Fire shapes the distribution of savanna and forest through complex interactions involving climate, resources and species traits. Based on data from central Brazil, we propose that these interactions are governed by two critical thresholds. The fire-resistance threshold is reached when individual trees have accumulated sufficient bark to avoid stem death, whereas the fire-suppression threshold is reached when an ecosystem has sufficient canopy cover to suppress fire by excluding grasses. Surpassing either threshold is dependent upon long fire-free intervals, which are rare in mesic savanna. On high-resource sites, the thresholds are reached quickly, increasing the probability that savanna switches to forest, whereas low-resource sites are likely to remain as savanna even if fire is infrequent. Species traits influence both thresholds; saplings of savanna trees accumulate bark thickness more quickly than forest trees, and are more likely to become fire resistant during fire-free intervals. Forest trees accumulate leaf area more rapidly than savanna trees, thereby accelerating the transition to forest. Thus, multiple factors interact with fire to determine the distribution of savanna and forest by influencing the time needed to reach these thresholds. Future work should decipher multiple environmental controls over the rates of tree growth and canopy closure in savanna.

Tropical savanna and forest are recognized to represent alternate stable states, primarily determined by feedbacks with fire. Vegetation-fire dynamics in each of these vegetation types are largely determined by the influence of the vegetation on fire behavior, as well as the effects of fire behavior on tree mortality, topkill (defined here as complete death of the aerial biomass, regardless of whether the plant recovers by resprouting), and rate of growth of resprouts. We studied the effect of fire on three savanna-forest boundaries in central Brazil. Fire intensity was greater in savanna than forest, as inferred by a twofold greater height of stem charring. Despite lower fire intensity, forest tree species exhibited higher rates of topkill, which was best explained by their thinner bark, relative to savanna species. Following topkill, there was no tendency for sprouts of savanna trees to grow faster than those of forest species, contrary to expectations, nor was whole-plant mortality higher in forest than in savanna. This contrasts with observations of high rates of postburn mortality in many other tropical forests. The low tree mortality in these transitional forests suggests that the dynamic of these natural savanna-forest boundaries is fundamentally different from that of forest boundaries originating from deforestation in the humid tropics. The forests studied here appear to be much more resilient to occasional incursion of fire from the savanna, despite being unable to invade frequently burned savanna. The thin bark of forest species makes them particularly susceptible to the "fire trap," whereby repeated topkill of small trees prevents recruitment into adult size classes. Rapid growth will be particularly important for forest species to escape the fire trap, so we predict that, where fire is frequent, forests should be restricted to high-resource sites. Here, Mg2+ and Ca2+ concentrations had particularly strong effects on postburn growth rates, suggesting that these elements may most strongly limit the distribution of forest in these fire-prone savannas.

Soils of the cerrado biome, mostly oxisols and deep sandy entisols, are acid, dystrophic and poor in available nutrients. These soils are not very different from soils that occur in the Amazon region. However, the open savanna physiognomies of cerrado with lower biomass of their different components are deficient in nutrients at the ecosystem level, unlike the Amazon forests which retain high nutrient reserves in their live biomass. Field crops are susceptible to aluminum and manganese toxicities, besides nutrient deficiencies, in cerrado soils and do not grow well in the absence of liming and fertilization. However, concepts of nutrient deficiencies and toxicities, well established for cultivated plants, should not be extended to native species in natural ecosystems, indiscriminately. Many native plants in the cerrado biome are resistant or tolerant to soil conditions deemed unfavorable for cultivated plants but their geographic distribution, frequency in native communities, growth and productivity are determined by water and nutrient availability and other edaphic conditions. Species growing on acid soils are aluminum tolerant or resistant, since their capacity to absorb essential nutrients, growth and reproduction is not affected by high aluminum levels in the soil. Many common species of the cerrado, instead of excluding aluminum, absorb and transport it to leaves and accumulate it in different tissues including leaves and seeds whereas others do not survive in the absence of exchangeable aluminum, even though no specific role of Al in plant metabolism is yet established. Key words: aluminum toxicity, latosol, mineral nutrition, soil fertility Adaptações de plantas nativas do cerrado em solos ácidos: Os solos do bioma cerrado, na maioria oxisols e entisols, são ácidos e distróficos, e apresentam baixa disponibilidade de nutrientes. Estes solos não são muito diferentes dos solos da região amazônica. Entretanto, as formas abertas de vegetação do cerrado com baixa biomassa de seus diferentes componentes são deficientes em nutrientes em nível do ecossistema ao contrário das florestas amazônicas que possuem uma maior reserva de nutrientes na sua biomassa vegetal. As plantas cultivadas são susceptíveis a toxicidade de alumínio e manganês nos solos do cerrado independente da deficiência de nutrientes e não crescem bem na ausência de calagem e adubação. Entretanto, os conceitos de deficiência de nutrientes e toxicidade, bem estabelecidos na agricultura, não devem ser estendidos às plantas nativas em ecossistemas naturais, indiscriminadamente. As inúmeras espécies das plantas nativas que ocorrem no bioma são resistentes ou tolerantes às condições edáficas consideradas desfavoráveis às plantas cultivadas, mas sua distribuição, freqüência nas comunidades nativas, crescimento e produtividade são determinados pela disponibilidade de nutrientes, regime hídrico do solo e outros fatores edáficos. As espécies crescendo em solos ácidos são tolerantes ou resistentes ao alumínio por que sua capacidade de absorção de nutrientes ...

Upland tropical forests have expanded and contracted in response to past climates, but it is not clear whether similar dynamics were exhibited by gallery (riparian) forests within savanna biomes. Because such forests generally have access to ample water, their extent may be buffered against changing climates. We tested the long-term stability of gallery forest boundaries by characterizing the border between gallery forests and savannas and tracing the presence of gallery forest through isotopic analysis of organic carbon in the soil profile. We measured leaf area index, grass vs. shrub or tree coverage, the organic carbon, phosphorus, nitrogen and calcium concentrations in soils and the carbon isotope ratios of soil organic matter in two transitions spanning gallery forests and savanna in a Cerrado ecosystem. Gallery forests without grasses typically show a greater leaf area index in contrast to savannas, which show dense grass coverage. Soils of gallery forests have significantly greater concentrations of organic carbon, phosphorus, nitrogen and calcium than those of savannas. Soil organic carbon of savannas is significantly more enriched in 13 C compared with that of gallery forests. This difference in enrichment is in part caused by the presence of C 4 grasses in savanna ecosystem and its absence in gallery forests. Using the 13 C abundance as a signature for savanna and gallery forest ecosystems in 1 m soil cores, we show that the borders of gallery forests have expanded into the savanna and that this process initiated at least 3000-4000 BP based on 14 C analysis. Gallery forests, however, may be still expanding as we found more recent transitions according to 14 C activity measurements. We discuss the possible mechanisms of gallery forest expansion and the means by which nutrients required for the expansion of gallery forest might accumulate.

Summary 1.Leaf traits are commonly associated with the life history, distribution and resource requirements of a species. To improve our understanding of the ecological and physiological differences between tropical savanna and forest trees, we compared leaf traits of species native to savanna and gallery (riverine) forests in the Cerrado region of central Brazil. 2. Congeneric species pairs from 14 different taxonomic families were studied, each with a savanna species and a forest species present at the study site. Only individuals growing in savanna conditions under full sun were studied. We measured foliar nutrients, δ 13 C, δ 15 N and specific leaf area (SLA: leaf area per unit leaf mass). We used phylogenetically independent contrasts to compare savanna and forest species and to test for correlations among species traits. 3. Overall, leaves of forest species had 17% higher N concentration, 32% higher P concentration, and 37% higher K concentration, despite growing in similar soils. Concentrations of all three elements were strongly and positively correlated with SLA. 4. Forest species had 52% greater SLA, on average, than savanna species, which accounts for the higher foliar nutrient concentrations of these species. 5. Savanna species had higher δ 13 C values than forest species, indicating higher wateruse efficiency. The SLA was negatively correlated with δ 13 C, suggesting that SLA may also account for the higher water-use efficiency of savanna species. 6. There was no difference in foliar δ 15 N between savanna and forest species, but foliar δ 15 N was negatively correlated with soil pH. 7. These results contribute to recent studies showing that tropical savanna and forest species represent two distinct functional types, with large differences in ecology and physiology, that have important consequences for the dynamics of savanna-forest boundaries.