Does low temperature cause the dominance of Acacia on the central Australian mountains? Evidence from a latitudinal gradient from 11° to 26° South in the Northern Territory, Australia

A latitudinal gradient, from the central north coast of the Northern Territory (11oS) to the South Australian State border (26oS), was defined to subsample a large 20 mx20 m quadrat data set (N>2000 quadrats) collected during the course of the Northern Territory 1:106 vegetation mapping programme. The mean and standard error of a range of environmental data, and structural and floristic variables pertaining to woody species were calculated for fifteen cells (3.5oE wide and 1oS long) on the transect using a total of 1050 quadrats. It was found that the interrelated measures of mean canopy height, mean canopy cover and mean total basal area steadily declined from the north coast to reach their minimum levels at 18.5oS. There was little variation in these variables south of this latitude. This pattern is probably controlled by precipitation given that there is a highly significant (r2=98%) negative exponential decay of mean annual rainfall with latitude for ten meteorological stations on or near the transect, and that the southern limit of the summer monsoon rains is at about 18oS. The mean percentage of the woody species quadrat richness attributable to Acacia species was found to increase at around 18.5oS. However, the mean Acacia basal area and the percentage of the total basal area composed of Acacia basal area increased at 21.5oS. At this latitude the mean Eucalyptus basal area, the percentage of the total basal area made up of Eucalyptus species, and the mean percentage of woody species quadrat richness composed of Eucalyptus species all decreased to minimum levels. South of 21.5oS mean landscape elevation ranges between 400 and 700 m above sea level while north of this latitude mean landscape elevation ranges 10-300 m above sea level. The combined effects of continentality and environmental lapse rates result in a highly significant (r2=82%) negative exponential decay of mean July (winter) minimum temperature with latitude for the five climate stations on the transect for which data are available. Mean slope angle, rock cover, surface soil gravel content and surface soil clay content were unrelated to any of the above patterns. It is suggested that the sharp change-over in Eucalyptus to Acacia dominance at 21.5oS is related to minimum temperature, but this hypothesis requires testing with detailed ecophysiological studies. None the less, the local dominance of Acacia shirleyi at 16.5oS suggests that environmental history also may have a strong influence on the contemporary latitudinal distribution of Acacia and Eucalyptus in the northern half of the Australian continent. Regrettably, few data are available to evaluate critically the importance of long-term environmental change on current distributional patterns of Acacia and Eucalyptus.