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.

Effects and response of the Cerrado ground-layer to frost along the canopy cover gradient.

Frost effects on savanna plant communities have been considered as analogous to those from fire, both changing community structure and filtering species composition. However, while frost impacts have been well-studied for the woody component of savannas, it is still poorly explored for the ground-layer community. Here, we investigated effects of frost in the Cerrado along a gradient of tree cover, focusing on ground-layer plant species, near the southern limit of the Cerrado in Brazil. We aimed to elucidate if the pattern already described for the tree layer also extends to the ground layer in terms of mimicking the effects of fire on vegetation structure and composition. We assessed how damage severity differs across species and across the tree-cover gradient, and we examined the recovery process after frost in terms of richness and community structure along the canopy cover gradient. Frost caused immediate and widespread dieback of the perennial ground-layer, with greatest impact on community structure where tree cover was lowest. However, frost did not reduce the number of species, indicating community resilience to this natural disturbance. Although frost mimicked the effects of fire in some ways, in other ways it differed substantially from fire. Unlike fire, frost increases litter cover and decreases the proportion of bare soil, likely hindering crucial processes for recovery of plant populations, such as seed dispersal, seed germination and plant resprouting. This finding calls attention to the risk of misguided conclusions when the ground layer is neglected in ecological studies of tropical savannas and grasslands.

Flammability thresholds or flammability gradients? Determinants of fire across savanna-forest transitions.

Vegetation-fire feedbacks are important for determining the distribution of forest and savanna. To understand how vegetation structure controls these feedbacks, we quantified flammability across gradients of tree density from grassland to forest in the Brazilian Cerrado. We experimentally burned 102 plots, for which we measured vegetation structure, fuels, microclimate, ignition success and fire behavior. Tree density had strong negative effects on ignition success, rate of spread, fire-line intensity and flame height. Declining grass biomass was the principal cause of this decline in flammability as tree density increased, but increasing fuel moisture contributed. Although the response of flammability to tree cover often is portrayed as an abrupt, largely invariant threshold, we found the response to be gradual, with considerable variability driven largely by temporal changes in atmospheric humidity. Even when accounting for humidity, flammability at intermediate tree densities cannot be predicted reliably. Fire spread in savanna-forest mosaics is not as deterministic as often assumed, but may appear so where vegetation boundaries are already sharp. Where transitions are diffuse, fire spread is difficult to predict, but should become increasingly predictable over multiple fire cycles, as boundaries are progressively sharpened until flammability appears to respond in a threshold-like manner.

Correction to: Allocation to leaf area and sapwood area affects water relations of co-occurring savanna and forest trees.

The original version of this article unfortunately contained a mistake. The Electronic supplementary material (ESM) was accompanying this article by mistake.

Convergence of bark investment according to fire and climate structures ecosystem vulnerability to future change.

Fire regimes in savannas and forests are changing over much of the world. Anticipating the impact of these changes requires understanding how plants are adapted to fire. In this study, we test whether fire imposes a broad selective force on a key fire-tolerance trait, bark thickness, across 572 tree species distributed worldwide. We show that investment in thick bark is a pervasive adaptation in frequently burned areas across savannas and forests in both temperate and tropical regions where surface fires occur. Geographic variability in bark thickness is largely explained by annual burned area and precipitation seasonality. Combining environmental and species distribution data allowed us to assess vulnerability to future climate and fire conditions: tropical rainforests are especially vulnerable, whereas seasonal forests and savannas are more robust. The strong link between fire and bark thickness provides an avenue for assessing the vulnerability of tree communities to fire and demands inclusion in global models.

Stomatal acclimation to vapour pressure deficit doubles transpiration of small tree seedlings with warming.

Future climate change is expected to increase temperature (T) and atmospheric vapour pressure deficit (VPD) in many regions, but the effect of persistent warming on plant stomatal behaviour is highly uncertain. We investigated the effect of experimental warming of 1.9-5.1 °C and increased VPD of 0.5-1.3 kPa on transpiration and stomatal conductance (gs ) of tree seedlings in the temperate forest understory (Duke Forest, North Carolina, USA). We observed peaked responses of transpiration to VPD in all seedlings, and the optimum VPD for transpiration (Dopt ) shifted proportionally with increasing chamber VPD. Warming increased mean water use of Carya by 140% and Quercus by 150%, but had no significant effect on water use of Acer. Increased water use of ring-porous species was attributed to (1) higher air T and (2) stomatal acclimation to VPD resulting in higher gs and more sensitive stomata, and thereby less efficient water use. Stomatal acclimation maintained homeostasis of leaf T and carbon gain despite increased VPD, revealing that short-term stomatal responses to VPD may not be representative of long-term exposure. Acclimation responses differ from expectations of decreasing gs with increasing VPD and may necessitate revision of current models based on this assumption.

Shifts in functional traits elevate risk of fire-driven tree dieback in tropical savanna and forest biomes.

Numerous predictions indicate rising CO2 will accelerate the expansion of forests into savannas. Although encroaching forests can sequester carbon over the short term, increased fires and drought-fire interactions could offset carbon gains, which may be amplified by the shift toward forest plant communities more susceptible to fire-driven dieback. We quantify how bark thickness determines the ability of individual tree species to tolerate fire and subsequently determine the fire sensitivity of ecosystem carbon across 180 plots in savannas and forests throughout the 2.2-million km(2) Cerrado region in Brazil. We find that not accounting for variation in bark thickness across tree species underestimated carbon losses in forests by ~50%, totaling 0.22 PgC across the Cerrado region. The lower bark thicknesses of plant species in forests decreased fire tolerance to such an extent that a third of carbon gains during forest encroachment may be at risk of dieback if burned. These results illustrate that consideration of trait-based differences in fire tolerance is critical for determining the climate-carbon-fire feedback in tropical savanna and forest biomes.

Temperature alone does not explain phenological variation of diverse temperate plants under experimental warming.

Anthropogenic climate change has altered temperate forest phenology, but how these trends will play out in the future is controversial. We measured the effect of experimental warming of 0.6-5.0 °C on the phenology of a diverse suite of 11 plant species in the deciduous forest understory (Duke Forest, North Carolina, USA) in a relatively warm year (2011) and a colder year (2013). Our primary goal was to dissect how temperature affects timing of spring budburst, flowering, and autumn leaf coloring for functional groups with different growth habits, phenological niches, and xylem anatomy. Warming advanced budburst of six deciduous woody species by 5-15 days and delayed leaf coloring by 18-21 days, resulting in an extension of the growing season by as much as 20-29 days. Spring temperature accumulation was strongly correlated with budburst date, but temperature alone cannot explain the diverse budburst responses observed among plant functional types. Ring-porous trees showed a consistent temperature response pattern across years, suggesting these species are sensitive to photoperiod. Conversely, diffuse-porous species responded differently between years, suggesting winter chilling may be more important in regulating budburst. Budburst of the ring-porous Quercus alba responded nonlinearly to warming, suggesting evolutionary constraints may limit changes in phenology, and therefore productivity, in the future. Warming caused a divergence in flowering times among species in the forest community, resulting in a longer flowering season by 10-16 days. Temperature was a good predictor of flowering for only four of the seven species studied here. Observations of interannual temperature variability overpredicted flowering responses in spring-blooming species, relative to our warming experiment, and did not consistently predict even the direction of flowering shifts. Experiments that push temperatures beyond historic variation are indispensable for improving predictions of future changes in phenology.

Carbon accumulation and nitrogen pool recovery during transitions from savanna to forest in central Brazil.

The expansion of tropical forest into savanna may potentially be a large carbon sink, but little is known about the patterns of carbon sequestration during transitional forest formation. Moreover, it is unclear how nutrient limitation, due to extended exposure to fire-driven nutrient losses, may constrain carbon accumulation. Here, we sampled plots that spanned a woody biomass gradient from savanna to transitional forest in response to differential fire protection in central Brazil. These plots were used to investigate how the process of transitional forest formation affects the size and distribution of carbon (C) and nitrogen (N) pools. This was paired with a detailed analysis of the nitrogen cycle to explore possible connections between carbon accumulation and nitrogen limitation. An analysis of carbon pools in the vegetation, upper soil, and litter shows that the transition from savanna to transitional forest can result in a fourfold increase in total carbon (from 43 to 179 Mg C/ha) with a doubling of carbon stocks in the litter and soil layers. Total nitrogen in the litter and soil layers increased with forest development in both the bulk (+68%) and plant-available (+150%) pools, with the most pronounced changes occurring in the upper layers. However, the analyses of nitrate concentrations, nitrate:ammonium ratios, plant stoichiometry of carbon and nitrogen, and soil and foliar nitrogen isotope ratios suggest that a conservative nitrogen cycle persists throughout forest development, indicating that nitrogen remains in low supply relative to demand. Furthermore, the lack of variation in underlying soil type (>20 cm depth) suggests that the biogeochemical trends across the gradient are driven by vegetation. Our results provide evidence for high carbon sequestration potential with forest encroachment on savanna, but nitrogen limitation may play a large and persistent role in governing carbon sequestration in savannas or other equally fire-disturbed tropical landscapes. In turn, the link between forest development and nitrogen pool recovery creates a framework for evaluating potential positive feedbacks on savanna-forest boundaries.

Are winter-active species vulnerable to climate warming? A case study with the wintergreen terrestrial orchid, Tipularia discolor.

In the eastern United States, winter temperature has been increasing nearly twice as fast as summer temperature, but studies of warming effects on plants have focused on species that are photosynthetically active in summer. The terrestrial orchid Tipularia discolor is leafless in summer and acquires C primarily in winter. The optimum temperature for photosynthesis in T. discolor is higher than the maximum temperature throughout most of its growing season, and therefore growth can be expected to increase with warming. Contrary to this hypothesis, experimental warming negatively affected reproductive fitness (number of flowering stalks, flowers, fruits) and growth (change in leaf area from 2010 to 2012) in T. discolor. Temperature in June-July was critical for flowering, and mean July temperature greater than 29 °C (i.e., 2.5 °C above ambient) eliminated reproduction. Warming of 1.2 °C delayed flowering by an average of 10 days and fruiting by an average of 5 days. Warming of 4.4 °C reduced relative growth rates by about 60%, which may have been partially caused by the direct effects of temperature on photosynthesis and respiration. Warming indirectly increased vapor pressure deficit (VPD) by 0.2-0.5 kPa, and leaf-to-air VPD over 1.3 kPa restricted stomatal conductance of T. discolor to 10-40% of maximum conductance. These results highlight the need to account for changes in VPD when estimating temperature responses of plant species under future warming scenarios. Increasing temperature in the future will likely be an important limiting factor to the distribution of T. discolor, especially along the southern edge of its range.

Stem heating results in hydraulic dysfunction in Symplocos tinctoria: implications for post-fire tree death.

Fire-induced heating of stems can impair plant water transport by deforming xylem and increasing vulnerability to cavitation, but it is not clear whether these effects can result in tree death, or how quickly this may occur. In field experiments, we heated stems of Symplocos tinctoria (L.) L'Hér saplings to 90 °C using a thin-film resistive heater, and we monitored stomatal conductance, leaf water potential, sap flow and hydraulic conductivity until stem death. Sap flow and stomatal conductance declined quickly after heating, while whole-plant hydraulic conductance and leaf water potential remained high for the first week. In fact, leaf water potential increased during the first days after heating, indicating that stomatal closure was not initially caused by leaf water deficit induced by impaired water transport. After 1 week, leaf water potential and whole-plant conductance declined below unheated controls, while stomatal conductance and sap flow continued declining, approaching zero after 2 weeks. To better understand the cause of these declines, we directly measured hydraulic conductivity of heated stems. Stems underwent a progressive decline in conductivity after heating, and by the time that samples were severely wilted or desiccated, the heated portion of stems had little or no conductivity. Importantly, conductivity of heated stems was not recovered by flushing stems to remove embolisms, suggesting the existence of physical occlusions. Scanning electron micrographs did not reveal deformed cell walls, nor did it identify alternative causes of blockages. These results reveal that stem heating can result in xylem dysfunction and mortality, but neither response is immediate. Dysfunction was likely caused by wound responses rather than embolism, but improved understanding of the mechanisms of heat-induced hydraulic failure is needed.

Ecological thresholds at the savanna-forest boundary: how plant traits, resources and fire govern the distribution of tropical biomes.

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.

Caught in a fire trap: recurring fire creates stable size equilibria in woody resprouters.

Globally, fire maintains many mesic habitats in an open canopy state by killing woody plants while reducing the size of those able to resprout. Where fire is frequent, tree saplings are often suppressed by a "fire trap" of repeated topkill (death of aerial biomoass) and resprouting, preventing them from reaching adult size. The ability to tolerate repeated topkill is an essential life-history trait that allows a sapling to persist until it experiences a long fire-free interval, during which it can escape the fire trap. We hypothesized that persistence in the fire trap results from a curvilinear relationship between pre-burn size and resprout size, which causes a plant to approach an equilibrial size in which post-fire biomass recovery is equal to fire-induced biomass loss. We also predicted that the equilibrial stem size is positively related to resource availability. To test these hypotheses, we collected data on pre-burn and resprout size of five woody plant species at wetland ecotones in longleaf pine savanna subjected to frequent burning. As expected, all species exhibited similar curvilinear relationships between pre-burn size and resprout size. The calculated equilibrial sizes were strong predictors of mean plant size across species and growing conditions, supporting the persistence equilibrium model. An alternative approach using matrix models yielded similar results. Resprouting was less vigorous in dry sites than at wet sites, resulting in smaller equilibrial stem sizes in drier sites; extrapolating these results provides an explanation for the absence of these species in xeric uplands. This new framework offers a straightforward approach to guide data collection for experimental, comparative, and modeling studies of plant persistence and community dynamics in frequently burned habitats.

Deciphering the distribution of the savanna biome.

• We aimed to identify the limits of savanna across Africa, Australia and South America. We based our investigation on the rich history of hypotheses previously examined: that the limits of savanna are variously determined by rainfall, rainfall seasonality, soil fertility and disturbance. • We categorized vegetation on all continents as 'savanna' (open habitats with a C(4) grass layer) or 'not-savanna' (closed habitats with no C(4) grass layer) and used a combination of statistical approaches to examine how the presence of savanna varied as a function of five environmental correlates. • The presence of savanna is constrained by effective rainfall and rainfall seasonality. Soil fertility is regionally important, although the direction of its effect changes relative to rainfall. We identified three continental divergences in the limits of savanna that could not be explained by environment. • Climate and soils do not have a deterministic effect on the distribution of savanna. Over the range of savanna, some proportion of the land is always 'not-savanna'. We reconciled previous contradictory views of savanna limits by developing a new conceptual framework for understanding these limits by categorizing environmental factors into whether they had a positive or negative effect on woody growth and the frequency of disturbance.

Recent vicariance and the origin of the rare, edaphically specialized Sandhills lily, Lilium pyrophilum (Liliaceae): evidence from phylogenetic and coalescent analyses.

Establishing the phylogenetic and demographic history of rare plants improves our understanding of mechanisms that have led to their origin and can lead to valuable insights that inform conservation decisions. The Atlantic coastal plain of eastern North America harbours many rare and endemic species, yet their evolution is poorly understood. We investigate the rare Sandhills lily (Lilium pyrophilum), which is endemic to seepage slopes in a restricted area of the Atlantic coastal plain of eastern North America. Using phylogenetic evidence from chloroplast, nuclear internal transcribed spacer and two low-copy nuclear genes, we establish a close relationship between L. pyrophilum and the widespread Turk's cap lily, L. superbum. Isolation-with-migration and coalescent simulation analyses suggest that (i) the divergence between these two species falls in the late Pleistocene or Holocene and almost certainly post-dates the establishment of the edaphic conditions to which L. pyrophilum is presently restricted, (ii) vicariance is responsible for the present range disjunction between the two species, and that subsequent gene flow has been asymmetrical and (iii) L. pyrophilum harbours substantial genetic diversity in spite of its present rarity. This system provides an example of the role of edaphic specialization and climate change in promoting diversification in the Atlantic coastal plain.

Hydraulic segmentation does not protect stems from acute water loss during fire.

The heat plume associated with fire has been hypothesized to cause sufficient water loss from trees to induce embolism and hydraulic failure. However, it is unclear whether the water transport path remains sufficiently intact during scorching or burning of foliage to sustain high water loss. We measured water uptake by branches of Magnolia grandiflora while exposing them to a range of fire intensities and examined factors influencing continued water uptake after fire. Burning caused a 22-fold mean increase in water uptake, with greatest rates of water loss observed at burn intensities that caused complete consumption of leaves. Such rapid uptake is possible only with steep gradients in water potential, which would likely result in substantial cavitation of xylem and loss of conductivity in intact stems. Water uptake continued after burning was complete and was greatest following burn intensities that killed leaves but did not consume them. This post-fire uptake was mostly driven by rehydration of the remaining tissues, rather than evaporation from the tissues. Our results indicate that the fire plume hypothesis can be expanded to include a wide range of burning conditions experienced by plants. High rates of water loss are sustained during burning, even when leaves are killed or completely consumed.

Decadal changes in fire frequencies shift tree communities and functional traits.

Global change has resulted in chronic shifts in fire regimes. Variability in the sensitivity of tree communities to multi-decadal changes in fire regimes is critical to anticipating shifts in ecosystem structure and function, yet remains poorly understood. Here, we address the overall effects of fire on tree communities and the factors controlling their sensitivity in 29 sites that experienced multi-decadal alterations in fire frequencies in savanna and forest ecosystems across tropical and temperate regions. Fire had a strong overall effect on tree communities, with an average fire frequency (one fire every three years) reducing stem density by 48% and basal area by 53% after 50 years, relative to unburned plots. The largest changes occurred in savanna ecosystems and in sites with strong wet seasons or strong dry seasons, pointing to fire characteristics and species composition as important. Analyses of functional traits highlighted the impact of fire-driven changes in soil nutrients because frequent burning favoured trees with low biomass nitrogen and phosphorus content, and with more efficient nitrogen acquisition through ectomycorrhizal symbioses. Taken together, the response of trees to altered fire frequencies depends both on climatic and vegetation determinants of fire behaviour and tree growth, and the coupling between fire-driven nutrient losses and plant traits.

Evidence for range stasis during the latter Pleistocene for the Atlantic Coastal Plain endemic genus, Pyxidanthera Michaux.

The general phylogeographical paradigm for eastern North America (ENA) is that many plant and animal species retreated into southern refugia during the last glacial period, then expanded northward after the last glacial maximum (LGM). However, some taxa of the Gulf and Atlantic Coastal Plain (GACP) demonstrate complex yet recurrent distributional patterns that cannot be explained by this model. For example, eight co-occurring endemic plant taxa with ranges from New York to South Carolina exhibit a large disjunction separating northern and southern populations by >300 km. Pyxidanthera (Diapensiaceae), a plant genus that exhibits this pattern, consists of two taxa recognized as either species or varieties. We investigated the taxonomy and phylogeography of Pyxidanthera using morphological data, cpDNA sequences, and amplified fragment length polymorphism markers. Morphological characters thought to be important in distinguishing Pyxidanthera barbulata and P. brevifolia demonstrate substantial overlap with no clear discontinuities. Genetic differentiation is minimal and diversity estimates for northern and southern populations of Pxyidanthera are similar, with no decrease in rare alleles in northern populations. In addition, the northern populations harbour several unique cpDNA haplotypes. Pyxidanthera appears to consist of one morphologically variable species that persisted in or near its present range at least through the latter Pleistocene, while the vicariance of the northern and southern populations may be comparatively recent. This work demonstrates that the refugial paradigm is not always appropriate and GACP endemic plants, in particular, may exhibit phylogeographical patterns qualitatively different from those of other ENA plant species.

Allocation to leaf area and sapwood area affects water relations of co-occurring savanna and forest trees.

Water availability is a principal factor limiting the distribution of closed-canopy forest in the seasonal tropics, suggesting that forest tree species may not be well adapted to cope with seasonal drought. We studied 11 congeneric species pairs, each containing one forest and one savanna species, to test the hypothesis that forest trees have a lower capacity to maintain seasonal homeostasis in water relations relative to savanna species. To quantify this, we measured sap flow, leaf water potential (Psi(L)), stomatal conductance (g (s)), wood density, and Huber value (sapwood area:leaf area) of the 22 study species. We found significant differences in the water relations of these two species types. Leaf area specific hydraulic conductance of the soil/root/leaf pathway (G (t)) was greater for savanna species than forest species. The lower G (t) of forest trees resulted in significantly lower Psi(L) and g (s) in the late dry season relative to savanna trees. The differences in G (t) can be explained by differences in biomass allocation of savanna and forest trees. Savanna species had higher Huber values relative to forest species, conferring greater transport capacity on a leaf area basis. Forest trees have a lower capacity to maintain homeostasis in Psi(L) due to greater allocation to leaf area relative to savanna species. Despite significant differences in water relations, relationships between traits such as wood density and minimum Psi(L) were indistinguishable for the two species groups, indicating that forest and savanna share a common axis of water-use strategies involving multiple traits.