amid the mat of grass roots and the roots of woody plant seedlings probing for soil moisture and the mineral nutrient resources that this water conveys. The competitive interaction enters a second stage once established tree saplings emerge from the grass layer, only to be burnt back periodically by the recurrent fires sustained by the grasses. While awaiting a sufficient interval between fires to raise their foliage above the flame zone, juvenile trees are exposed to further tissue losses and damage from browsing herbivores. Grasses are superbly adapted to accommodate variable rainfall, withstand fires and tolerate herbivory, as the chapters forming this section will reveal. The feature defining savanna formations is specifically the presence of a grass layer sufficiently well-developed to support fires. Hence grasslands lacking trees are functionally allied in a broadened category of tropical grassy biomes.1 What needs explanation is where and how woody plants manage somehow to establish and persist in regions where grasses dominate.
Figure II.1
African biomes, distinguishing evergreen forest in green, savanna in tan, grassland in brown, ericaceous shrubland in red, arid shrubland and desert in yellow, plus lakes or wetlands in blue.
(from World Wildlife Fund Ecoregions, reproduced in Jones et al. (2013) Soil Atlas of Africa)
Textbooks commonly place savannas climatically where precipitation is too low to support closed woodland or forest; effectively as a seasonally dry form of forest (Figure II.2). Grasslands are located where precipitation is too low to support trees. Both placements are misleading. The feature distinguishing savanna climates is the recurrence of a dry season receiving too little rainfall to support much plant growth. The amount of rain falling during the wet season months may be no different from that associated with forests. South Africa’s open grasslands extend into regions that are wetter than those associated with savanna, not drier (Figure II.3). Where rainfall is generally low but more evenly distributed through the year, savanna grades into low thicket or shrub steppe. Where winter rather than summer months are wet, ericaceous shrubland prevails. Closed-canopy forests develop where soil moisture remains ample for tree growth year-round.
Figure II.2
Ordination of world biomes in relation to temperature and precipitation. Savanna is misleadingly located as a form of seasonally dry forest.
(source: Wikipedia, modified from Whittaker (1975) Communities and Ecosystems)
Figure II.3
Biome distribution in South Africa (from Mucina, L; Rutherford, M. (2006) Strelitzia 19: The Vegetation of South Africa, Lesotho and Swaziland). Note the broad extent of grassland in the north-eastern interior, the savanna extension into the arid western interior and the savanna mosaic interspersed with thicket patches spreading along the coast.
The dependence of the vegetation cover on soil moisture generates a further characteristic feature of savannas: spatial heterogeneity in the tree canopy cover. Forest patches form in moist hollows or flanking rivers. Impermeable layers in the soil may keep trees out. Soil texture and depth along with topography modify the tree cover. Recognising this heterogeneity, it has been proposed that 80 percent or more of the landscape should have an adequately robust grass layer in order to be mapped within the savanna biome.2
My experience of African savannas was honed within the Hluhluwe-iMfolozi Park, situated in foothills below the interior escarpment in the northern region of KwaZulu-Natal formerly known as Zululand. It encloses a mean annual rainfall (MAR) gradient from 600 to 950 mm within its 900-km2 bounds. Thorn savanna prevails in the dry thornveld in the low-lying south-west and a grassland–forest–thicket mosaic in the moist north-east. The local composition of the trees and grasses, and thus where I was most likely to encounter white rhinos, depended further on the underlying soils, in places modified by past human settlements. Similar contributions to spatial heterogeneity occur in savannas elsewhere, but gradients were especially compressed because of the topographic diversity associated with the transitional location of this park.
Savanna ecology is thus intimately entwined with the ecology of grasses. In the first chapter introducing Part II, I will describe how climate, soil features, hydrology and fire regimes generate heterogeneity in tree:grass ratios. This concludes with an appendix introducing some of the prominent tree and grass species. The next chapter details the physiological processes governing competition between grasses and trees, taking place largely underground. The third chapter shifts the focus above-ground to population interactions mediated by repeated fires. The final chapter looks back on how vegetation features have changed through time, particularly during the period when hominins evolved. Not covered in this section are the profound impacts that large mammalian herbivores, in interaction with fire, can have in modifying vegetation and hence ecosystem function. This is deferred to Part III of the book, after the large herbivore fauna has been introduced.
References
1.Parr, CL, et al. (2014) Tropical grassy biomes: misunderstood, neglected, and under threat. Trends in Ecology & Evolution 29:205–213.
2.Scholes, RJ; Walker, BH. (1993) An African Savanna. Synthesis of the Nylsvley Study. Cambridge University Press, Cambridge.
3.Mucina, L; Rutherford, M. (2006) Strelitzia 19: The Vegetation of South Africa, Lesotho and Swaziland. South African National Biodiversity Institute, Pretoria.