Vegetation Mosaics of Arid Western New South Wales, Australia: Considerations of Their Origin and Persistence

The drylands of Australia contain areas of striking and spatially extensive patterned vegetation. The difficulty in understanding emergent structures in dryland eco-geomorphology springs in part from a fundamental challenge of separating function or operation in a system once fully-developed from the processes that drive the initial emergence of patterning. There is an additional challenge over much of the Australian drylands, connected with the very marked inter-annual variability in rainfall, which drives large swings in soil moisture, plant cover, and floristics, as well as in herbivore grazing pressure, abundance of soil biota, and other factors. There are reasons to believe that the rainfall variability, including particularly the ecosystem impacts of the dry years lying near the extremes of the annual series, is important to the emergence of vegetation and soil patterning. In the study area of western New South Wales discussed in this chapter, soil characteristics are tightly linked with the evolution of vegetation patterning. Diverse kinds of vegetation mosaics are present in the study area, ranging from gilgai-like depressions and mounds, in which plants surround the moist depressions, to irregular patches or groves lying within bare soil or stone mantled soil, to strikingly regular, contour-aligned vegetation banding. Groves are favourable locations for plant growth because of the combined role of the enhanced infiltrability generated within groves, and the increased soil water availability that is created by the crabholes and other shallow closed depressions. Field data suggest that crabholes and closed depressions are more likely to account for the presence of shrubs than the reverse. The production of overland flow in the intergroves within patterned vegetation can be remarkably efficient. Consequently, groves are not uniform hydrologically, and the upslope margins receive runon water more frequently than the lower parts of the groves. For the field context of western NSW, existing models fail to touch upon key mechanisms that appear to be pivotal to the development and operation of emergent vegetation patterns, including soil shrink-swell and spatial differences in soil shear strength. In order to explore in a general way the development of vegetation banding of the kind seen in western NSW, a cellular automaton (CA) model was employed. Crabhole formation during very dry years and runs of years appears to be a very important aspect of the mechanisms that create and sustain the banded mosaic vegetation communities of the study area. Thus, it may be that soil pattern development is the leading process, and the emergence of patterned vegetation is an effect consequent upon the soil behaviour. It is likely, also, that the vegetation patterning establishes a feedback process to further strengthen the hydrologic compartmentalisation resulting from the soil collapse processes.