Southern African Savanna Research Articles

Integrating in-situ, Landsat, and MODIS data for mapping in Southern African savannas: experiences of LCCS-based land-cover mapping in the Kalahari in Namibia

Integrated ecosystem assessment initiatives are important steps towards a global biodiversity observing system. Reliable earth observation data are key information for tracking biodiversity change on various scales. Regarding the establishment of standardized environmental observation systems, a key question is: What can be observed on each scale and how can land cover information be transferred? In this study, a land cover map from a dry semi-arid savanna ecosystem in Namibia was obtained based on the UN LCCS, in-situ data, and MODIS and Landsat satellite imagery. In situ botanical relevé samples were used as baseline data for the definition of a standardized LCCS legend. A standard LCCS code for savanna vegetation types is introduced. An object-oriented segmentation of Landsat imagery was used as intermediate stage for downscaling in-situ training data on a coarse MODIS resolution. MODIS time series metrics of the growing season 2004/2005 were used to classify Kalahari vegetation types using a tree-based ensemble classifier (Random Forest). The prevailing Kalahari vegetation types based on LCCS was open broadleaved deciduous shrubland with an herbaceous layer which differs from the class assignments of the global and regional land-cover maps. The separability analysis based on Bhattacharya distance measurements applied on two LCCS levels indicated a relationship of spectral mapping dependencies of annual MODIS time series features due to the thematic detail of the classification scheme. The analysis of LCCS classifiers showed an increased significance of life-form composition and soil conditions to the mapping accuracy. An overall accuracy of 92.48% was achieved. Woody plant associations proved to be most stable due to small omission and commission errors. The case study comprised a first suitability assessment of the LCCS classifier approach for a southern African savanna ecosystem.

Methods to determine the impact of rainfall on fuels and burned area in southern African savannas

In southern African savannas, grass production, and therefore the annual extent of fire, is highly dependent on rainfall. This response has repeatedly been noted in the literature but authors used different input variables and modelling approaches and the results are not comparable. Using long-term fire occurrence data from six protected areas in southern Africa we tested various methods for determining the relationship between antecedent rainfall and burned area. The types of regression model, the most appropriate index of accumulated rainfall, and the period over which to calculate annual burned area were all investigated. The importance of accumulating rainfall over more than one growing season was verified in all parks – improving the accuracy of the models by up to 30% compared with indices that only used the previous year’s rainfall. Up to 56% of the variance in burned area between years could be explained by an 18-month accumulated rainfall index. Linear models and probit models performed equally well. The method suggested in this paper can be applied across southern Africa. This will improve our understanding of the drivers of interannual variation in burned area in this globally important fire region.

Tree cover and biomass increase in a southern African savanna despite growing elephant population

The growing elephant populations in many parts of southern Africa raise concerns of a detrimental loss of trees, resulting in overall reduction of biodiversity and ecosystem functioning. Elephant distribution and density can be steered through artificial waterpoints (AWPs). However, this leaves resident vegetation no relief during dry seasons. We studied how the introduction of eight AWPs in 1996 affected the spatiotemporal tree-structure dynamics in central Chobe National Park, an unfenced savanna area in northern Botswana with a dry-season elephant density of approximately 3.34 individuals per square kilometer. We hypothesized that the impact of these AWPs amplified over time and expanded in space, resulting in a decrease in average tree density, tree height, and canopy volume. We measured height and canopy dimensions of all woody plants around eight artificial and two seasonal waterpoints for 172 plots in 1997, 2000, and 2008. Plots, consisting of 50 x 2 m transects for small trees (0.20-3.00 m tall) nested within 50 x 20 m transects for large trees (> or = 3.0 m tall), were located at 100, 500, 1000, 2000, and 5000 m distance classes. A repeated-measures mixed-effect model showed that tree density, cover, and volume had increased over time throughout the area, caused by a combination of an increase of trees in lower size classes and a decrease in larger size classes. Our results indicate that the decrease of large trees can be attributed to a growing elephant population. Decrease or loss of particular tree size classes may have been caused by a loss of browser-preferred species while facilitating the competitiveness of less-preferred species. In spite of 12 years of artificial water supply and an annual elephant population growth of 6%, we found no evidence that the eight AWPs had a negative effect on tree biomass or tree structure. The decreasing large-tree component could be a remainder of a depleted but currently restoring elephant population.

The impact of shrub encroachment on savanna bird diversity from local to regional scale

AbstractAim Evidence is accumulating of a general increase in woody cover of many savanna regions of the world. Little is known about the consequences of this widespread and fundamental ecosystem structural shift on biodiversity.Location South Africa.Methods We assessed the potential response of bird species to shrub encroachment in a South African savanna by censusing bird species in five habitats along a gradient of increasing shrub cover, from grassland/open woodland to shrubland dominated by various shrub species. We also explored historical bird species population trends across southern Africa during the second half of the 20th century to determine if any quantifiable shifts had occurred that support an ongoing impact of shrub encroachment at the regional scale.Results At the local scale, species richness peaked at intermediate levels of shrub cover. Bird species composition showed high turnover along the gradient, suggesting that widespread shrub encroachment is likely to lead to the loss of certain species with a concomitant decline in bird species richness at the landscape scale. Finally, savanna bird species responded to changes in vegetation structure rather than vegetation species composition: bird assemblages were very similar in shrublands dominated by Acacia mellifera and those dominated by Tarchonanthus camphoratus.Main conclusions Shrub encroachment might have a bigger impact on bird diversity in grassland than in open woodland, regardless of the shrub species. Species recorded in our study area were associated with historical population changes at the scale of southern Africa suggesting that shrub encroachment could be one of the main drivers of bird population dynamics in southern African savannas. If current trends continue, the persistence of several southern African bird species associated with open savanna might be jeopardized regionally.

The importance of post-fire regrowth for sable antelope in a Southern African savanna

Burning is commonly used in savannas to stimulate grass regrowth for grazing ungulates. We recorded the relative use of burns occurring at different stages in the seasonal cycle, as well as in different regions of the landscape by two herds of sable. We also recorded behavioural measures of foraging efficiency and faecal nutrient contents as an indication of nutrient gains. Sable consistently concentrated their grazing on burned areas provided there was sufficient green regrowth during the dry season. In these circumstances they grazed for longer per feeding station, showed a slower step rate while foraging, and shorter between-patch moves, and a higher probability of encountering acceptable food per step taken while foraging than on unburnt areas. In the year when only a burn with insufficient regrowth was available, sable continued to forage in the area that had been burned during the previous year. Faecal crude protein was substantially higher at the end of the dry season in the year when burned areas were utilized. Accordingly early dry season fires can be important in helping sable bridge the nutritional limitations posed by the dry season, provided sufficient soil moisture remains to promote adequate grass regrowth. Resume Les feux sont frequemment utilises dans les savanes pour stimuler la repousse des herbes consommees par les ongules herbivores. Nous avons enregistre l’utilisation relative des aires brulees a differents stades du cycle saisonnier, et dans differentes regions du paysage, par deux hardes d’antilopes sable. Nous avons aussi enregistre des mesures comportementales de l’efficacite du nourrissage et le contenu en nutriments des matieres fecales pour avoir une indication du gain en nutriments. Les antilopes sable concentraient leur pâturage de facon continue sur les aires brulees pour autant que la repousse de verdure fut suffisante au cours de la saison seche. Dans ce cas-la, elles paissaient plus longtemps par station de nourrissage, presentaient un nombre de pas plus faible en mangeant et se deplacaient moins entre les arrets, et la probabilite qu’elles rencontrent une nourriture acceptable par pas franchi en cherchant leur nourriture etait plus grande que dans les zones non brulees. L’annee ou ne fut disponible qu’une zone brulee ou la repousse etait insuffisante, les antilopes ont continuea brouter dans la zone qui avait ete brulee l’annee precedente. Le taux de proteines brutes des matieres fecales etait substantiellement plus haut a la fin de la saison seche l’annee ou les zones brulees ont ete utilisees. En fonction de cela, on peut dire que les feux du debut de la saison seche peuvent etre importants pour aider les antilopes sable a passer le cap des limites nutritionnelles imposees par la saison seche, a condition qu’il reste suffisamment d’humidite dans le sol pour favoriser une repousse d’herbes adequate.

Variability and recent trends in the African terrestrial carbon balance

Abstract. We modeled the African terrestrial carbon balance over the past century using a spatially resolved process based vegetation model (ORCHIDEE). The model is forced by changing climate and by human-induced changes in land use. It includes a simple parameterization of natural fires, but the natural vegetation dynamics was ignored. The period analyzed is 1901–2002. Overall, we found that the African net terrestrial carbon balance (Net Biome Productivity, NBP) increased from a net CO2 source to the atmosphere of 0.14 Pg C yr−1 in the 1980s to a net sink of 0.15 Pg C yr−1 in the 1990s. The land use flux alone is estimated to be a source of 0.13 Pg C yr−1 caused by deforestation. This implies that climatic trends (mainly increasing precipitation) and CO2 increase (fertilization effect), are causing a sink of 0.28 Pg C yr−1 which offsets the land-use source. We found that the interannual variability of NBP is large, and mostly driven by photosynthesis variability. Over savannas, photosynthesis changes from one year to the next are strongly correlated with rainfall changes (R2=0.77 in northern Africa, and R2=0.42 in southern African savannas). Over forests, such a control by rainfall is not found. The main spatial pattern of interannual variability in NBP and photosynthesis/ecosystem respiration fluxes is related with ENSO, with dryer conditions prevailing over savannas during El Niño and wetter conditions over forests. Climate induced variations in fire emissions respond to this ENSO forcing, but do not determine strongly the NBP interannual variability. Finally, we model that ecosystem respiration variations (mostly autotrophic respiration) are correlated with those of photosynthesis, on interannual as well as on decadal time scales, but this result is uncertain given the potential for acclimation for autotrophic respiration processes.

Predictive fire occurrence modelling to improve burned area estimation at a regional scale: A case study in East Caprivi, Namibia

Fires threaten human lives, property and natural resources in Southern African savannas. Due to warming climate, fire occurrence may increase and fires become more intense. It is crucial, therefore, to understand the complexity of spatiotemporal and probabilistic characteristics of fires. This study scrutinizes spatiotemporal characteristics of fires and the role played by abiotic, biotic and anthropogenic factors for fire probability modelling in a semiarid Southern African savanna environment. The MODIS fire products: fire hot spots (MOD14A2 and MYD14A2) and burned area product MODIS (MCD45A1), and GIS derived data were used in analysis. Fire hot spots occurrence was first analysed, and spatial autocorrelation for fires investigated, using Moran's I correlograms. Fire probability models were created using generalized linear models (GLMs). Separate models were produced for abiotic, biotic, anthropogenic and combined factors and an autocovariate variable was tested for model improvement. The hierarchical partitioning method was used to determine independent effects of explanatory variables. The discriminating ability of models was evaluated using area under the curve (AUC) from the receiver operating characteristic (ROC) plot. The results showed that 19.2–24.4% of East Caprivi burned when detected using MODIS hot spots fire data and these fires were strongly spatially autocorrelated. Therefore, the autocovariate variable significantly improved fire probability models when added to them. For autologistic models, i.e. models accounting for spatial autocorrelation, discrimination was good to excellent (AUC 0.858–0.942). For models not counting spatial autocorrelation, prediction success was poor to moderate (AUC 0.542–0.745). The results of this study clearly showed that spatial autocorrelation has to be taken in to account in the fire probability model building process when using remotely sensed and GIS derived data. This study also showed that fire probability models accounting for spatial autocorrelation proved to be superior in regional scale burned area estimation when compared with MODIS burned area product (MCD45A1).

Precipitation–NDVI relationships in eastern and southern African savannas vary along a precipitation gradient

Precipitation is a major determinant of vegetation production at the regional scale, especially in tropical areas. Recent works, however, have shown that this relationship is weak at the interannual temporal scale, particularly in mesic ecosystems, where limitations other than water constrain vegetation production. We investigated whether this holds true for eastern and southern African savannas, by studying the relationship between precipitation and the normalized difference vegetation index (NDVI) at the interannual time‐scale in 33 protected areas. We also used extreme precipitation and NDVI events to reveal the overlooked influence of precipitation along the mean annual precipitation (MAP) gradient. Only the semi‐arid ecosystems showed significant precipitation–NDVI relationships. We found that maximum NDVI was associated with opposite precipitation conditions along the MAP gradient; maximum NDVI was associated with some of the largest precipitation in semi‐arid sites and with some of the lowest precipitation in mesic sites. Although untested, these results are consistent with the hypothesis of an interaction between water and nutrient limitations along the MAP gradient. Our results extend to African savannas the previous finding that ecosystem sensitivity to annual precipitation decreases with increasing MAP, and highlight that, even in mesic ecosystems, precipitation patterns condition the likelihood of reaching maximum NDVI.

Herbivore and nutrient control of lawn and bunch grass distributions in a southern African savanna

In African savannas, bottom-up soil nutritional factors and top-down herbivory have both been suggested to control the distribution patterns of bunch and lawn grasses. We tested the separate and combined roles of these factors on grassland distribution in a South African savanna by focusing on three main objectives, namely (1) are grazing patches and lawns restricted to specific soils or sites, (2) does herbivory alter rates of nutrient cycling which facilitates lawns and (3) are there any differences in foliage quality between lawn and bunch grass-dominated sites that might influence animals in choosing to feed on lawns? We set up ten sites along a grazing gradient in the Hluhluwe-iMfolozi Park in KwaZulu-Natal, South Africa. At these sites, we measured total soil nutrients, soil nutrient turnover rates and grass foliage nutrient concentrations. We found that in the Hluhluwe-iMfolozi Park the spatial distribution of lawn and bunch grassland types appears to be an animal-driven phenomenon and not dependent on specific soil properties. The short-structured and distinct species found in lawns do not appear to be restricted to nutrient-enriched patches. However, the grasses of these lawns had significantly higher nutrient concentrations, in their foliage, which might explain the high attraction to these patches by herbivores. We also did not find any animal-induced stimulation of nutrient cycling rates that are often associated with lawn grass species.

Sexual and vegetative regeneration of three leguminous tree species in South African savannas

The legume trees Dichrostachys cinerea, Acacia karroo and Acacia nilotica are encroaching a number of African savannas. I quantified the extent to which these species regenerated by seeds and root suckers respectively, and how land management affects the mode of regeneration. I used transects at three sites inside the Hluhluwe-iMfolozi Park and at three sites in adjacent villages outside the park to score all individuals shorter than 20 cm. I also excavated the roots to determine whether the plant was a gamete or a ramet connected to an established mother tree. I found D. cinerea regenerated to equal extent vegetatively and sexually, while A. karroo and A. nilotica mainly regenerated sexually. The present results suggest that the capacity of sexual regeneration of trees is important to consider in future management of bush encroachment in southern African savannas.

RAPD markers demonstrate genetic diversity in Pterocarpus angolensis from Zimbabwe and Zambia

Understanding the availability, extent and apportionment of genetic variability in natural populations of the southern African savanna tree Pterocarpus angolensis can provide reliable information on the evolutionary history and phylogenetic relationships of P. angolensis populations. This information will aid strategies for successful domestication, conservation and sustainable exploitation of the species. In this study random amplified polymorphic DNA (RAPD) analysis was employed to assess genetic diversity of P. angolensis, characterising 60 genotypes originating from four Zimbabwean and two Zambian provenances. Four RAPD primers each generated 11 to 15 scorable bands, with a total of 50 reproducible bands of which 88% were polymorphic. Analysis of molecular variance partitioned the greatest proportion of variability to within population (80.82%). Euclidean distance gave a phiST value of 0.192, indicating a moderately high level of population differentiation. The degree of population diversity found in this study, with two clear strong population structures for most genotypes from Nyazura and Gokwe and the rest of the populations with slight weak structures, suggests that all provenances tested could be beneficially utilised in a domestication programme for P. angolensis.

Soil carbon and nitrogen dynamics in southern African savannas: the effect of vegetation-induced patch-scale heterogeneities and large scale rainfall gradients

Savanna ecosystems are mixed plant communities in which trees and grasses co-exist thereby providing a heterogeneous landscape with a mosaic of tree-dominated and grass-dominated soil patches. Despite the important role that nutrient availability plays in these systems, detailed knowledge of differences in carbon and nitrogen cycling in soil patches predominantly covered by tree canopies or by grasses is still lacking. In this study, a process-based model was used to investigate the carbon and nitrogen dynamics in soil plots located in grass-dominated and tree/shrub-dominated soil patches along the Kalahari Transect (KT). The KT in southern Africa traverses a dramatic aridity gradient, across relatively homogenous soils, providing an ideal setting for global change studies. Here we show that there are distinctly different dynamics for soil moisture, decomposition and nitrogen mineralization between soil plots located under tree canopies and in open canopy areas, especially at the dryer end of the KT. Such differences diminished when approaching the wetter end of this transect. This study shows that in savanna ecosystems, water availability determines the patterns and rates of nutrient cycling at large scales, while at the local scales, vegetation patchiness plays an important role in nutrient cycling. Savannas are relatively stable ecosystems, resilient to small rainfall modifications, although irreversible changes may occur with stronger shifts in climate conditions.

Spatial heterogeneity and sources of soil carbon in southern African savannas

Knowledge of the southern Africa soil carbon pool, its heterogeneity, sources (from trees or grasses), and potential response to climate is extremely limited. In this study the Kalahari Transect (KT) was used as a representative savanna ecosystem to quantitatively evaluate the spatial heterogeneity of the soil carbon pool and its contributing sources. The KT encompasses a dramatic aridity gradient on relatively homogenous soils. Two sites were chosen along the KT, representing dry and wet conditions. In February–March 2005, soil samples were collected at each site along a 300 m transect. Stable carbon isotope (δ 13C) and organic carbon content (%C) of the soils were utilized in the assessment in conjunction with geostatistical analysis of the spatial patterns of soil δ 13C and %C. At the dry savanna site, well-defined patterns in both δ 13C and %C were observed that were related to the distribution of woody vegetation. At the wet savanna site, the spatial patterns of δ 13C and %C were somewhat less pronounced, but still were impacted by the distribution of woody vegetation. The relative contributions from C 3 and C 4 vegetation to the soil carbon pool at the wet site were independent of tree locations, but dependent on woody plant locations at the dry site. At the dry site, ~ 40% of the soil carbon was derived from C 3 vegetation, whereas at the wet site ~ 90% of the soil carbon originated from C 3 vegetation. These results represent an important step in understanding the impact of regional climate change (e.g., rainfall variations) on carbon sequestration in southern Africa by providing quantitative information on soil carbon spatial distributions and sources under different climatic conditions (e.g., different rainfall regimes).

The sustainability of southern African savannas

The sustainability of southern African savannas

Patterns and implications of plant-soil δ13C and δ15N values in African savanna ecosystems

Southern African savannas are mixed plant communities where C 3 trees co-exist with C 4 grasses. Here foliar δ 15N and δ 13C were used as indicators of nitrogen uptake and of water use efficiency to investigate the effect of the rainfall regime on the use of nitrogen and water by herbaceous and woody plants in both dry and wet seasons. Foliar δ 15N increased as aridity rose for both C 3 and C 4 plants for both seasons, although the magnitude of the increase was different for C 3 and C 4 plants and for two seasons. Soil δ 15N also significantly increased with aridity. Foliar δ 13C increased with aridity for C 3 plants in the wet season but not in the dry season, whereas in C 4 plants the relationship was more complex and non-linear. The consistently higher foliar δ 15N for C 3 plants suggests that C 4 plants may be a superior competitor for nitrogen. The different foliar δ 13C relationships with rainfall may indicate that the C 3 plants have an advantage when competing for water resources. The differences in water and nitrogen use likely collectively contribute to the tree–grass coexistence in savannas. Such differences facilitate interpretations of palaeo-vegetation composition variations and help predictions of vegetation composition changes under future climatic scenarios.

What controls woodland regeneration after elephants have killed the big trees?

Summary Top‐down regulation of ecosystems by large herbivores is a topic of active debate between scientists and managers, and a prime example is the interaction between elephants Loxodonta africana and trees in African savannas. A common assumption among wildlife managers is that a local reduction in elephant numbers will ultimately allow woodland to self‐restore to a desired former state. Such regeneration is, however, dependent on the survival of seedlings of impacted tree species. We conducted a field experiment to investigate seedling predation in the elephant‐transformed Chobe riparian woodland of northern Botswana. We planted seedling gardens in (i) complete exclosures that excluded all herbivores except small rodents and invertebrates, (ii) semi‐permeable exclosures that excluded ungulates but included primates, lagomorphs, all rodents, gallinaceous birds, etc, and (iii) completely open plots. Seedlings were of two tree species decreasing in the area (Faidherbia albida and Garcinia livingstonei) and two that are increasing (Combretum mossambicense and Croton megalobotrys). After 9 months, seedling survival ranged from >75% for all species in the complete exclosure to <20% for Faidherbia albida in the open plots. Survival of all seedlings except C. megalobotrys declined precipitously in open plots during the dry season when invertebrates are largely dormant but when impalas Aepyceros melampus (locally abundant ungulates) increase the browse components of their diets. Seedling survival in the open plots was negatively related to local impala density but unrelated to that of any other browser. Synthesis and applications. Our findings relate to the current debate about managing elephants to restore southern African savanna landscapes to desired historical states. Various seedling predators, including the ubiquitous impala Aepyceros melampus, regulate the regeneration of trees from seedlings, and our experiments support the hypothesis that tall closed‐canopy woodlands originate during episodic windows of opportunity for seedling survival. To artificially recreate such a window would require the decimation of seedling predators as well as elephants, which is impractical at the landscape scale.

The effects of land tenure change on sustainability: human security and environmental change in southern African savannas

Abstract Using a human security perspective, we investigate how the history of land tenure changes has driven sustainability in southern African savannas. The paper examines four countries—Botswana, Mozambique, South Africa and Zimbabwe. Two research questions are addressed: first, how has colonial land tenure and distribution affected people's options and capacity to end, mitigate or adapt to risks to their human, environmental and social rights, and what have been the related effects on land uses and degradation? Second, to what extent have post-independence land reforms addressed threats to human security and political causes of land degradation? The inequitable distribution of land and colonial legacy of dual or pluralistic systems of tenure are found to be at the root of many agrarian and environmental problems. Post-independence land reforms have largely failed to address these fundamental issues, sometimes even reinforcing threats to social, economic and environmental sustainability. The skewed distribution of land and resources, insecure rights, and the marginalisation and restriction of savanna livelihood systems have persisted, undermining human security and environmental integrity in the region as well as leading to mounting conflict and insecurity.

The dynamic context of southern African savannas: investigating emerging threats and opportunities to sustainability

Abstract The Southern African Savannas Network and the Southern African Vulnerability Initiative used study areas and information from a wide range of sources to assess threats and opportunities to savanna sustainability from a natural and social science perspective, respectively. This paper describes an attempt to synthesise their findings using structural analysis. Key driving processes, main dependent processes, and processes that are inherent to most interactions by having both high driving power and high dependency, were identified. Qualitative examination reveals that while many of the changes within savannas reinforce each other, many of the key driving factors are generated externally, such as economic globalisation and climate change. In order to take advantage of opportunities and address challenges within such a complex and dynamic system, future interventions must address the different elements of savannas in a coherent and multi-sectoral manner. Within the large social, economic and environmental changes taking place over the past decade, changes in institutions and technologies and in indigenous fauna and flora continue to play a central role both in driving, and being dependent on, other processes and must continue to be a key focus of any coherent savanna policy.

Satellite prediction of soil δ13C distributions in a southern African savanna

Stable carbon isotopes have been frequently used to indicate carbon pools and processes in soils, plants, and the atmosphere. Carbon isotope compositions are particularly useful in partitioning soil carbon sources between C 3 and C 4 vegetation because of the distinct δ 13C distributions for C 3 and C 4 vegetation. Remote sensing is a powerful tool used to identify ecosystem patterns and processes at larger scales. A union of these two approaches would hold promise for spatially continuous estimates of carbon isotope compositions. In the current study, a framework is presented for using high spatial resolution remote sensing to predict soil δ 13C distributions across a southern Africa savanna ecosystem. The results suggest that if the vegetation–soil δ 13C relationship can be established, soil δ 13C distributions can be estimated by high-resolution satellite images (e.g., IKONOS, Quickbird). Despite limitations remote sensing is a promising tool to expand estimates of terrestrial δ 13C spatial patterns and dynamics.

Interacting effects of grass height and herbivores on the establishment of an encroaching savanna shrub

Shrub encroachment is a widely observed problem in Southern African savannas. Although the effects of herbivory and grass height on woody species recruitment have been studied individually, little information exists about how these factors interact. In this study seeds and seedlings of the encroaching shrub Dichrostachys cinerea were planted in clipped and unclipped grass plots, with and without large herbivores present. Seed germination, seedling survival and seedling predation were monitored for 8 months. Germination started earlier in plots where herbivores were excluded. Overall, the earlier the seeds germinated, the longer the seedlings survived. Clipping positively affected the number of germinated seeds, seedling growth and survival but effects varied among herbivore exclusion treatments and sites. Invertebrates caused the majority of the seedling damage. We conclude the recruitment of D. cinerea is influenced by the interplay of grass height and herbivory. In this study, the presence of large herbivores early in the wet season, and the absence of simulated grazing later on, affected the regeneration of D. cinerea negatively. However, differences in effects among sites suggest that the mechanisms found here may work differently in other habitats.