Southern African Savanna Research Articles

Habitat selection by large herbivores in a southern African savanna: the relative roles of bottom‐up and top‐down forces

Animals must often balance food availability and predation risk when selecting habitat. Here, we examined habitat preferences of large mammalian herbivores in a long‐term fire experiment in the Kruger National Park, South Africa to assess the role of bottom‐up (e.g., forage quantity/quality) and top‐down (e.g., predation risk) processes in driving herbivore distributions. We focused on experimental plots (∼10 ha on average) that have been burned in the winter (August) since 1954 at 1‐ and 3‐yr intervals, or left unburned (n = 3 per burn type). Herbivore distributions (during both day and night) and plant community structure were surveyed on each plot during the growing seasons (November–March) of 2005–2006, 2006–2007, and 2007–2008. Overall, we sighted 4,187 individuals representing twelve species of mammalian herbivores. Impala (Aepyceros melampus), zebra (Equus quagga), and wildebeest (Connochaetes taurinus) comprised 37%, 28%, and 18% of all individuals observed, respectively. Several species such as African buffalo (Syncerus caffer), wildebeest, and giraffe (Giraffa camelopardalis) exhibited a significant trade‐off between food acquisition and minimizing predation risk by foraging in areas with lower density of woody vegetation. We also observed significant day vs. night dynamics in herbivore habitat selection. For example, zebra utilized annual or triennial burns during the day depending on which years the plots were burned, but they avoided triennial burns with dense woody vegetation in favor of more open annual burns at night when predators such as lions (Panthera leo) are more active. Similarly, the smaller, mixed‐feeding impala appeared to use riskier habitats with more diverse forage options during the day (triennial burns and unburned plots) but used less risky habitats at night (annual and triennial burns). Diurnal vs. nocturnal patterns are often overlooked in studies of habitat selection but are necessary for understanding the factors that shape distribution. The variation we observed in herbivore distribution patterns during this three‐year period indicates that different species exhibit different trade‐offs with respect to food and predation risk. Factors such as body size, nutritional requirements, prey escape tactics, and recent fire history likely mediated these interspecific differences.

Beyond precipitation: physiographic gradients dictate the relative importance of environmental drivers on Savanna vegetation.

BackgroundUnderstanding the drivers of large-scale vegetation change is critical to managing landscapes and key to predicting how projected climate and land use changes will affect regional vegetation patterns. This study aimed to improve our understanding of the role, magnitude and spatial distribution of the key environmental factors driving vegetation change in southern African savanna, and how they vary across physiographic gradients.Methodology/Principal FindingsWe applied Dynamic Factor Analysis (DFA), a multivariate times series dimension reduction technique to ten years of monthly remote sensing data (MODIS-derived normalized difference vegetation index, NDVI) and a suite of environmental covariates: precipitation, mean and maximum temperature, soil moisture, relative humidity, fire and potential evapotranspiration. Monthly NDVI was described by cyclic seasonal variation with distinct spatiotemporal patterns in different physiographic regions. Results support existing work emphasizing the importance of precipitation, soil moisture and fire on NDVI, but also reveal overlooked effects of temperature and evapotranspiration, particularly in regions with higher mean annual precipitation. Critically, spatial distributions of the weights of environmental covariates point to a transition in the importance of precipitation and soil moisture (strongest in grass-dominated regions with precipitation<750 mm) to fire, potential evapotranspiration, and temperature (strongest in tree-dominated regions with precipitation>950 mm).Conclusions/SignificanceWe quantified the combined spatiotemporal effects of an available suite of environmental drivers on NDVI across a large and diverse savanna region. The analysis supports known drivers of savanna vegetation but also uncovers important roles of temperature and evapotranspiration. Results highlight the utility of applying the DFA approach to remote sensing products for regional analyses of landscape change in the context of global environmental change. With the dramatic increase in global change research, this methodology augurs well for further development and application of spatially explicit time series modeling to studies at the intersection of ecology and remote sensing.

Disentangling the Relationships between Net Primary Production and Precipitation in Southern Africa Savannas Using Satellite Observations from 1982 to 2010

To obtain a better understanding of the variability in net primary production (NPP) in savannas is important for the study of the global carbon cycle and the management of this particular ecosystem. Using satellite and precipitation data sets, we investigated the variations in NPP in southern African savannas from 1982 to 2010, and disentangled the relationships between NPP and precipitation by land cover classes and mean annual precipitation (MAP) gradients. Specifically, we evaluate the utility of the third generation Global Inventory Monitoring and Modeling System (GIMMS3g) normalized difference vegetation index (NDVI) dataset, in comparison with Moderate-resolution Imaging Spectroradiometer (MODIS) derived NPP products, and find strong relationships between the overlapping data periods (2000–2010), such that we can apply our model to derive NPP estimates to the full 29-year NDVI time-series. Generally, the northern portion of the study area is characterized by high NPP and low variability, whereas the southern portion is characteristic of low NPP and high variability. During the period 1982 through 2010, NPP has reduced at a rate of −2.13 g∙C∙m−2∙yr−1 (p &lt; 0.1), corresponding to a decrease of 6.7% over 29 years, and about half of bush and grassland savanna has experienced a decrease in NPP. There is a significant positive relationship between mean annual NPP and MAP in bush and grassland savannas, but no significant relationship is observed in tree savannas. The relationship between mean annual NPP and MAP varies with increases in MAP, characterized as a linear relationship that breaks down when MAP exceeding around 850–900 mm.

Investigating potential determinants of the distribution limits of a savanna woody plant:Colophospermum mopane

AbstractQuestionsEarly bioclimatic models predict that climate change in southern African savannas will cause a huge southward and westward range shift of the savanna treeColophospermum mopane(Kirk ex Benth.) Kirk ex J.Léon.C. mopaneis an economically and ecologically important subtropical savanna tree that forms mono‐dominant stands across 30% of southern African savannas. We investigate the validity of these initial range expansion predictions to answer the following questions: what are the regional‐scale drivers of the distribution ofC. mopanein southern African savannas; and what are the landscape‐scale distribution patterns of this species?LocationCentral Lowveld, Kruger National Park, South Africa.MethodsWe investigate the validity of very early range expansion modelling predictions using a regional‐scale, climate envelope niche model, and fine‐scale field mapping of the current boundary, to understand which environmental variables may determine the distribution limit of this signature species.ResultsOur findings indicate that both non‐climatic (dry season day length) and climatic (minimum temperatures) variables limit the regional distribution ofC. mopane. At the landscape scale, the distribution of this species is restricted to the warmer parts of the landscape, suggesting minimum temperature appears to be the primary factor determining its landscape‐scale distribution.ConclusionsThis study provides the first detailed model of environmental factors that may limit the regional distribution ofC. mopane, and allows us to formulate testable hypotheses regarding the determinants of the range of a keystone species.

What do ecologists miss by not digging deep enough? Insights and methodological guidelines for assessing soil fertility status in ecological studies

Soil fertility is one of the major drivers of ecological processes and is therefore frequently investigated in ecological research. Although often referred to in studies, soil fertility is not well quantified. Consequently some studies have resorted to classifying site soil fertility according to the potential fertility associated with underlying geology, ignoring the soil nutrient status of the rootzone. A common protocol is for ecologists to sample the upper soil layers only (<20 cm). Unfortunately these surface layers are those most likely to be altered by the vegetation itself and may not necessarily reflect the influence of the geological substrate. Using examples, we attempt to provide some practical guidelines on how to determine the intrinsic nutrient status of soils. Soil data from five sites in southern African savannas were used to demonstrate: a) when deeper soil sampling may not be needed, b) how to determine which nutrients may be limiting at a site, c) the importance of bulk density measurements and d) the effect of three different sampling methods. Our data illustrate that the effects of fine scale landscape variability on soil nutrients were evident to variable soil depths. Frequent fires affected soils only to depths of <5 cm, the presence of tree canopies affected soils up to 50 cm, while topographic position affected soil nutrients to a depth of 90 cm. Bulk density did not differ between depths nor between treatments within sites, but differed amongst sites. None of the alternative methods used to collect soil samples (i.e. augering vs. digging soil pits and sampling by depth or horizon) resulted in significant differences in nutrient measures. Standardised sampling from at least three depths together with bulk density measurements allow for calculation of nutrient stocks as a measure of intrinsic soil nutrient status, while also providing insights into nutrient distributions with depth, thereby allowing meaningful cross-site comparisons.

Multifunctional Rangeland in Southern Africa: Managing for Production, Conservation, and Resilience with Fire and Grazing

Residents of Southern Africa depend on rangeland for food, livelihoods, and ecosystem services. Sustainable management of rangeland ecosystems requires attention to interactive effects of fire and grazing in a changing climate. It is essential to compare rangeland responses to fire and grazing across space and through time to understand the effects of rangeland management practices on biodiversity and ecosystem services in an era of global climate change. We propose a paradigm of ecologically-analogous rangeland management within the context of multifunctional landscapes to guide design and application of ecosystem-based rangeland research in Southern Africa. We synthesize range science from the North American Great Plains and Southern African savannas into a proposal for fire and grazing research on rangeland in Southern Africa. We discuss how management for the fire-grazing interaction might advance multiple goals including agricultural productivity, biodiversity conservation, and resilience to increased variability under global change. Finally, we discuss several ecological and social issues important to the effective development of sustainable rangeland practices especially within the context of global climate change. The associated literature review serves as a comprehensive bibliography for sustainable rangeland management and development across the savanna biomes of Southern Africa.

Comparing terrain and vegetation-based visibility for explaining sable flight behaviour in a Southern African savanna

In this study, we tested whether terrain-based visibility modelled from a remotely sensed ASTER Digital Elevation Model (DEM) explains sable flight initiation distance (FID) better than vegetation-based visibility measured in the field. We also tested whether the effect of hunting on sable FID varies with spatial scale. We first performed a linear regression analysis relating FID to standardized coefficients of both vegetation- and terrain-based visibility where the variable with the larger coefficient was the better predictor of FID. We latter performed an analysis of covariance (ANCOVA) comparing the slopes relating FID to both measures of visibility, first at the large scale and later at the small scale within the hunting area. Our results suggest that remotely sensed terrain-based visibility predicts the FID of sable better than vegetation-based visibility. We also found that the effect of hunting on sable FID varies with spatial scale.

Interactive Effects of Fire, Rainfall, and Litter Quality on Decomposition in Savannas: Frequent Fire Leads to Contrasting Effects

One of the many ecological processes expected to undergo alteration due to global change is the decomposition of organic matter, with little known concerning the effects that changing disturbance regimes may have. Fire, a critical process in many habitats, is expected to become more common. We measured the decomposition rates of four grass species that differed in litter quality, investigating them under different fire regimes across a savanna rainfall gradient in South Africa. We also collected data on the abundance and activity of fungus-growing termites and recorded measurements of temperature and canopy cover. Overall, decomposition rate followed global models, increasing under warmer and wetter conditions. Litter quality was also significant with higher quality grasses decomposing faster; however, this effect was less pronounced than expected. Fire regimes did not have a consistent effect on decomposition rate along the rainfall gradient. In the most arid savanna type examined, fire had no effect, whereas in the intermediate rainfall savanna burning increased decomposition rate under higher levels of fungus-growing termite activity. In the wetter savannas, fire slowed decomposition, possibly through modification of vegetation structure and potential effects on other invertebrates. Our results demonstrate that grass decomposition in African savannas varies significantly along precipitation gradients, with different factors becoming influential in different habitats. Importantly, we demonstrate that fire does not always act to slow decomposition and that it interacts with other factors to influence the process. These findings have important implications for decomposition in the light of global change models that predict wetter climates and a higher frequency of fires for southern African savannas.

Physiological Cross-Sectional Area of the Masticatory Muscles in the Giraffe (Giraffa camelopardalis)

Numerous studies have investigated how the skeletal morphology is related to feeding behavior (Christiansen and Adolfssen 2005; Wroe et al. 2005; Christiansen and Wroe 2007; Ellis et al. 2009; Koyabu and Endo 2009; Koyabu et al. 2009; Koyabu and Endo 2010), although remarkably few studies have focused on the architecture of masticatory muscles from which bite forces are produced. In this regard, the quantification of physiological cross-sectional areas (PCSA) of muscle is critical for estimation of bite forces. PCSA is generally proportional to and regulates maximum force-generating capacity (Santana et al. 2010). To date, PCSA of masticatory muscles are reported for bats (Burke et al. 1974; Santana et al. 2010), primates (Weijs and Hillen 1985; Anton 1999; Anapol et al. 2008), sloths (Naples 1985), rodents (Weijs and Dantuma 1975; Weijs 1980; Druzinsky 2010), rabbits (Weijs and Dantuma 1980), and pigs (Herring 1980, 1985). Since quantification of muscle PCSA is critical to estimate bite forces in animals, and since yet little is understood about the relationship between the variation of masticatory muscle PCSA and diversity in feeding behavior among mammals, further accumulation of knowledge on masticatory muscle PCSA is highly needed. Here, we report the first quantification of masticatory muscle PCSA in the giraffe. Among ruminants, giraffes have been reported that they can be clearly classified as browsers (Leuthold and Leuthold 1972; Dutoit 1990; Codron et al. 2007). Therefore, investigation on this species should provide useful information on the dichotomy of masticatory biomechanics between grazers and browsers. Their browsing behaviors are supported by some traits such as longer tongue, narrower muzzle, and narrower incisor row. Pellew (1984) have reported that Acacia shoots were most selected by Giraffa camelopardalis tippelskirchi to obtain high protein in the Serengeti National Park, Tanzania. Acacia flowers are also reported to be an important food resource for G. camelopardalis in a Southern African savanna (Dutoit 1990). To date, there have been very few studies on giraffe masticatory anatomy. Sasaki et al. (2001) has previously described the detailed macroscopic anatomy of the masseter muscles in the giraffe. However, the PCSA of masticatory muscles were not quantified, and other masticatory muscles were not incorporated in their study. Thus, in the present study, we macroscopically dissected three carcasses of giraffes and examined the muscle weight and PCSA of masticatory muscles, the medial and lateral pterygoid muscles, masseter muscle, temporal muscle, for the purpose of better understanding of their feeding strategy.

Predicting the spatial distribution of hunted and non-hunted Sable antelope (Hippotragus niger niger) using remotely sensed woody cover in a Southern African savanna

From remotely sensed woody cover, we tested whether sables under hunting pressure preferred closed woodland habitats and whether those not under hunting preferred more open woodland habitats. We applied a two factorial logistic regression analysis to model the probability of occurrence of sable antelope in hunted and non-hunted areas of northwest Zimbabwe as a function of vegetation cover density (estimated by a normalized difference vegetation index (NDVI)). We validated the results by high-spatial resolution imagery derived tree canopy area. We subsequently compared the predictions from the two models in order to compare sable cover selection between hunted and non-hunted areas. Our results suggest that hunted sables are likely to select closed woodland, while non-hunted ones would prefer more open woodland habitats. We also established a significant positive relationship between NDVI and tree canopy cover, thus emphasizing the importance of remote sensing in studies that measure the impact of hunting on habitat selection of targeted species.

The vegetation of Malilangwe Wildlife Reserve, south-eastern Zimbabwe

The vegetation of Malilangwe Wildlife Reserve in Zimbabwe (39 378 ha) was classified and mapped in 2003 using an integrated approach of ground-based measurement and remote sensing. Environmental variables, structure and composition of the tree, shrub and herbaceous layers were sampled at 201 sites across the reserve. Thirty-eight vegetation types were identified using TWINSPAN, and a map of the types was created by supervised classification of a Landsat 7 ETM+ image (2000). A discriminant function analysis showed that the types were primarily separated along a soil textural gradient. A secondary gradient, associated with changes in soil depth, rockiness, slope and topographical position, was also important. Ordination indicated that several other independent environmental gradients appeared to be influencing the vegetation. Comparisons with other southern African savannas are made, and similarities and differences are discussed.

Inter-specific variation in bud banks and flowering effort among semi-arid African savanna grasses

Population viability and productivity of grasses in southern African savannas are dependent upon both successful seed production and tiller recruitment from the belowground bud bank. Relative recruitment rates from buds versus seeds influence population dynamics, genetic diversity, and patterns of vegetation productivity. We assessed patterns in bud bank size and flowering effort in fourteen semi-arid savanna grass species in the Kalahari region of Botswana. There was high inter-specific variability and between-year variability in flowering effort (percentage of tillers flowering). Bud production (number of buds per tiller) exhibited high inter-specific variability, but was more consistent between-years than flowering effort. Relative allocation to flowering versus bud production varied with life history, with longer-lived perennial grasses showing higher bud production and lower flowering effort relative to shorter-lived grasses. Several species showed higher bud production and lower flowering effort in a wet year compared to a dry year, and grass species that are regularly grazed maintained significantly larger bud banks than non-grazed species. These differential demographic responses among co-occurring species suggest that environmental change in semi-arid savannas may alter the composition, relative abundances and diversity of grasses, and that the maintenance of a belowground bud bank is an important factor influencing their resiliency, their capacity to recover from grazing and/or drought, and their persistence and sustainability under changing environmental conditions. Meristem-limitation in species that maintain few viable buds may constrain their population viability under changing conditions in semi-arid savannas.

Patterns of mistletoe infection in four Acacia species in a semi-arid southern African savanna

In a range of systems, studies on mistletoe distribution on the host plant have documented a number of factors that affect their occurrence and spread (Aukema &amp; Martinez del Rio 2002a, Bowie &amp; Ward 2004, Overton 1996, Reid et al. 1995). These patterns can be determined by host specificity, environmental conditions, host plant characteristics (Martinez del Rio et al. 1995) and the movement patterns of dispersal agents (Aukema &amp; Martinez del Rio 2002a, 2002b). In mistletoe plants, host choice can be considerably influenced by the advantages of interacting with relatively abundant hosts (Norton &amp; Carpenter 1998, Norton &amp; De Lange 1999). Besides the relative abundance of host species, characteristics such as branch size, age and height can have a strong effect on mistletoe attachment resulting in size-related mistletoe infection patterns (Overton 1994). Generally positive relationships between mistletoe infection and host size have been demonstrated worldwide (Donohue 1995, Martinez del Rio et al. 1996, Norton et al. 1997, Reid &amp; Stafford Smith 2000) and they have been interpreted in terms of the preferences by dispersing birds to perch and feed in taller trees (Aukema &amp; Martinez del Rio 2002a) and trees accumulating infections as they age (Overton 1994). Aukema &amp; Martinez del Rio (2002a) reported more frequent perching in taller-than-average trees by the phainopepla (Phainopepla nitens), which is the principal disperser of the desert mistletoe Phoradendron californicum. Thus, visits by mistletoe-seed-dispersing birds, and therefore mistletoe seeds received, tend to increase with tree height (Aukema &amp; Martinez del Rio 2002a). Using a simple metapopulation model, Overton (1994) predicted the frequency of parasitized trees to increase with host age. Therefore, assuming that size is a good proxy for age, large trees are likely to be more infected than smaller trees. Reid &amp; Stafford Smith (2000), using experimentally disinfected trees, found that larger trees were disproportionately re-infected with mistletoes. This size–intensity relationship may be used to describe mistletoe infection patterns. However, several previous studies have shown size–intensity relationships to be weak (Aukema &amp; Martinez del Rio 2002a, Donohue 1995, Overton 1994, Reid &amp; Stafford Smith 2000). This indicates that other factors may be important in determining mistletoe infection intensity, including that already parasitized hosts of a specific height are more likely to receive seeds than non-parasitized hosts of the same height or dispersers are likely to be attracted to trees for reasons other than size (Aukema &amp; Martinez del Rio 2002a).

Do freeze events create a demographic bottleneck for Colophospermum mopane?

Frost disturbance is often mentioned in southern African savanna literature, but it is seldom discussed or investigated further. However, it can represent an above-ground disturbance as effective as fire or browsing, depending on the resistance capacity of the effected plants. A severely freeze-damaged stand of Colophospermum mopane along a slope in the Venetia Limpopo Nature Reserve provided an opportunity to investigate the nature of freeze-damage impacts on C. mopane. Is this disturbance a possible demographic limitation of C. mopane preventing its southwards spread? Freeze-damage of individual trees was assessed according to tree height and landscape position — with lower elevations representing the most severe freeze zones and higher elevations representing the least severe. Lower elevation trees were relatively small (2.24m) and coppicing, whilst higher elevation trees were taller (3.65m) with no coppice present. No freeze-damage was observed on tree canopies above 4m in height. Trees<4m in height that had experienced 100% freeze-damage, failed to regrow to their original heights of the previous season. This is a possible driver of the pre-freeze height differences seen across the slope; with trees at low elevations having to recover from freeze events and subsequent topkill more frequently, resulting in a net decrease in tree height for that growing season. It appears that C. mopane has limited resistance to freeze events, and this may be linked to the absence of this species at colder latitudes in the Southern Hemisphere.

Explaining elephant (Loxodonta africana) and buffalo (Syncerus caffer) spatial distribution in the Zambezi Valley using maximum entropy modelling

We used the maximum entropy (MAXENT) technique to explore how anthropogenic and physical factors influence the habitat of elephants and buffalo in a Southern African savanna landscape. Results show that for both species the distance to agricultural fields is the most important factor determining the choice of habitat. Without having used the nature reserves’ and conservancy's boundary as an input in the maximum entropy model, the model was mapping out the high probability of buffalo and elephant presence to coincide within their boundaries. The analysis of the habitat condition show that higher presence of wildlife in the nature reserves compared with cultivated areas is not dependent on the abundance of green biomass, proximity to surface water, or terrain slope, since these do not significantly differ between nature reserves and communal lands. We conclude that the presence of agricultural fields contributes more in shaping the habitats of both buffalo and elephant than other environmental variables. In the communal lands where the agricultural fields are concentrated along the rivers elephants and buffalo are pushed away from the rivers.

An Overview of Nitrogen Cycling in a Semiarid Savanna: Some Implications for Management and Conservation in a Large African Park

Nitrogen (N) is a major control on primary productivity and hence on the productivity and diversity of secondary producers and consumers. As such, ecosystem structure and function cannot be understood without a comprehensive understanding of N cycling and dynamics. This overview describes the factors that govern N distribution and dynamics and the consequences that variable N dynamics have for structure, function and thresholds of potential concern (TPCs) for management of a semiarid southern African savanna. We focus on the Kruger National Park (KNP), a relatively intact savanna, noted for its wide array of animal and plant species and a prized tourist destination. KNP's large size ensures integrity of most ecosystem processes and much can be learned about drivers of ecosystem structure and function using this park as a baseline. Our overview shows that large scale variability in substrates exists, but do not necessarily have predictable consequences for N cycling. The impact of major drivers such as fire is complex; at a landscape scale little differences in stocks and cycling were found, though at a smaller scale changes in woody cover can lead to concomitant changes in total N. Contrasting impacts of browsers and grazers on N turnover has been recorded. Due to the complexity of this ecosystem, we conclude that it will be complicated to draw up TPCs for most transformations and pools involved with the N cycle. However, we highlight in which cases the development of TPCs will be possible.

Biomass burning aerosol absorption measurements with MODIS using the critical reflectance method

[1] This research uses the critical reflectance technique, a space-based remote sensing method, to measure the spatial distribution of aerosol absorption properties over land. Choosing two regions dominated by biomass burning aerosols, a series of sensitivity studies were undertaken to analyze the potential limitations of this method for the type of aerosol to be encountered in the selected study areas, and to show that the retrieved results are relatively insensitive to uncertainties in the assumptions used in the retrieval of smoke aerosol. The critical reflectance technique is then applied to Moderate Resolution Imaging Spectrometer (MODIS) data to retrieve the spectral aerosol single-scattering albedo (SSA) in South African and South American biomass burning events. The retrieved results were validated with collocated Aerosol Robotic Network (AERONET) retrievals. Approximately 67% of the comparisons show a difference between MODIS and AERONET smaller than 0.03, the magnitude of the AERONET uncertainty. The overlap of the two retrievals increases to 88%, allowing for measurement variance in the MODIS retrievals, as well. The ensemble average of MODIS-derived SSA for the Amazon forest station is 0.92 at 670 nm, and 0.84–0.89 for the southern African savanna stations. The critical reflectance technique allows evaluation of the spatial variability of SSA and shows that SSA in South America exhibits higher spatial variation than in South Africa. The accuracy of the retrieved aerosol SSA from MODIS data indicates that this product can help to better understand how aerosols affect the regional and global climate.

Mound-Building Termite (Blattaria: Isoptera) Mound Density and Ecological Correlates in a Southern African Savanna Reserve

Termites are widely distributed in tropical and subtropical regions, where they are dominant members of the soil fauna and are often important ecological engineers (Whitford 2000; Davies et al. 2003). They generally feed below-ground on dead plant material or on standing, dead herbaceous or woody plants and vegetation litter (Jouquet et al. 2005; Berge et al. 2008). Termite mounds can reduce plant stands, with plant colonization hampered by hardness of mound soil or by active termite foraging (Malaka 1977; Pomeroy 1983). Yet sometimes, plant growth can increase on and around intact termitaria (Flemming & Loveridge 2003; Jouquet et al. 2005), while rainfall runoff at the base of the termitarium enables grasses to survive drought (Steinke & Nel 1989). Termitaria can also serve as seed reserves (Steinke & Nel 1989), as some termite species relocate sparsely-distributed food resources into the mound, which in turn, acts as a concentrated central seed store (Dangerfield 1990), and alters rate of seed growth and senescence (Curry 1987).

The nutritional significance of a winter‐flowering succulent for opportunistic avian nectarivores

The winter‐flowering succulent Aloe marlothii provides nectar for many opportunistic avian nectarivores in southern African savannas. We assessed the importance of A. marlothii nectar sugar for opportunistic nectarivores by analysing temporal changes in stable carbon isotope ratios (δ13C) in the tissues of birds in Suikerbosrand Nature Reserve, South Africa. The blood of the 11 most common non‐granivorous opportunistic nectarivores at our site was enriched in 13C by 3.4 ± 1.5‰ during the flowering period of A. marlothii, reflecting the enriched crassulacean acid metabolism (CAM) isotopic signature of nectar (−12.6 ± 0.5‰). This relatively small contribution of A. marlothii nectar to assimilated carbon in whole blood contrasted with that of exhaled CO2 in African Red‐eyed Bulbuls Pycnonotus nigricans and Cape White‐eyes Zosterops capensis. In both these species, the δ13C of breath samples was significantly enriched compared with blood and feathers, and closely resembled that of the nectar, revealing combustion of ingested nectar rather than assimilation. Although our analysis was complicated by the presence of C4 grasses, whose δ13C values are similar to those of CAM photosynthesizers, when considered with previously published feeding observations our data reveal that opportunistic nectarivores feeding on A. marlothii nectar obtain a relatively small fraction of their assimilated carbon, but most of their metabolized carbon, from this seasonally available carbohydrate food resource. Because the δ13C values of insects associated with C3 plants also became enriched during the flowering season, some insect‐eating opportunistic nectarivores may have assimilated A. marlothii carbon indirectly from insects. This study highlights the importance of understanding isotopic routing when assessing the nutritional significance of specific dietary items to consumer communities.

What are the effects of substrate and grass removal on recruitment of Acacia mellifera seedlings in a semi-arid environment?

Acacia mellifera is one of the most important encroaching woody plants in southern African savannas. Previous studies found that this species encroaches far more readily on rocky areas than on sandy substrates, although it grows larger on sandy substrates. Rocky substrates are known to retain more water than sandy substrates, which may be of vital importance during recruitment in semi-arid and arid environments. A number of studies have also indicated that competition with grasses may reduce the recruitment and biomass of tree seedlings. We created an experiment in a semi-arid environment (mean annual rainfall = 388 mm) that tested for the effects of rockiness on A. mellifera recruitment. We also tested the hypothesis that grasses effectively compete with A. mellifera in this environment by simulating the effect of grazing by clipping grasses from half the plots in both the rocky and sandy treatments. Significantly more A. mellifera seedlings established in plots where grasses were clipped than in control plots. A. mellifera seedlings had greater biomass on sandy substrates than on rocky substrates. No significant interaction effects were found between substrate and grass clipping treatment for either seedling number or biomass. We conclude that A. mellifera seedlings are more likely to encroach in habitats with low grass density, although they may achieve greater biomass on sandy soils. Thus, it may be the lower grass density rather than rockiness, which increases the encroachment observed in naturally rocky habitats. These results are also consistent with our observations that adult A. mellifera trees are larger on sandy soils than on rocky soils.