The causes and effects of indigenous C4 grass expansion into a hyper-diverse fynbos shrubland.

The cool season rainfall at our study site should favour C3 rather than C4 grasses. There are, however, several locations where C4 grasses have become dominant, suggesting that rainfall seasonality is not a constraint on distribution. Here, we explored the limitations on C4 grass distribution in a fynbos shrubland. Using δ13C values of SOM, we determined when these grasses had established. We also looked at the role of roads as conduits for establishment and asked what impact these grasses may have on fynbos species richness. We then conducted a field experiment designed to examine the extent to which soil moisture, nutrient availability, and competition with fynbos for resources influence the establishment and growth of the grasses. Finally using aerial photography, we explored the role of changes in land use on distribution. Our results showed that the establishment is recent, that roads may be acting as conduits, and that with establishment, there is a reduction in fynbos species richness and diversity. Disturbance and removing below-ground competition for resources open the system to establishment in wetter areas. This study is the first to look at the potential for C4 grasses expanding into cool season rainfall shrublands such as in Western Australia and South Africa. What is interesting about these results is that C4 grasses can establish and dominate in a cool season rainfall regime. Rather than temperature of the growing season, it is competition for resources from fynbos that prevents these grasses from encroaching.

Woody encroachment over 70 years in South African savannahs: overgrazing, global change or extinction aftershock?

Woody encroachment in 'open' biomes like grasslands and savannahs is occurring globally. Both local and global drivers, including elevated CO2, have been implicated in these increases. The relative importance of different processes is unresolved as there are few multi-site, multi-land-use evaluations of woody plant encroachment. We measured 70 years of woody cover changes over a 1020 km(2) area covering four land uses (commercial ranching, conservation with elephants, conservation without elephants and communal rangelands) across a rainfall gradient in South African savannahs. Different directions of woody cover change would be expected for each different land use, unless a global factor is causing the increases. Woody cover change was measured between 1940 and 2010 using the aerial photo record. Detection of woody cover from each aerial photograph was automated using eCognitions' Object-based image analysis (OBIA). Woody cover doubled in all land uses across the rainfall gradient, except in conservation areas with elephants in low-rainfall savannahs. Woody cover in 2010 in low-rainfall savannahs frequently exceeded the maximum woody cover threshold predicted for African savannahs. The results indicate that a global factor, of which elevated CO2 is the likely candidate, may be driving encroachment. Elephants in low-rainfall savannahs prevent encroachment and localized megafaunal extinction is a probable additional cause of encroachment.This article is part of the themed issue 'Tropical grassy biomes: linking ecology, human use and conservation'.

Herbivore population regulation and resource heterogeneity in a stochastic environment.

Large-mammal herbivore populations are subject to the interaction of internal density-dependent processes and external environmental stochasticity. We disentangle these processes by linking consumer population dynamics, in a highly stochastic environment, to the availability of their key forage resource via effects on body condition and subsequent fecundity and mortality rates. Body condition and demographic rate data were obtained by monitoring 500 tagged female goats in the Richtersveld National Park, South Africa, over a three-year period. Identifying the key resource and pathway to density dependence for a population allows environmental stochasticity to be partitioned into that which has strong feedbacks to population stability, and that which does not. Our data reveal a density- dependent seasonal decline in goat body condition in response to concomitant density-dependent depletion of the dry-season forage resource. The loss in body condition reduced density-dependent pregnancy rates, litter sizes, and pre-weaning survival. Survival was lowest following the most severe dry season and for juveniles. Adult survival in the late-dry season depended on body condition in the mid-dry season. Population growth was determined by the length of the dry season and the population size in the previous year. The RNP goat population is thereby dynamically coupled primarily to its dry-season forage resource. Extreme environmental variability thus does not decouple consumer resource dynamics, in contrast to the views of nonequilibrium protagonists.

Diverse functional responses to drought in a Mediterranean-type shrubland in South Africa.

• Mediterranean-type ecosystems contain 20% of all vascular plant diversity on Earth and have been identified as being particularly threatened by future increases in drought. Of particular concern is the Cape Floral Region of South Africa, a global biodiversity hotspot, yet there are limited experimental data to validate predicted impacts on the flora. In a field rainout experiment, we tested whether rooting depth and degree of isohydry or anisohydry could aid in the functional classification of drought responses across diverse growth forms. • We imposed a 6-month summer drought, for 2 yr, in a mountain fynbos shrubland. We monitored a suite of parameters, from physiological traits to morphological outcomes, in seven species comprising the three dominant growth forms (deep-rooted proteoid shrubs, shallow-rooted ericoid shrubs and graminoid restioids). • There was considerable variation in drought response both between and within the growth forms. The shallow-rooted, anisohydric ericoid shrubs all suffered considerable reductions in growth and flowering and increased mortality. By contrast, the shallow-rooted, isohydric restioids and deep-rooted, isohydric proteoid shrubs were largely unaffected by the drought. • Rooting depth and degree of iso/anisohydry allow a first-order functional classification of drought response pathways in this flora. Consideration of additional traits would further refine this approach.

Strontium isotope analyses of large herbivore habitat use in the Cape Fynbos region of South Africa.

The Cape Fynbos region of South Africa, a global biodiversity hotspot, hosted a diverse large mammal fauna till shortly after permanent European settlement (1652). How these animals survived in this exceptionally nutrient-poor environment is puzzling and it is generally believed that they restricted their movements to the more fertile shale areas. We tested the hypothesis that large herbivores avoid nutrient-poor limestone and sandstone fynbos shrublands in favour of shale-derived renosterveld vegetation using strontium (Sr) isotope analysis. If this technique could reconstruct the preferred feeding habitats of the contemporary fauna, it might also be useful for reconstructing the preferred feeding grounds of an extinct fauna. Using the assumption that small rodents have spatially restricted foraging activities, we determined the (87)Sr/(86)Sr isotope ratios of rodent teeth to establish the isotopic signal characteristic of the different geological substrates in the area. We then analysed (87)Sr/(86)Sr isotope ratios in the bones of a number of different large herbivores found in De Hoop Nature Reserve using laser ablation multi-collector inductively coupled plasma mass spectrometry. These values were compared to the bioavailable (rodent) values on the respective geological substrates. The technique identified differences in feeding substrate selection between different species and groups of the same species. The results also showed that shale renosterveld shrubland is not the exclusive source of nutrition for the large herbivores. Strikingly different isotope ratios among individuals in some populations pointed to significant dispersal events from distant sources. However, we were unable to pinpoint the exact feeding areas using Sr isotope analysis probably because some animals use a combination of substrates for feeding and because the geology of the study area is complex with graded isotope signals. We suggest that this technique is a valuable additional tool for exploring large mammal foraging behaviour on habitats associated with contrasting and less complex geology.

The global distribution of ecosystems in a world without fire.

This paper is the first global study of the extent to which fire determines global vegetation patterns by preventing ecosystems from achieving the potential height, biomass and dominant functional types expected under the ambient climate (climate potential). To determine climate potential, we simulated vegetation without fire using a dynamic global-vegetation model. Model results were tested against fire exclusion studies from different parts of the world. Simulated dominant growth forms and tree cover were compared with satellite-derived land- and tree-cover maps. Simulations were generally consistent with results of fire exclusion studies in southern Africa and elsewhere. Comparison of global 'fire off' simulations with landcover and treecover maps show that vast areas of humid C(4) grasslands and savannas, especially in South America and Africa, have the climate potential to form forests. These are the most frequently burnt ecosystems in the world. Without fire, closed forests would double from 27% to 56% of vegetated grid cells, mostly at the expense of C(4) plants but also of C(3) shrubs and grasses in cooler climates. C(4) grasses began spreading 6-8 Ma, long before human influence on fire regimes. Our results suggest that fire was a major factor in their spread into forested regions, splitting biotas into fire tolerant and intolerant taxa.

Seed size and seedling emergence: an allometric relationship and some ecological implications.

We develop a geometric model predicting that maximum seedling emergence depth should scale as the cube root of seed weight. We tested the prediction by planting seeds from 17 species ranging in weight from 0.1 to 100 mg at a variety of depths in a sand medium. The species were spread across 16 genera and 13 families, all occurring in fire-prone fynbos shrublands of South Africa. Maximum emergence depth was found to scale allometrically with seed weight with an exponent of 0.334, close to the predicted value. We used the allometry to predict recruitment response to experimentally simulated variation in fire intensity. Five species with small (<2 mg) seeds and five with large (>10 mg) seeds were planted at ≤20-mm and 40-mm depths and exposed to low and high heat treatments and a control. The allometric equation predicted that species with large seeds would be able to emerge from a depth of 40 mm but those with small seeds would not. Only 1% of 481 seedlings from small-seeded species emerged from the 40-mm planting compared with 40% of 626 seedlings from the large-seeded group. The simulated fire treatments killed seeds in shallow, but not deeper, soil layers. At simulated high fire intensities, seedling emergence was poor in small-seeded species but good in large-seeded species, with most seedlings emerging from the 40-mm planting depth. Seed size could be a useful general predictor of recruitment success under different fire intensities in this system. We suggest that allometric relationships in plants deserve wider attention as predictive tools.

Evidence for widespread pollen limitation of fruiting success in Cape wildflowers.

We used spot checks of stigmatic pollen deposition and hand-pollination experiments to test whether fruit production in Cape wildflower populations is limited by pollen availability. Natural levels of stigmatic pollen deposition were very low (median = 30.0% of flowers) in populations of 33 orchid species. We found similarly low levels of fruit set (median = 32% of flowers per plant) in six Orchidaceae and four Amaryllidaceae species. Experimental hand pollination at the whole plant level caused significant increases in fruit production in 11 of the 12 study populations. These results indicate that pollen limitation of fruit set may occur frequently among some plant families in the Cape flora.

Dry mass allocation, water use efficiency and delta(13)C in clones of Eucalyptus grandis, E. grandis x camaldulensis and E. grandis x nitens grown under two irrigation regimes.

Clonal variation in water use efficiency (WUE), dry mass accumulation and allocation, and stable carbon isotope ratio (delta(13)C) of crude leaf fiber extracts was determined in six clones of Eucalyptus grandis W. Hill ex Maiden. grown for 16 months in field lysimeters in two soil water regimes. The relationships between delta(13)C and WUE calculated on the basis of leaf, harvestable stem, shoot and whole-plant dry mass accumulation were investigated. There was no clonal variation in dry mass accumulation but clonal allocation to roots, harvestable stems, branches and leaves varied. Water use efficiencies (mass of plant or plant part/water used over 16 months) differed significantly between clones. The clonal ranking of WUE varied depending on the units of dry mass accumulation used. Significant relationships between delta(13)C values and instantaneous water use efficiencies and ratios of internal leaf to ambient CO(2) concentrations were found only in the high soil water treatment. There were no relationships between delta(13)C values and whole-plant, shoot and harvestable stem water use efficiencies and soil water availability. Values of delta(13)C were negatively correlated with dry mass accumulation in the low soil water treatment. At the whole-plant level, WUE was positively correlated with dry mass accumulation in the high soil water treatment. We found significant differences in delta(13)C values between clones and the clonal rankings in delta(13)C and WUE were maintained in both soil water treatments.

Growth and water use of Eucalyptus grandis and Pinus radiata plantations irrigated with effluent.

We studied the growth and water balance of young plantations of Pinus radiata D. Don and Eucalyptus grandis W. Hill ex Maiden irrigated with effluent for 3 years in a climate of high net evaporation. The plantations were irrigated weekly with secondary-treated municipal effluent at the estimated water-use rate, or at nominally twice or half this rate. Control plots were irrigated with bore water at their estimated water-use rate. Both species grew rapidly when irrigated with either effluent or bore water. The eucalypts irrigated with effluent at the estimated water-use rate closed canopy in 24 months, and at 34 months, mean dominant height was 12.1 m, stand basal area was 12.2 m(2) ha(-1), volume was 51.2 m(3) ha(-1), LAI was 5.7, and foliage mass was 6.5 Mg ha(-1). The pines in the corresponding effluent treatment had not closed canopy by 34 months. At this time, mean height was 5.0 m, stand basal area was 9.6 m(2) ha(-1), volume was 29.7 m(3) ha(-1), LAI was 3.5, and foliage mass was 7.3 Mg ha(-1). Water use by eucalypts was consistently higher than by pines, commensurate with their more rapid early growth, but the difference was not in proportion to the difference in leaf area. In the third year (when the eucalypts had a closed canopy), the eucalypts used 22% more water than the pines, but the annual mean LAI of the eucalypts was three times greater than that of the pines. The results suggest that (1) plantation water use by the two species on the same site will be similar for the same stage of canopy development, (2) eucalypts are not inherently more profligate consumers of water than pines when soil water is not limiting, and (3) stomatal control limits growth and water use of E. grandis in arid environments.

Desertification of the eastern Karoo, South Africa: Conflicting paleoecological, historical, and soil isotopic evidence.

The desertification debate in South Africa has benefitted greatly in recent years from the contributions of a wide range of disciplines. In this paper we review the conflicting and supporting evidence for degradation in the eastern Karoo as reported in recent archaeological, historical, and stable carbon isotope studies as it relates to three key aspects of the debate: the precolonial environment, the rate and nature of change, and the relative contributions of humans and climate to the process. First, all studies suggest a greater grassiness at some time in the past, but researchers disagree on the timing of the switch to more shrubby conditions in the eastern Karoo. Second, regional rainfall records for the past 2 decades reveal an above-average rainfall period, and numerous long-term surveys show an increase in grass cover over the same period. These findings question the expanding Karoo hypothesis as well as the argument that the Karoo's carrying capacity has decreased in recent years. Finally, the relative responsibilities of humans and climate in the degradation process remain poorly understood and generally have not formed the focus of investigation.

Habitat dependent pollination success in a Cape orchid.

We examined the pollination success of Disa uniflora (Orchidaceae) in two habitats. Plants occurring in a rocky gorge were far more successful than plants occurring in an adjacent open valley. More than 55% of flowers in the gorge were pollinated and set fruit compared with less than 25% of flowers in the valley. These differences are explained by the preference of Meneris tulbaghia (Satyridae), the exclusive pollinator of the orchid, for rocky, sheltered habitats. Fruit set of hand-pollinated flowers did not differ significantly between the two habitats, indicating that resources did not account for the variation in fruiting success.

Evidence for rapid faunal changes on islands in a man-made lake : I. Ants.

Few studies of island biogeography have been made on islands in which the time of insularization is precisely known. We tested the effects of island formation on ant species diversity in a man-made lake in South Africa, to determine whether island effects are detectable after only 16 years of insularization. The number of ant species observed at trap-line censuses on islands was significantly correlated with island size (r=0.608; P<0.05) and ant species diversity was generally low compared with similar mainland habitats. Mean species number for all islands, including landbridge islands, was 5.5±3.3 species, and on mainland sites was 7.9±2.85 species. Island effects were more marked on islands <20 ha, which had a mean of 3.3±2.5 species per island. Species number on islands was inversely related to densities of the aggressive Anoplolepis custodiens and A. steingroeveri. These two species were only patchily distributed on mainlands, but these ants were nearly ubiquitous on small islands. Several lines of evidence suggest that this single species domination may be responsible for island effects. Island sites also differed in the number of ant species in different trophic groupings, tending to have fewer granivorous species than the mainland sites, but species in other diet groups were similar in both island and mainland habitats. We conclude that there have been marked changes in the ant faunas on islands smaller than 20 ha apparently due to changes in abundance of the dominant ant species. However, the causes of these changes are unknown.

The costs of leaving home: ants disperse myrmecochorous seeds to low nutrient sites.

Leucospermum conocarpodendron (L.) Buek (Proteaceae) seedlings were excavated several months after a fire in Cape fynbos. Seedlings under burnt parental skeletons had short hypocotyls (mean 25 mm) indicating passive dispersal whereas seedlings in the open were more deeply buried (mean 48 mm) by ants. Soil nutrient concentrations at the site of germination were negatively related to depth of burial and distance from parent. Ant dispersal resulted in seedlings emerging in soils with lower nutrient concentrations than passively dispersed seeds. Tissue analysis supported the soil results with lower P content in seedlings from open (ant dispersed) sites. Seedling survival in the first year of establishment was also lower in open sites, but not significantly so. However seedlings were slightly taller in the open. The results of this study, the first on naturally occurring intraspecific variation in myrmecochory, strongly contradict current explanations for the high incidence of myrmecochory in nutrient poor environments.

When is an island not an island? Insular effects and their causes in fynbos shrublands.

According to the equilibrium theory of island biogeography, insularisation will lead to species loss from habitat remnants. Extinctions will continue untill species number equilibrates at a level appropriate for the size and isolation of the island remnants. We tested whether insularisation leads to species loss by comparing plant species numbers on islands of fynbos shrublands surrounded by Afrotemperate evergreen forest with extensive "mainland" tracts of fynbos. Species area curves for islands and subsamples of mainland had significantly different slopes (z island=0.43, z mainland=0.16). Small islands had the fewest species (less than one fifth) relative to mainland samples of similar size. The species area curves intersect at 590 ha so that reserve sizes of this order of magnitude are needed to avoid species losses relative to extensive areas of fynbos.We compared traits of species on islands and mainlands to determine processes most affected by insularisation. Island floras did not differ from the mainland in the mix of dispersal types, pollinator syndromes or proportion of dioecious species. Islands did have significantly fewer species of low stature and significantly more species that survive fire only as seed and not by resprouting. We infer that the main cause of species loss is change in disturbance frequency. Islands have fewer fires and lose species dependent on frequent fires. We predict that island effects could be reduced by judicious fire management of small reserves.