An idealized model for tree–grass coexistence in savannas: the role of life stage structure and fire disturbances

1.  We discuss a simple implicit-space model for the competition of trees and grasses in an idealized savanna environment. The model represents patch occupancy dynamics within the habitat and introduces life stage structure in the tree population, namely adults and seedlings. A tree can be out-competed by grasses only as long as it is a seedling.
2.  The model is able to predict grassland, forest, savanna and bistability between forest and grassland, depending on the different characteristics of the ecosystem, represented by the model's parameters.
3.  The inclusion of stochastic fire disturbances significantly widens the parameter range where coexistence of trees and grasses is found. At the same time, grass-fire feedback can induce bistability between forest and grassland.
4.   Synthesis . These results suggest that tree–grass coexistence in savannas can be either deterministically stable or stabilized by random disturbances, depending on prevailing environmental conditions and on the types of plant species present in the ecosystem.     

Herbivore–vegetation feedbacks can expand the range of savanna persistence: insights from a simple theoretical model

Theoretical models of tree–grass coexistence in savannas have focused primarily on the role of resource availability and fire. It is clear that herbivores heavily impact vegetation structure in many savannas, but their role in driving tree–grass coexistence and the stability of the savanna state has received less attention. Theoretical models of tree–grass dynamics tend to treat herbivory as a constant rather than a dynamic variable, yet herbivores respond dynamically to changes in vegetation structure in addition to modifying it. In particular, many savannas host two distinct herbivore guilds, grazers and browsers, both of which have the potential to exert profound effects on tree/grass balance. For example, grazers may indirectly favor tree recruitment by suppressing the destructive effects of fire, and browsers may facilitate the expansion of grassland by reducing the competitive dominance of trees. We use a simple theoretical model to explore the role of grazer and browser dynamics on savanna vegetation structure and stability across fire and resource availability gradients. Our model suggests that herbivores may expand the range of conditions under which trees and grasses are able to stably coexist, as well as having positive reciprocal effects on their own niche spaces. In addition, we suggest that given reasonable assumptions, indirect mutualisms can arise in savannas between functional groups of herbivores because of the interplay of consumption and ecosystem feedbacks.     

The Enemy of My Enemy Hypothesis: Why Coexisting with Grasses May Be an Adaptive Strategy for Savanna Trees

Savannas are characterized by the coexistence of trees and flammable grasses. Yet, tree–grass coexistence has been labeled as paradoxical—how do these two functional groups coexist over such an extensive area, despite being generally predisposed to excluding each other? For instance, many trees develop dense canopies that limit grass growth, and many grasses facilitate frequent/intense fires, increasing tree mortality. This study revisits tree–grass coexistence with a model of hierarchical competition between pyrogenic grasses, “forest trees” adapted to closed-canopy competition, and “savanna trees” that are inferior competitors in closed-canopy communities, but more resistant to fire. The assumptions of this model are supported by empirical observations, including a systematic review of savanna and forest tree community composition reported here. In general, the model simulations show that when savanna trees exert weaker competitive effects on grasses, a self-reinforcing grass community is maintained, which limits forest tree expansion while still allowing savanna trees to persist (albeit as a subdominant to grasses). When savanna trees exert strong competitive effects on grasses, savanna trees cover increases initially, but as grasses decline their inhibitory effect on forest trees weakens, allowing forest trees to expand and exclude grasses and savanna trees. Rather than paradoxical, these results suggest that having weaker competitive effects on grasses may be advantageous for savanna trees, leading to greater long-term abundance and stability. We label this the “enemy of my enemy hypothesis,” which might apply to species coexistence in communities defined by hierarchical competition or with species capable of generating strong ecological feedbacks.     

The Impact of Horning by Wildebeest on Woody Vegetation of the Serengeti Ecosystem

Abstract Horning vegetation, an expression of aggression predominately among adult males, may be universal among horned ungulates. We found that horning by wildebeest (Connochaetes taurinus) males had an important impact on the Serengeti ecosystem, Africa, from the 1960s to the 1980s, as the wildebeest population increased from 0.25 million to 1.5 million. Between 1979 and 2003, we sampled 2,626 trees and bushes to assess horning impacts. In the 1986 survey, 57% (n = 1,416) of trees and bushes had suffered moderate to severe horning injury. Severe damage frequency was highest (68%) in open grassland, where a few trees were exposed to many wildebeests, and lowest (24%) inside savanna woodland where wildebeest rarely go. Horning by 300,000–400,000 adult male wildebeest contributed to converting savanna woodland into tree savanna and open grassland. Horning by wildebeest, in combination with known impacts such as grazing, manuring, and trampling, may result in ecological impacts to Serengeti ecosystems only exceeded by the elephant (Loxodonta africana) and fire. More research is needed to understand the ecological and management implications of horning.     

Notes on habitat relationships of ungulates in Yankari Game Reserve, Nigeria

Yankari Game Reserve in northeastern Nigeria consists largely of savanna woodland with trees on the better soils growing to 15 m and with spreading crowns. On shallow and stony soils the tree height is generally less and the canopy is discontinuous. The Gaji River riparian zone supports a wide variety of vegetation types ranging from evergreen, closed canopy forest to sedge meadows and patches of open grassland.
Elephant ( Loxodonta africana ) range backwards and forwards along the riparian strip, feeding on perennial grasses and a variety of browse material and utilizing closed canopy forest patches for shade cover. The major movement patterns of other important herbivore species are perpendicular to the riparian strip. Areas used intensively are: waterbuck ( Kobus defassa )–open savanna woodland immediately behind the riparian strip: Western hartebeest ( Alcelaphus buselaphus major )– open grassy habitat in relatively poor woodland at middle distances from the river; Roan antelope ( Hippotragus equinus )–patches of well-developed and infrequently burned woodland, often at major distances from the river. Buffalo ( Syncerus caffer brachyceros ) during the dry season ranged between the riparian grassland areas and the more open sections of nearby savanna woodland, but travelled out to distant sections of the reserve after rainwater pools had formed.
A major problem in management was the development of a burning policy that would maintain an appropriate balance between perennial and annual grasses and the shade providing trees.     

Tree–grass coexistence in savannas revisited – insights from an examination of assumptions and mechanisms invoked in existing models

Several explanations for the persistence of tree–grass mixtures in savannas have been advanced thus far. In general, these either concentrate on competition‐based mechanisms, where niche separation with respect to limiting resources such as water lead to tree–grass coexistence, or demographic mechanisms, where factors such as fire, herbivory and rainfall variability promote tree–grass persistence through their dissimilar effects on different life‐history stages of trees. Tests of these models have been largely site‐specific, and although different models find support in empirical data from some savanna sites, enough dissenting evidence exists from others to question their validity as general mechanisms of tree–grass coexistence. This lack of consensus on determinants of savanna structure and function arises because different models: (i) focus on different demographic stages of trees, (ii) focus on different limiting factors of tree establishment, and (iii) emphasize different subsets of the potential interactions between trees and grasses. Furthermore, models differ in terms of the most basic assumptions as to whether trees or grasses are the better competitors. We believe an integration of competition‐based and demographic approaches is required if a comprehensive model that explains both coexistence and the relative productivity of the tree and grass components across the diverse savannas of the world is to emerge. As a first step towards this end, we outline a conceptual framework that integrates existing approaches and applies them explicitly to different life‐history stage of trees.     

Is the lack of leguminous savanna trees in grasslands of South Africa related to nutritional constraints?

As with many grasslands globally, the Highveld grasslands of South Africa are tree-less, despite having a climate that can support tree growth. Models predict that fire maintains these grasslands. The question arises as to why fire-tolerant savanna trees do not survive in these ecosystems? Savanna tree survival in mesic areas is restricted by demographic bottlenecks, specifically limitations to sapling-escape from fire. It was hypothesised that ancient highly leached soils from grassland areas would prevent saplings from growing fast enough to escape the fire-trap. Growth rates of savanna tree seedlings (Acacia karroo Hayne and Acacia sieberiana Burtt Davy) were measured in a common garden experiment using soils from ten sites collected along a savanna-grassland continuum. Soils from grassland sites were relatively nutrient-poor compared to those from savannas with lower pH, and associated cations. A. sieberiana growth rates responded to pH and these nutrients, whereas A. karroo growth was less strongly linked to specific nutrients. Even so, both species accumulated more biomass when grown in soils from savanna sites compared to grassland sites. An exception was a low elevation low nutrient savanna site that resulted in poor growth, yet sustains high tree biomass in situ. Differences between growth in grassland and savanna soils were small. They may contribute to, but are unlikely to explain, the treeless nature of these grasslands.     

A lepidopteran (<Emphasis Type="Italic">Imbrasia belina</Emphasis>) might influence tree-grass balance of <Emphasis Type="Italic">Colophospermum mopane</Emphasis> savanna

Traditional explanations of tree-grass coexistence in African savannas are based on competition between these growth forms or demographic bottlenecks of trees maintained by fire or mammalian browsers. Perturbation of their “balance” may result in an alternate system state of woody encroachment. Invertebrate herbivory has never been offered as an explanation. We developed a consumer-resource model which illustrated that annual irruptions of a lepidopteran (Imbrasia belina), known as mopane worm, can determine the tree-grass balance of semi-arid Colophospermum mopane savanna in southern Africa. Model performance was sensitive to the abundance, hence mortality, of mopane worms, owing to their complete defoliation of tree leaf biomass resulting in altered competitive relations between trees and grasses. Invertebrate herbivores have been recognized in other systems as agents for effecting a state change of host tree populations; this modeling study offers a first indication of such a role for the well-researched tree-grass relations of African savannas.     

Structure, floristic composition, and vegetation forming factors of three vegetation types in Senegal

Six 1 ha plots were established in a coastal savanna, called Fathala Forest, in Delta du Saloum National Park, Senegal. Two plots were placed in woodland, two in wooded grassland, and two in transition woodland in order to describe structure and floristic composition of the vegetation. All trees ≥ 5 cm dbh were sampled. The three selected vegetation types showed distinct differences in structure as well as in species composition. Woodland had high density (440–449 individuals per ha), many small trees, and high basal area (13.4 m²per ha). Transition woodland was characterised by low density (54–118 individuals per ha) but many large trees and a relatively large basal area (8.6–12.8 m² per ha). Wooded grassland was characterised by medium sized trees, it had low density (86–102 individuals per ha) and low basal area (3.8–5.7 m² per ha). Species richness ranged between 17 and 27 species per ha in the six plots. Only two species were found in all plots, Daniellia oliveri (Caesalpiniaceae) and Prosopis africana (Mimosaceae). Legumes dominated all plots. Wooded grassland and transition woodland had many characteristics of fire-affected vegetation in contrast to woodland. Today wooded grassland encroaches on woodland and transition woodland. Management of the latter two vegetation types should be given priority as they maintain structural and floristic characteristics that are essential to conserve biodiversity and original features of the vegetation, and they are also important for local people who are allowed to make sustainable use of the vegetation.     

Tree growth in an African woodland savanna affected by disturbance

Questions: How does tree growth in a tropical woodland savanna vary as a function of size, and how is it affected by competition from neighbours, site attributes, and damage caused by disturbance? Location: western Zimbabwe. Methods: Trees of common species were tagged, mapped, and measured annually between 2001 and 2003 in a Kalahari sand woodland savanna. Diameter increments were analysed with mixed model regressions for the largest ramet in each genet. Stem diameter and damage, soil texture, and indices of competition at multiple spatial scales were used as covariates. Results: Stem diameter increased initially and then declined as a function of size in undamaged trees, which grew faster than damaged trees. Growth in damaged trees declined with size. No site differences were detected, and there was evidence for between‐tree competition on growth only in the fastest‐growing species, Brachystegia spiciformis. In several species the growth rate of the largest ramet increased as a function of the basal area of secondary ramets, contrary to expectations. For many species, the growth models showed poor explanatory power. Conclusions: Growth in Kalahari sand savanna trees varies as a function of size and changes in tree architecture caused by disturbance agents such as fire, frost, and elephant browsing. Disturbance may thus play an important role on vegetation dynamics through its effects on growth in the post‐disturbance phase. Growth is highly stochastic for some species in this system, and more deterministic in others. It is hypothesized that this dichotomy may be driven by differences in rooting depth among species.     

Spatial pattern of dry rainforest colonizing unburnt Eucalyptus savanna

Abstract The spatial pattern of dry rainforest and savanna tree species was analysed in a 1.56‐ha plot within an unburnt eucalypt savanna woodland in north Queensland, Australia. Rainforest colonization constituted only 1.3% of the basal area and mostly consisted of individuals less than 3 m high. The distribution of rainforest trees was highly clumped around the large savanna eucalypt trees. Ecological mechanisms generating the clumped distribution are discussed in light of evidence from this study and the literature. Herbaceous biomass was not reduced under trees, suggesting that relief from grass competition has not favoured rainforest colonization under tree crowns. Edaphic facilitation through nutrient enrichment under savanna tree crowns appears to be only minor on the moderate fertility soils of the area. The highly clumped pattern of colonizing dry rainforest may be a consequence of seeds dropped from birds roosting in savanna trees.     

Root dynamics influence tree–grass coexistence in an Australian savanna

Both resource and disturbance controls have been invoked to explain tree persistence among grasses in savannas. Here we determine the extent to which competition for available resources restricts the rooting depth of both grasses and trees, and how this may influence nutrient cycling under an infrequently burned savanna near Darwin, Australia. We sampled fine roots <2 mm in diameter from 24 soil pits under perennial as well as annual grasses and three levels of canopy cover. The relative proportion of C₃ (trees) and C₄ (grasses) derived carbon in a sample was determined using mass balance calculations. Our results show that regardless of the type of grass both tree and grass roots are concentrated in the top 20 cm of the soil. While trees have greater root production and contribute more fine root biomass grass roots contribute a disproportional amount of nitrogen and carbon to the soil relative to total root biomass. We postulate that grasses maintain soil nutrient pools and provide biomass for regular fires that prevent forest trees from establishing while savanna trees, are important for increasing soil N content, cycling and mineralization rates. We put forward our ideas as a hypothesis of resource‐regulated tree–grass coexistence in tropical savannas.     

Tree–grass coexistence in the Brazilian cerrado: demographic consequences of environmental instability

Aim Tropical savanna ecosystems are uniquely characterized by the co‐dominance of both trees and grasses. An operational understanding of the ecological processes involved in maintaining this condition is essential for understanding both the functioning of savanna systems as well as their potential response to environmental change. A simple model is presented to explore the potential for a demographic mechanism of long‐term tree persistence and temporal physiognomic stability in the Brazilian cerrado. Location The model is developed based on data from the humid cerrado of Brazil. Methods In contrast to many existing models of tree–grass dynamics a model is presented which is based on data from the humid cerrado of Brazil, which is both qualitatively and quantitatively different from many of the more arid savannas of the palaeotropics. The model focuses on the dynamics of a synthetic tree population, with particular attention given to reproduction, seedling establishment and fire effects; with separate sub‐models for grass production, fire and rainfall. Results The model successfully predicts coexistence across the full range of observed vegetation physiognomies, but only under limited conditions. Under coexistence conditions, the dynamics of the tree population are characterized by long periods of gradual decline, punctuated by occasional bursts of growth. However, in agreement with earlier studies, the model consistently over‐predicts domination by the tree component. Fire is identified as an overriding factor in determining model behaviour, and the response of reproduction and sapling recruitment to variance in the frequency of fire ignition is identified to be of potential importance in the functioning of the Brazilian cerrado. The key dynamics of the model which promote tree–grass coexistence are consistent with a number of established determinants of ecological resilience in savanna systems. Main conclusions The model identifies the importance of the effective exploitation of rare opportunities for favourable recruitment (e.g. exclusion from fire) by the tree population, in promoting coexistence within a predominantly adverse environment. Support is provided for an alternative demographic mechanism of tree–grass coexistence in the cerrado (the storage effect), which is not based on the limiting assumption of niche partitioning through differences in rooting depth. The results are consistent with those presented by recent modelling work based on the more arid savannas of southern Africa. The model presented here differs in the emphasis given to particular environmental and life‐history attributes which are critical in determining the tree–grass balance, but provides further general support for the potential role of demographic mechanisms (such as the storage effect) in determining the structure of tropical savannas. Despite having clear limitations, models can serve as valuable heuristic tools to aid the integration and exploration of existing data sets as well as our present understanding of key ecological processes.     

The invasion of Lantana camara L. in Forty Mile Scrub National Park,north Queensland

Abstract Seventy-three per cent of dry rainforest in Forty Mile Scrub National Park and large areas in adjacent savanna woodland have more than 5000 individuals per ha of lantana (Lantana camara L.). Transect studies in dry rainforest and savanna woodland across varying intensities of lantana infestation show a negative correlation between the density of lantana and tree cover in rainforest. The density of pig rooting is very high in areas of the dry rainforest on deep soil that was not heavily infested with lantana. It is suggested that the digging activities of these animals may cause tree death and subsequent increased light penetration, which favours lantana. The species richness of the dry rainforest declines as the density of lantana increases. However, the saplings and seedlings and the soil seed bank of dry rainforest and savanna woodland tree species have comparable densities in heavy and light lantana infestations. The proliferation of lantana results in the build up of heavy fuel loads across the boundary of dry rainforest and savanna woodland. Recent fires have killed the canopy trees in a large area of dry rainforest within the Park. Active management of Forty Mile Scrub National Park is urgent and some initiatives are suggested.     

Structure and biomass along an Acacia zanzibarica woodland–savanna gradient in a former ranching area in coastal Tanzania

Questions: What factors influence the density, size and growth form of trees in secondary Acacia zanzibarica woodlands on a former humid savanna rangeland? How does tree density relate to variation in tree foliage and spines, and woody and grass biomass? Location: Tropical coastal Tanzania (former Mkwaja Ranch, now in Saadani National Park). Methods: We surveyed 97 circular plots (4‐m radius) representing a gradient from open savanna to dense woodland. Within each plot, we measured all trees and estimated the biomass of spines. Foliage biomass of tree and grass layers was estimated on three occasions, twice during the wet season and once in the dry season. Soil samples were taken from each plot and analysed for texture and nutrient content. Interrelationships among various variables were investigated using linear multiple regression and mixed effects models. Results: Tree densities were highest on more nutrient‐rich, heavy soils. Spinescence was highest on trees in open savanna. Biomass of tree foliage in the wet season was best explained by numbers of ant nests and tree live‐wood ratio. Foliage biomass in the dry season was less than half that in the wet season and best predicted by grass biomass. Variables related to biomass of the grass layer were strongly influenced by fire; living grass biomass also decreased with increasing tree density. Conclusions: A. zanzibarica is a tree with a high water demand, and the association with heavy soils is probably due to greater availability of water on these sites. Establishment of A. zanzibarica woodlands significantly reduced grazing resources at Mkwaja Ranch. Under post‐ranching conditions, however, fires and soil conditions predominate. The woodlands may, therefore, represent a transient state of woody density in a still resilient humid savanna.     

Seasonal leaf dynamics across a tree density gradient in a Brazilian savanna

Interactions between trees and grasses that influence leaf area index (LAI) have important consequences for savanna ecosystem processes through their controls on water, carbon, and energy fluxes as well as fire regimes. We measured LAI, of the groundlayer (herbaceous and woody plants <1-m tall) and shrub and tree layer (woody plants >1-m tall), in the Brazilian cerrado over a range of tree densities from open shrub savanna to closed woodland through the annual cycle. During the dry season, soil water potential was strongly and positively correlated with grass LAI, and less strongly with tree and shrub LAI. By the end of the dry season, LAI of grasses, groundlayer dicots and trees declined to 28, 60, and 68% of mean wet-season values, respectively. We compared the data to remotely sensed vegetation indices, finding that field measurements were more strongly correlated to the enhanced vegetation index (EVI, r ²=0.71) than to the normalized difference vegetation index (NDVI, r ²=0.49). Although the latter has been more widely used in quantifying leaf dynamics of tropical savannas, EVI appears better suited for this purpose. Our ground-based measurements demonstrate that groundlayer LAI declines with increasing tree density across sites, with savanna grasses being excluded at a tree LAI of approximately 3.3. LAI averaged 4.2 in nearby gallery (riparian) forest, so savanna grasses were absent, thereby greatly reducing fire risk and permitting survival of fire-sensitive forest tree species. Although edaphic conditions may partly explain the larger tree LAI of forests, relative to savanna, biological differences between savanna and forest tree species play an important role. Overall, forest tree species had 48% greater LAI than congeneric savanna trees under similar growing conditions. Savanna and forest species play distinct roles in the structure and dynamics of savanna–forest boundaries, contributing to the differences in fire regimes, microclimate, and nutrient cycling between savanna and forest ecosystems.     

黄土丘陵沟壑区土壤水分环境及植被生长响应——以燕沟流域为例

调查了黄土丘陵沟壑区燕沟流域刺槐(Robinia pseudoacacia)林地、辽东栎(Quercus liaotungensis)林地、荒草地、农地等不同植被类型条件下7种地类的土壤水分环境,分析不同植被类型对水分环境的生长响应.结果认为,各地类均存在一定程度的水分亏缺,亏缺量由大到小依次为:阳坡刺槐林地991.57mm、阳坡荒草地941.21mm、阴坡刺槐林地866.53mm、阳坡辽东栎林地815.89mm、阴坡荒草地790.27mm、阴坡辽东栎林地745.20mm、农地325.55mm.土壤水分的交换深度农地达320cm,阴坡荒草地为240cm,阴坡辽东栎林地为200cm,阴坡刺槐林地和阳坡辽东栎林地均为160cm,阳坡荒草地为140cm,阳坡刺槐林地为120cm.试验期间,林地、荒草地和农地分别约有10%、14%、30%的降水储存于土壤中,林地、荒草地600cm深土壤水库可利用水量62.6～309.0mm,与农地728.6mm相比土壤水库的调节能力很有限.受林木耗水量和土壤供水能力的双重影响,阳坡刺槐林枯梢现象严重,有整株枯死林木;阴坡刺槐林有明显的枯梢,但没有整株枯死的林木;辽东栎林也存在枯梢现象,但较刺槐林轻微,林木生长仍然十分旺盛.人工林地植被较高的截留和蒸腾耗水是造成土壤干燥化的主要原因,在植被建设中应遵循区域植被的演替规律,以水定植,尽量选择低耗水的适生乡土树种,采取自然修复为主、人工栽植为辅的措施,同时实施好水土保持措施.黄土丘陵区天然辽东栎林是当地植被演替的顶级群落,林地土壤的干燥化是黄土高原气候整体趋于旱化造成的,并不是人为干扰导致植被过度耗水造成的,这种土壤干燥化不宜归属于干层的范畴.判别土壤干层应以当地稳定天然植被群落的生物量水平和土壤水分状况为基准.     

Resilience and Thresholds in Savannas: Nitrogen and Fire as Drivers and Responders of Vegetation Transition

Resilience theory suggests that ecosystems can persist for long periods, before changing rapidly to a new vegetation phase. Transition between phases occurs when ecological thresholds have been crossed, and is followed by a reorganization of biotic and environmental interactions, leading to the emergence of a new vegetation phase or quasi-stable state. Savannas are dynamic, complex systems in which fire, herbivory, water and nutrient availability interact to determine tree abundance. Phase and transition has been observed in savannas, but the role of these different possible drivers is not always clear. In this study, our objectives were to identify phase and transition in the fossil pollen record, and then to explore the role of nitrogen and fire in these transitions using δ¹⁵N isotopes and charcoal abundance. We present palaeoenvironmental data from the Kruger National Park, South Africa, which show transition between grassland and savanna phases. Our results show transition at the end of the ninth century A.D. from a nutrient- and herbivore-limited grazing lawn, in which fire was absent and C₄ grasses were the dominant and competitively superior plant form, to a water-, fire- and herbivory-limited semi-arid savanna, in which C₄ grasses and C₃ trees and shrubs co-existed. The data accord with theoretical frameworks that predict that variability in ecosystems clusters in regions of higher probability space, interspersed by rapid transitions between these phases. The data are also consistent with the idea that phase transitions involve switching between different dominant driving processes or limiting factors.     

Seasonality and facilitation drive tree establishment in a semi-arid floodplain savanna

A popular hypothesis for tree and grass coexistence in savannas is that tree seedlings are limited by competition from grasses. However, competition may be important in favourable climatic conditions when abiotic stress is low, whereas facilitation may be more important under stressful conditions. Seasonal and inter-annual fluctuations in abiotic conditions may alter the outcome of tree–grass interactions in savanna systems and contribute to coexistence. We investigated interactions between coolibah (Eucalyptus coolabah) tree seedlings and perennial C₄ grasses in semi-arid savannas in eastern Australia in contrasting seasonal conditions. In glasshouse and field experiments, we measured survival and growth of tree seedlings with different densities of C₄ grasses across seasons. In warm glasshouse conditions, where water was not limiting, competition from grasses reduced tree seedling growth but did not affect tree survival. In the field, all tree seedlings died in hot dry summer conditions irrespective of grass or shade cover, whereas in winter, facilitation from grasses significantly increased tree seedling survival by ameliorating heat stress and protecting seedlings from herbivory. We demonstrated that interactions between tree seedlings and perennial grasses vary seasonally, and timing of tree germination may determine the importance of facilitation or competition in structuring savanna vegetation because of fluctuations in abiotic stress. Our finding that trees can grow and survive in a dense C₄ grass sward contrasts with the common perception that grass competition limits woody plant recruitment in savannas.