Statistical analysis of the spatial symmetry of periodic vegetation patterns in semi-arid lands

Abstract

The existence of spatially periodic distributions of plants in arid and semi-arid areas is a topic of ongoing research. Many biological models have been developed to try to explain this phenomenon. Recent models predict the following sequence of patterns for increasing aridity: gaps of bare ground in a dense vegetation cover, alternating stripes of vegetation and bare ground and spots of vegetation surrounded by bare ground. The gaps and spots are arranged according to a hexagonal symmetry. This statement has been investigated through the use of spatial statistics. In addition to the near-hexagonal symmetry hypothesis, other near-regular symmetries were also investigated (rhomboidal, square and triangular). The statistical analysis was performed on a gapped vegetation pattern in Kenya. The Complete Spatial Randomness hypothesis was rejected in favour of the regularity alternative. The rhomboidal with angle 52° was the most likely point process for the observed data.