Simplifying the savanna: the trajectory of fire‐sensitive vegetation mosaics in northern Australia

Aim Fire is a key agent in savanna systems, yet the capacity to predict fine‐grained population phenomena under variable fire regime conditions at landscape scales is a daunting challenge. Given mounting evidence for significant impacts of fire on vulnerable biodiversity elements in north Australian savannas over recent decades, we assess: (1) the trajectory of fire‐sensitive vegetation elements within a particularly biodiverse savanna mosaic based on long‐term monitoring and spatial modelling; (2) the broader implications for northern Australia; and (3) the applicability of the methodological approach to other fire‐prone settings. Location Arnhem Plateau, northern Australia. Methods We apply data from long‐term vegetation monitoring plots included within Kakadu National Park to derive statistical models describing the responses of structure and floristic attributes to 15 years of ambient (non‐experimental) fire regime treatments. For a broader 28,000 km² region, we apply significant models to spatial assessment of the effects of modern fire regimes (1995–2009) on diagnostic closed forest, savanna and shrubland heath attributes. Results Significant models included the effects of severe fires on large stems of the closed forest dominant Allosyncarpia ternata, stem densities of the widespread savanna coniferous obligate seeder Callitris intratropica, and fire frequency and related fire interval parameters on numbers of obligate seeder taxa characteristic of shrubland heaths. No significant relationships were observed between fire regime and eucalypt and non‐eucalypt adult tree components of savanna. Spatial application of significant models illustrates that more than half of the regional closed forest perimeters, savanna and shrubland habitats experienced deleterious fire regimes over the study period, except in very dissected terrain. Main conclusions While north Australia’s relatively unmodified mesic savannas may appear structurally intact and healthy, this study provides compelling evidence that fire‐sensitive vegetation elements embedded within the savanna mosaic are in decline under present‐day fire regimes. These observations have broader implications for analogous savanna mosaics across northern Australia, and support complementary findings of the contributory role of fire regimes in the demise of small mammal fauna. The methodological approach has application in other fire‐prone settings, but is reliant on significant long‐term infrastructure resourcing.     

The Fire Refuge Value of Patches of a Fire‐Sensitive Tree in Fire‐prone Savannas: Callitris intratropica in Northern Australia

Patches of fire‐sensitive vegetation often occur within fire‐prone tropical savannas, and are indicative of localized areas where fire regimes are less severe. These may act as important fire refugia for fire‐sensitive biota. The fire‐sensitive tree Callitris intratropica occurs in small patches throughout the fire‐prone northern Australian savannas, and is widely seen as an indicator of low‐severity fire regimes and of good ecosystem health. Here, we address the question: to what extent do Callitris patches act as refuges for other fire‐sensitive biota, and therefore play a broader conservation role? We contrast floral and faunal species composition between Callitris patches and surrounding eucalypt savanna, using three case studies. In the first case study, a floristic analysis of 47 Callitris patches across Western Australia's Kimberley region showed that woody species in these patches were overwhelmingly widespread, fire‐tolerant savanna taxa. No species of special conservation concern occurred disproportionately within Callitris patches. Similarly, there was no concentration of fire‐sensitive fauna or flora in five Callitris patches in the East Kimberley. Finally, there was no difference in ant species composition among 12 Callitris patches and surrounding eucalypt savannas in Kakadu National Park, Northern Territory, and there were no fire‐sensitive ant species in Callitris patches. Our three case studies from throughout the northwestern Australia provide no evidence that Callitris patches act as important refuges for fire‐sensitive flora or fauna within fire‐prone eucalypt savannas. This calls into question the notion that Callitris is a strong indicator of general ecosystem health.     

Post‐fire habitat use of the golden‐backed tree‐rat (Mesembriomys macrurus) in the northwest Kimberley,Western Australia

Fire regimes are changing throughout the world. Changed fire patterns across northern Australian savannas have been proposed as a factor contributing to recent declines of small‐ and medium‐sized mammals. Despite this, few studies have examined the mechanisms that underpin how species use habitat in fire‐affected landscapes. We determined the habitats and resources important to the declining golden‐backed tree‐rat (Mesembriomys macrurus) in landscapes partially burnt by recent intense fire. We aimed to (i) compare the relative use of rainforest and savanna habitats; (ii) examine the effect of fire history on use of savanna habitats; and (iii) identify key foraging and denning resources. Habitat selection was examined by comparing the availability of eight habitat types around real (used) and generated (available) location points. Individuals used a range of habitats, but consistently selected long unburnt rainforest in preference to recently burnt savanna (1–12 months post‐fire); however, recently burnt savanna was used in preference to long unburnt savanna. Tree‐rats foraged in Terminalia hadleyana, Planchonia rupestris, Celtis philippensis and Owenia vernicosa, tree species that are found in a variety of habitat types. Individuals used a range of den sites, including cliffs, trees, logs, scree and stags found throughout the study area. Although multiple factors may have led to the decline of Mes. macrurus across its range, these results are consistent with the idea that changes in the savanna structure as a consequence of contemporary fire patterns could also have a role. The continued persistence of Mes. macrurus in the northwest Kimberley may be supported by land management strategies that conserve fruiting and hollow‐bearing trees, and maintain the availability of fire‐sensitive vegetation types.     

Fire resilience of ant assemblages in long‐unburnt savanna of northern Australia

Abstract Tropical savannas and rainforests contrast in their flammability and the fire resilience of their associated species. While savanna species generally exhibit high resilience to burning, there is much debate about the fire resilience of forest‐associated species, and the persistence of forest patches in a flammable savanna matrix. Where fire has been excluded, savanna tends on a trajectory towards forest, with an increase in forest‐associated plants and animal species. This study tested the idea that given the high proportion of forest‐associated taxa in long‐unburnt savanna, the fauna of these areas would be expected to exhibit less resilience to fire than the fauna in frequently burnt savannas. The study investigated the immediate and short‐term effects on ant assemblages of re‐introducing fire into long‐unburnt savanna in northern Australia. The ant fauna exhibited high resistance to fires, with no significant short‐term change in mean abundance or species richness; instead, seasonality had a far stronger influence on overall ant activity. Fire caused dramatic declines in dominance of the patchily distributed forest‐associated species Oecophylla smaragdina and Papyrius sp., but had no effect on overall dominance by open savanna species of Iridomyrmex. Dominance by Iridomyrmex pallidus declined, but this was compensated for by increases in I. reburrus, while two other species of Iridomyrmex showed no change. This indicates a high level of functional redundancy among dominant species of Iridomyrmex, which universally dominate open savanna communities, but not of dominant forest‐associated species. Overall, our findings demonstrate a high degree of fire‐resilience of the long‐unburnt savanna ant fauna. Despite the occurrence of forest‐associated species, the high proportion of savanna species persisting in this habitat means that long‐unburnt savanna retains the general response characteristics of frequently burnt savanna.     

Fire‐mediated niche‐separation between two sympatric small mammal species

Fire is a key ecological process influencing the population dynamics of small mammals. Whilst shifting competitive advantage amongst small mammal species following a single fire event is well‐documented, there has been little investigation of the potential influence of fire frequency on small mammal interspecific interactions. In this study, we investigated the effect of fire frequency on the abundance of two small dasyurid mammals, Antechinus stuartii and A. flavipes, which occur sympatrically in some parts of their range. The two antechinus species are known to have different habitat preferences, so it is possible that fire regimes may promote their coexistence in areas of sympatry by altering vegetation structure. To investigate this possibility, we estimated the abundance of both species using replicate sites which differed in the number of times burnt (1–4) during the last four decades, but with identical time‐since‐fire. Proportionally, we captured greater numbers of A. stuartii in less frequently burnt sites and greater numbers of A. flavipes in more‐frequently burnt sites. Hence, fire may mediate niche‐separation between these two species. To clarify further this pattern of response to fire frequency, we investigated which structural habitat variables differed between fire frequencies, and compared antechinus abundances with structural vegetation characteristics. We found a trend for lower ground cover density under higher fire frequencies. This offers one potential explanation of the patterns of abundance that we observed. Our study provided insights into the complexities of small mammal responses to fire, and strongly suggests that fire could mediate competitive interactions between species.     

Influence of fire history on small mammal distributions: insights from a 100‐year post‐fire chronosequence

Aim Fire affects the structure and dynamics of ecosystems world‐wide, over long time periods (decades and centuries) and at large spatial scales (landscapes and regions). A pressing challenge for ecologists is to develop models that explain and predict faunal responses to fire at broad temporal and spatial scales. We used a 105‐year post‐fire chronosequence to investigate small mammal responses to fire across an extensive area of ‘tree mallee’ (i.e. vegetation characterized by small multi‐stemmed eucalypts). Location The Murray Mallee region (104,000 km²) of semi‐arid Australia. Methods First, we surveyed small mammals at 260 sites and explored the fire responses of four species using nonlinear regression models. Second, we assessed the predictive accuracy of models using cross‐validation and by testing with independent data. Third, we examined our results in relation to an influential model of animal succession, the habitat accommodation model. Results Two of four study species showed a clear response to fire history. The distribution of the Mallee Ningaui Ningaui yvonneae, a carnivorous marsupial, was strongly associated with mature vegetation characterized by its cover of hummock grass. The occurrence of breeding females was predicted to increase up to 40–105 years post‐fire, highlighting the extensive time periods over which small mammal populations may be affected by fire. Evaluation of models for N. yvonneae demonstrated that accurate predictions of species occurrence can be made from fire history and vegetation data, across large geographical areas. The introduced House Mouse Mus domesticus was the only species positively associated with recently burnt vegetation. Main conclusions Understanding the impact of fire over long time periods will benefit ecological and conservation management. In this example, tracts of long‐unburnt mallee vegetation were identified as important habitat for a fire‐sensitive native mammal. Small mammal responses to fire can be predicted accurately at broad spatial scales; however, a conceptual model of post‐fire change in community structure developed in temperate Australia is not, on its own, sufficient for small mammals in semi‐arid systems.     

Herbaceous vegetation responses to experimental fire in savannas and forests depend on biome and climate

Fire–vegetation feedbacks potentially maintain global savanna and forest distributions. Accordingly, vegetation in savanna and forest ecosystems should have differential responses to fire, but fire response data for herbaceous vegetation have yet to be synthesized across biomes. Here, we examined herbaceous vegetation responses to experimental fire at 30 sites spanning four continents. Across a variety of metrics, herbaceous vegetation increased in abundance where fire was applied, with larger responses to fire in wetter and in cooler and/or less seasonal systems. Compared to forests, savannas were associated with a 4.8 (±0.4) times larger difference in herbaceous vegetation abundance for burned versus unburned plots. In particular, grass cover decreased with fire exclusion in savannas, largely via decreases in C₄ grass cover, whereas changes in fire frequency had a relatively weak effect on grass cover in forests. These differential responses underscore the importance of fire for maintaining the vegetation structure of savannas and forests.     

Pyrodiversity interacts with rainfall to increase bird and mammal richness in African savannas

Fire is a fundamental process in savannas and is widely used for management. Pyrodiversity, variation in local fire characteristics, has been proposed as a driver of biodiversity although empirical evidence is equivocal. Using a new measure of pyrodiversity (Hempson et al.), we undertook the first continent‐wide assessment of how pyrodiversity affects biodiversity in protected areas across African savannas. The influence of pyrodiversity on bird and mammal species richness varied with rainfall: strongest support for a positive effect occurred in wet savannas (> 650 mm/year), where species richness increased by 27% for mammals and 40% for birds in the most pyrodiverse regions. Range‐restricted birds were most increased by pyrodiversity, suggesting the diversity of fire regimes increases the availability of rare niches. Our findings are significant because they explain the conflicting results found in previous studies of savannas. We argue that managing savanna landscapes to increase pyrodiversity is especially important in wet savannas.     

Contrasting Effects of Fire on Arboreal and Ground‐Dwelling Ant Communities of a Neotropical Savanna

Ants are a dominant group in tropical savannas and here we examined the responses of the arboreal and ground‐dwelling ant fauna to a fire in a Neotropical savanna (cerrado) reserve in Central Brazil. Ants were collected using pitfall traps and baits placed in trees and on the ground beneath each tree. Of the 36 trees marked along two transects, half (from each transect) were burned and half not. The same trees were sampled 1 wk before and again 3 and 12 mo after the fire. Rarefaction curves and ordination analyses using data from all trees from each side of each transect indicated that overall ant species richness and composition did not change after fire. Fire, however, reduced the mean number of ant species per tree, and increased the mean number of species on the ground. Fire increased the average abundance of specialist predators, Camponotini, and opportunistic species, and decreased that of arboreal specialists. Changes in the ground‐dwelling fauna were only detected 12 mo after the fire, while those in the arboreal fauna occurred earlier and were no longer apparent 12 mo after the fire. We suggest that these contrasting results represent mainly an indirect response of the ant communities to fire‐induced changes in vegetation. Given the temporary and small scale nature of the effects detected and the overall resilience of the ant fauna, our results indicate that a single fire in the cerrado vegetation does not greatly impact the structure of ant communities in the short term.     

Ant diversity in an Amazonian savanna: Relationship with vegetation structure,disturbance by fire,and dominant ants

Abstract The savannas of South America support a relatively diverse ant fauna, but little is known about the factors that influence the structure and dynamics of these assemblages. In 1998 and 2002, we surveyed the ground‐dwelling ant fauna and the fauna associated with the woody vegetation (using baits and direct sampling) from an Amazonian savanna. The aim was to evaluate the influence of vegetation structure, disturbance by fire and dominant ants on patterns of ant species richness and composition. Variations in the incidence of fires among our 39 survey plots had no or only limited influence on these patterns. In contrast, spatial variations in tree cover and cover by tall grasses (mostly Trachypogon plumosus), significantly affected ant species composition. Part of the variation in species richness among the study plots correlated with variations in the incidence of a dominant species (Solenopsis substituta) at baits. Ant species richness and composition also varied through time, possibly as an indirect effect of changes in vegetation cover. In many plots, and independently of disturbance by fire, there was a major increase in cover by tall grasses, which occupied areas formerly devoid of vegetation. Temporal changes in vegetation did not directly explain the observed increase in the number of ant species per plot. However, the incidence of S. substituta at baits declined sharply in 2002, especially in plots where changes in vegetation cover were more dramatic, and that decline was correlated with an increase in the number of ground‐dwelling species, a greater turnover of bait‐recruiting species and the appearance of the little fire ant Wasmannia auropunctata. The extent to which these changes in fact resulted from the relaxation of dominance by S. substituta is not clear. However, our results strongly suggest that the ant fauna of Amazonian savannas is affected directly and indirectly by the structure of the vegetation.     

Mammalian Insectivores Exert Top‐Down Effects on Azteca Ants

Insectivorous mammals are hypothesized to reduce the abundance of their insect prey. Using a 14‐yr mammal exclusion experiment, we demonstrate for the first time that a widespread and abundant Neotropical mammalian insectivore (Tamandua: Tamandua mexicana) reduced Azteca ant abundance. Azteca ant nests inside mammal exclosures were significantly larger than nests in control plots, where tamanduas were more abundant. These top‐down effects were caused not only by direct consumption, but also through non‐trophic direct effects, specifically nest damage. In contrast, tamanduas appeared to exert no significant top‐down effect on termite prey, which have strong chemical defenses. Our results are consistent with theory that strong defenses against predation can mitigate the top‐down effects of predators on some prey species. We argue that predicting the degree of top‐down effects caused by predators requires both a quantitative knowledge of prey choice and an understanding of the anti‐predator defenses of prey.     

Effects of a short fire‐return interval on resources and assemblage structure of birds in a tropical savanna

Fire frequency is a key land management issue, particularly in tropical savannas where fire is widely used and fire recurrence times are often short. We used an extended Before‐After‐Control‐Impact design to examine the impacts of repeated wet‐season burning for weed control on bird assemblages in a tropical savanna in north Queensland, Australia. Experimentally replicated fire treatments (unburnt, singularly bunt, twice burnt), in two habitats (riparian and adjacent open woodland), were surveyed over 3 years (1 year before the second burn, 1 year post the second burn, 2 years post the second burn) to examine responses of birds to a rapid recurrence of fire. Following the second burn, species richness and overall bird abundance were lower in the twice‐burnt sites than either the unburnt or singularly burnt sites. Feeding group composition varied across year of survey, but within each year, feeding guilds grouped according to fire treatment. In particular, abundance of frugivores and insectivores was lower in twice‐burnt sites, probably because of the decline of a native shrub that produces fleshy fruits, Carissa ovata. Although broader climatic variability may ultimately determine overall bird assemblages, our results show that a short fire‐return interval will substantially influence bird responses at a local scale. Considering that fire is frequently used as a land management tool, our results emphasize the importance of determining appropriate fire‐free intervals.     

Termite‐induced heterogeneity in African savanna vegetation: mechanisms and patterns

Objectives: To (1) assess the strength of evidence for the role of termites in vegetation heterogeneity in African savannas, and (2) identify the mechanisms by which termites induce such heterogeneity. Location: African savannas. Methods: We conducted a review of the literature, a meta‐analysis and qualitative systems analysis to identify mechanisms to explain the observed patterns. Results: The review provided evidence for termite‐induced heterogeneity in floristic composition and vegetation patterning in savannas across Africa. Termites induced vegetation heterogeneity directly or indirectly through their nest‐building and foraging activities, associated nutrient cycling and their interaction with mammalian herbivores and fire. The literature reviewed indicated that termite mounds essentially act as islands of fertility, which are responsible for ecosystem‐level spatial heterogeneity in savannas. This was supported by the meta‐analysis, which demonstrated that mounds of Ancistrotermes, Macrotermes, Odontotermes (family Macrotermitinae), Cubitermes (family Termitinae) and Trinervitermes (Nasutitermitinae) are significantly enriched in clay (75%), carbon (16%), total nitrogen (42%), calcium (232%), potassium (306%) and magnesium (154%) compared to the surrounding savanna soil. Conclusions: Termite activity is one of the major factors that induce vegetation patterning in African savannas. The implications of this are discussed and research questions for future studies and modelling efforts are indicated.     

Testing top‐down and bottom‐up effects on arid zone beetle assemblages following mammal reintroduction

Species extinctions and declines are occurring globally and commonly have cascading effects on ecosystems. In Australia, mammal extinctions have been extensive, particularly in arid areas, where precipitation drives ecosystems. Many ecologically extinct mammals feed on soil‐dwelling insects. However, how this top‐down pressure affected their prey and how this contrasts with the bottom‐up impacts of fluctuating precipitation remains unclear. We constructed a long‐term exclusion experiment in a multi‐species mammal reintroduction zone in semi‐arid Australia to test how top‐down (reintroduced mammals) and bottom‐up (precipitation) factors affect root‐feeding chafer beetles (Coleoptera: Melolonthinae). We used emergence traps in ten replicate 20 × 20 m plots of control, exclusion and procedural control treatments to trap chafers biannually from 2009 to 2015. Annual precipitation during this period varied from 173 to 481 mm. Mammal exclusion did not affect chafers, indicating that top‐down regulation was not important. Instead, chafer abundance, species density and biomass increased with precipitation. Chafer body size and assemblage composition were best predicted by sampling year, suggesting that random drift determined species abundances. Increased resource availability therefore favoured all species similarly. We thus found no evidence that mammal predation alters chafer populations and conclude that they may be driven primarily by bottom‐up processes. Further research should determine if the cascading effects of species loss are less important for herbivores generally than for higher level trophic groups and the role of ecosystem stability in mediating these patterns.     

Fire weather risk differs across rain forest—savanna boundaries in the humid tropics of north‐eastern Australia

Alternative stable state theory has been applied to understanding the control by landscape fire activity of pyrophobic tropical rain forest and pyrophytic eucalypt savanna boundaries, which are often separated by tall eucalypt forests. We evaluate the microclimate of three vegetation types across an elevational gradient and their relative fire risk as measured by McArthur's Forest Fire Danger Index (FFDI). Microclimatic data were collected from rain forest, tall eucalypt forest and savanna sites on eight vegetation boundaries throughout the humid tropics in north Queensland over a 3‐year period and were compared with data from a nearby meteorological station. There was a clear annual pattern in daily FFDI with highest values in the austral winter dry season and lowest values in the austral summer wet season. There was a strong association of the meteorological station FFDI values with those from the three vegetation types, albeit they were substantially lower. The rank order of FFDI values among the vegetation types decreased from savanna, tall eucalypt forest, then rain forest, a pattern that was consistent across each transect. Only very rarely would rain forest be flammable, despite being adjacent to highly flammable savannas. These results demonstrate the very strong effect of vegetation type on microclimate and fire risk, compared with the weak effect of elevation, consistent with a fire–vegetation feedback. This study is the first demonstration of how vegetation type influences microclimate and fire risk across a topographically complex tropical forest–savanna gradient.     

Contrasting demographics of tropical savanna and temperate forest eucalypts provide insight into how savannas and forests function. A case study using Corymbia clarksoniana from north‐eastern Australia

Eucalypts (Eucalyptus spp. and Corymbia spp.) dominate many communities across Australia, including frequently burnt tropical savannas and temperate forests, which receive less frequent but more intense fires. Understanding the demographic characteristics that allow related trees to persist in tropical savannas and temperate forest ecosystems can provide insight into how savannas and forests function, including grass–tree coexistence. This study reviews differences in critical stages in the life cycle of savanna and temperate forest eucalypts, especially in relation to fire. It adds to the limited data on tropical eucalypts, by evaluating the effect of fire regimes on the population biology of Corymbia clarksoniana, a tree that dominates some tropical savannas of north‐eastern Australia. Corymbia clarksoniana displays similar demographic characteristics to other tropical savanna species, except that seedling emergence is enhanced when seed falls onto recently burnt ground during a high rainfall period. In contrast to many temperate forest eucalypts, tropical savanna eucalypts lack canopy‐stored seed banks; time annual seed fall to coincide with the onset of predictable wet season rain; have very rare seedling emergence events, including a lack of mass germination after each fire; possess an abundant sapling bank; and every tropical eucalypt species has the ability to maintain canopy structure by epicormically resprouting after all but the most intense fires. The combination of poor seedling recruitment strategies, coupled with characteristics allowing long‐term persistence of established plants, indicate tropical savanna eucalypts function through the persistence niche rather than the regeneration niche. The high rainfall‐promoted seedling emergence of C. clarksoniana and the reduction of seedling survival and sapling growth by fire, support the predictions that grass–tree coexistence in savannas is governed by rainfall limiting tree seedling recruitment and regular fires limiting the growth of juvenile trees to the canopy.     

Short‐term response of threatened small macropods and their predators to prescribed burns in subtropical Australia

Fire is an important ecological process that shapes vegetation structure and habitat for faunal assemblages globally. Prescribed burns are increasingly being used in conservation and management to restore fire regimes in fire‐suppressed vegetation communities. Small threatened macropods require structurally complex habitat that allows them to evade detection by predators. Given that fire can alter vegetation structure, it can be viewed as a strong ecological force in shifting the dynamics between predator and prey species. Previous studies in temperate Australia have shown that prescribed burns in the presence of European Red Fox (Vulpes vulpes) and feral Cat (Felis catus) can have negative impacts on small macropods and medium‐sized mammals. Post‐fire response of threatened small macropods and their predators has not been experimentally examined in subtropical Australia despite this region providing refugia for the Long‐nosed Potoroo (Potorous tridactylus) and Red‐legged Pademelon (Thylogale stigmatica). We conducted a before‐after‐control‐impact fire experiment at two paired sites after low–moderate intensity burns typical of cool season prescribed burns. We used camera trapping to investigate changes in activity of threatened small macropods and their predators. We also recorded vegetation change. Despite large reductions in ground and shrub cover, activity of small macropods and the Dingo (Canis dingo) did not change in response to fires. Therefore, the threat of dingo predation appears to have remained unchanged following the fires. Although feral cats and foxes were present, they showed negligible activity across our sites. Our study suggests that small‐scale patchy ecological burns may not lead to increased predation of small macropods in our landscape. We attribute this to sufficient post‐fire refugia and very low densities of foxes.     

Episodic outbreaks of small mammals influence predator community dynamics in an east African savanna ecosystem

Little is known about the dynamics of small mammals in tropical savanna: a critical gap in our understanding of Africa's best known ecosystems. Historical evidence suggested small mammals peak in abundance (outbreak) in Serengeti National Park (SNP), as in agricultural systems. We asked 1) what are bottom–up drivers of small mammals and 2) do predators have top–down effects? We documented dynamics of small mammals, birds of prey, and mammalian carnivores in SNP and agricultural areas. We used climatic fluctuations and differences between unmodified and agricultural systems as perturbations to examine trophic processes, key to understanding responses to climate change and increasing human pressures. Data were derived from intermittent measures of abundance collected 1968–1999, combined with systematic sampling 2000–2010 to construct a 42‐year time series. Data on abundance of black‐shouldered kites (1968–2010), eight other species of rodent‐eating birds (1997–2010), and 10 carnivore species (1993–2010) were also collated. Outbreaks occurred every 3–5 years in SNP, with low or zero abundance between peaks. There was a positive relationship between rainfall in the wet season and 1) small mammal abundance and 2) the probability of an outbreak, both of which increased with negative Southern Oscillation Index values. Rodent‐eating birds and carnivores peaked 6–12 months after small mammals. In agricultural areas, abundance remained higher than in natural habitats. Abundances of birds of prey and mammalian carnivores were extremely low in these areas and not related to small mammal abundance. Small mammals are an important food resource for higher trophic levels in the Serengeti ecosystem. Changes in climate and land use may alter their future dynamics, with cascading consequences for higher trophic levels, including threatened carnivores. Although outbreaks cause substantial damage to crops in agricultural areas, small mammals also play a vital role in maintaining some of the diversity and complexity found in African savanna ecosystems.     

Long‐term variability and rainfall control of savanna fire regimes in equatorial East Africa

Fires burning the vast grasslands and savannas of Africa significantly influence the global carbon cycle. Projecting the impacts of future climate change on fire‐mediated biogeochemical processes in these dry tropical ecosystems requires understanding of how various climate factors influence regional fire regimes. To examine climate–vegetation–fire linkages in dry savanna, we conducted macroscopic and microscopic charcoal analysis on the sediments of the past 25 000 years from Lake Challa, a deep crater lake in equatorial East Africa. The charcoal‐inferred shifts in local and regional fire regimes were compared with previously published reconstructions of temperature, rainfall, seasonal drought severity, and vegetation dynamics to evaluate millennial‐scale drivers of fire occurrence. Our charcoal data indicate that fire in the dry lowland savanna of southeastern Kenya was not fuel‐limited during the Last Glacial Maximum (LGM) and Late Glacial, in contrast to many other regions throughout the world. Fire activity remained high at Lake Challa probably because the relatively high mean‐annual temperature (~22 °C) allowed productive C₄ grasses with high water‐use efficiency to dominate the landscape. From the LGM through the middle Holocene, the relative importance of savanna burning in the region varied primarily in response to changes in rainfall and dry‐season length, which were controlled by orbital insolation forcing of tropical monsoon dynamics. The fuel limitation that characterizes the region's fire regime today appears to have begun around 5000–6000 years ago, when warmer interglacial conditions coincided with prolonged seasonal drought. Thus, insolation‐driven variation in the amount and seasonality of rainfall during the past 25 000 years altered the immediate controls on fire occurrence in the grass‐dominated savannas of eastern equatorial Africa. These results show that climatic impacts on dry‐savanna burning are heterogeneous through time, with important implications for efforts to anticipate future shifts in fire‐mediated ecosystem processes.     

Variable rainfall has a greater effect than fire on the demography of the dominant tree in a semi‐arid Eucalyptus savanna

Rainfall, fire and competition are emphasized as determinants of the density and basal area of woody vegetation in savanna. The semi‐arid savannas of Australia have substantial multi‐year rainfall deficits and insufficient grass fuel to carry annual fire in contrast to the mesic savannas in more northern regions. This study investigates the influence of rainfall deficit and excess, fire and woody competition on the population dynamics of a dominant tree in a semi‐arid savanna. All individuals of Eucalyptus melanophloia were mapped and monitored in three, 1‐ha plots over an 8.5 year period encompassing wet and dry periods. The plots were unburnt, burnt once and burnt twice. A competition index incorporating the size and distance of neighbours to target individuals was determined. Supplementary studies examined seedling recruitment and the transition of juvenile trees into the sapling layer. Mortality of burnt seedlings was related to lignotuber area but the majority of seedlings are fire resistant within 12 months of germination. Most of the juveniles (≤1 cm dbh) of E. melanophloia either died in the dry period or persisted as juveniles throughout 8.5 years of monitoring. Mortality of juveniles was positively related to woody competition and was higher in the dry period than the wet period. The transition of juveniles to a larger size class occurred at extremely low rates, and a subsidiary study along a clearing boundary suggests release from woody competition allows transition into the sapling layer. From three fires the highest proportion of saplings (1–10 cm dbh) reduced to juveniles was only 5.6% suggesting rates of ‘top‐kill’ of E. melanophloia as a result of fire are relatively low. Girth growth was enhanced in wet years, particularly for larger trees (>10 cm dbh), but all trees regardless of size or woody competition levels are vulnerable to drought‐induced mortality. Overall the results suggest that variations in rainfall, especially drought‐induced mortality, have a much stronger influence on the tree demographics of E. melanophloia in a semi‐arid savanna of north‐eastern Australia than fire.