Topology matters: Network topology affects outcomes from community ecology neutral models

Abstract

Abstract. Topology affects outcomes of processes in planar networks. Hexagon tessellations have different performance than, and are often superior to, square tessellations in applications such as fluid dynamics, percolation theory, self-avoiding walks, survey sample design, and quantization. Hexagon tessellations and square tessellations using both von Neumann and Moore neighborhoods were examined as a network topology for neutral ecology community models, following the simulation approach of Graham Bell. These models, which assume identical life history and movement properties for each individual of each species, produce collective properties of communities, such as abundance distributions, range distributions, spatial variation in abundance, species-area curves, and spatial variation in species composition, that match many empirical patterns. The simulations in this study varied the dispersal rate but kept birth, death, and immigration rates constant. For these experiments, ending community populations, species richness, Shannon diversity, and Simpson diversity were clearly different for the different topologies, but the relationship between the topologies varied as the dispersal rate changed. Empirical distributions of the performance measures also showed clear differences among topologies. The interaction of topology with dispersal, spatial boundary effects, and other parameters of these models appears to be quite complex and warrants further research.

Topological explanations and robustness in biological sciences

“…Many of these studies have been done by simula- Fig. 3 Topologies of neighborhood used to simulate neutral ecological models (after White and Kiester 2008) tion. Recently, White and Kiester 2008 paid attention to the topology of the grid onto which one designs the simulation by cellular automata.…”

The influence of planning unit characteristics on the efficiency and spatial pattern of systematic conservation planning assessments

“…We also found that using hexagons produced more efficient results when compared with squares, and this improvement ranged between 0.2 and 1.8% reductions in portfolio extent. This is because hexagons can be arranged in a more spatially compact manner and so need fewer superfluous planning units to form patches containing the important conservation features (Culver et al 2004;Birch 2006;White and Kiester 2008). Using hexagons also generally produced portfolios that consisted of fewer smaller patches and there were similar trends when using larger planning units.…”

Eco-ISEA3H, a machine learning ready spatial database for ecometric and species distribution modeling

“…Differing grid topologies affect the results of ecological models simulating dispersal. White and Kiester 48 , for example, found the topology of the network of communities in a neutral community ecology model - in which simulated communities had hexagonal neighborhoods, or von Neumann, Moore, or Margolus neighborhoods - affected modeled species abundances and diversities, but in complex ways, which differed given different model parameter values. (Note that the four neighbors with which a square cell shares an edge are termed its rook , or von Neumann neighborhood, and these plus the four neighbors with which it shares a single vertex its queen , or Moore neighborhood.)…”