A grass–fire cycle in Australian tropical savannas has been postulated as driving the regional decline of the obligate-seeding conifer Callitris intratropica and other fire-sensitive components of the regional flora and fauna, due to proliferation of flammable native grasses. We tested the hypothesis that a high-biomass invasive savanna grass drives a positive feedback process where intense fires destroy fire-sensitive trees, and the reduction in canopy cover facilitates further invasion by grass. We undertook an observational and experimental study using, as a model system, a plantation of C. intratropica that has been invaded by an African grass, gamba (Andropogon gayanus) in the Northern Territory, Australia. We found that high grass biomass was associated with reduced canopy cover and restriction of foliage to the upper canopy of surviving stems, and mortality of adult trees was very high (>50%) even in areas with low fuel loads (1 t·ha−1). Experimental fires, with fuel loads >10 t·ha−1, typical of the grass-invasion front, caused significant mortality due to complete crown scorch. Lower fuel loads cause reduced canopy cover through defoliation of the lower canopy. These results help explain how increases in grass biomass are coupled with the decline of C. intratropica throughout northern Australia by causing a switch from litter and sparse perennial grass fuels, and hence low-intensity surface fires, to heavy annual grass fuel loads that sustain fires that burn into the midstorey. This study demonstrates that changes in fuel type can alter fire regimes with substantial knock-on effects on the biota.

Predicting the consequences of climate change on forest systems is difficult because trees may display species-specific responses to exaggerated droughts that may not be reflected by the climatic envelope of their geographic range. Furthermore, few studies have examined the postdrought recovery potential of drought-susceptible tree species. This study develops a robust ranking of the drought susceptibility of 21 tree species based on their mortality after two droughts (1990s and 2000s) in the savanna of north-eastern Australia. Drought-induced mortality was positively related to species dominance, negatively related to the ratio of postdrought seedlings to adults and had no relationship to the magnitude of extreme drought within the species current geographic ranges. These results suggest that predicting the consequences of exaggerated drought on species' geographic ranges is difficult, but that dominant species like Eucalyptus with relatively slow rates of population recovery and dispersal are the most susceptible. The implications for savanna ecosystems are lower tree densities and basal area.

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.

The rarity of native grasslands in agricultural districts heightens the requirement for optimal management to maintain diversity. Previous studies have suggested that disturbance is required to maintain species diversity in temperate Australian grasslands, but grasslands in semiarid environments do not have the same disturbance requirement. The current study examines the short-term responses to disturbance of subtropical grassland of the Darling Downs, south-eastern Queensland. We also compare temperate and subtropical grasslands in terms of biomass and rainfall. A field experiment was established with treatments, including burning in 2013, burning in 2014, burning in both years, mowing in both years, mowing and raking in both years, and an undisturbed control. Treatments were replicated at each of seven sampling stations in similar environments. The initial sampling after 2013 followed a wet summer and the final sampling was in 2015 after a dry summer. Non-metric multi-dimensional scaling showed that environmental differences, including silt content, soil pH, waterlogging and rainfall history, had more effect on the variation in species composition than did the treatments. The treatments engendered no significant response in species diversity. Of 51 widespread species, only four had a significant change in abundance in response to treatment. Herbaceous biomass was higher in temperate than subtropical grassland after a dry period. The grassland sward may be more open in the subtropics than in temperate grassland because of higher decomposition rates. A comparison of rainfall distribution between subtropical grassland and temperate grassland indicated that droughts are much more frequent in the former environments. These occasional droughts may provide a stress that reduces perennial grass cover, supplanting the requirement for grazing or fire to maintain plant diversity in grasslands. The management of grassland remnants in the subtropics, therefore, seems straightforward because there is little response in species richness or composition to disturbance. However, soil disturbance should be avoided to ensure that exotic species do not proliferate.

At fine spatial scales, savanna-rainforest-grassland boundary dynamics are thought to be mediated by the interplay between fire, vegetation and soil feedbacks. These processes were investigated by quantifying tree species composition, the light environment, quantities and flammability of fuels, bark thickness, and soil conditions across stable and dynamic rainforest boundaries that adjoin grassland and eucalypt savanna in the highlands of the Bunya Mountains, southeast Queensland, Australia. The size class distribution of savanna and rainforest stems was indicative of the encroachment of rainforest species into savanna and grassland. Increasing dominance of rainforest trees corresponds to an increase in woody canopy cover, the dominance of litter fuels (woody debris and leaf), and decline in grass occurrence. There is marked difference in litter and grass fuel flammability and this result is largely an influence of strongly dissimilar fuel bulk densities. Relative bark thickness, a measure of stem fire resistance, was found to be generally greater in savanna species when compared to that of rainforest species, with notable exceptions being the conifers Araucaria bidwillii and Araucaria cunninghamii. A transect study of soil nutrients across one dynamic rainforestgrassland boundary indicated the mass of carbon and nitrogen, but not phosphorus, increased across the successional gradient. Soil carbon turnover time is shortest in stable rainforest, intermediate in dynamic rainforest and longest in grassland highlighting nutrient cycling differentiation. We conclude that the general absence of fire in the Bunya Mountains, due to a divergence from traditional Aboriginal burning practices, has allowed for the encroachment of fire-sensitive rainforest species into the flammable biomes of this landscape. Rainforest invasion is likely to have reduced fire risk via changes to fuel composition and microclimatic conditions, and this feedback will be reinforced by altered nutrient cycling. The mechanics of the feedbacks here identified are discussed in terms of landscape change theory.Austral Ecology (2017) 42, 154-164 Grassland invaded by young forest, typically including dense stands of Acacia and rain forest species but also minor areas of Eucalyptus spp. Woody cover is >30% Rainforest (stable rainforest)Closed forest without eucalypts in the canopy