In mesic regions, patchy spatial vegetation patterns arise through the interplay of succession and disturbances. In the past, this has been shown in many studies of patch dynamics for mesic forests and grasslands, for example, summarized in the classic book by S. T. A. Pickett and P. S. White. However, patchy structures are also frequently observed in water-limited arid and semiarid ecosystems, where distinct banded and spotted vegetation patterns are often found. There are numerous hypotheses on the origin of the distinctive patterning. For example, banded vegetation may be a remnant of more complete vegetation cover diminished by climatic deterioration or by grazing disturbance. Other hypotheses assume these patterns to be natural with downslope water re-allocation from bare areas to vegetated bands as a key-process or are based solely on the intrinsic dynamics of the vegetation without slope-induced anisotropy.
A number of cellular simulation models were developed to investigate the robustness and origin of such patterns. For example, end of the 1990s, D. L. Dunkerley modeled banded vegetation communities in western New South Wales Australian grassland and shrublands to test the hypothesis that water partitioning in spatially unstructured plant communities may lead to the development ofbanding. The model shows that without any climatic change or external disturbance, strongly developed banding can emerge from an initially random distribution of plants.
Other models used differential equations to search for possible unifying mechanisms to explain these spatial patterns. One hypothesis is that spatial patterns establish themselves through a Turing-like spatial instability depending only on a tradeoff between facilitative and competitive interactions among plants. This hypothesis goes back to the 1950s where A. M. Turing described morphogenesis in chemical systems. These models produce patterns superficially similar to banded and spotted vegetation, which then is taken as evidence for the validity of the underlying hypothesis. However, reproduction of a pattern is not proof that the modeled processes represent the natural processes. Since these top-down models are usually not explicitly related to specific spatial and temporal scales, they are difficult to test and their ecological content remains unclear.
Was this article helpful?