Seminar:  ``River Networks: from Geomorphic Auto-Organization to Biodiversity Pattern Dynamics''
by Matteo Convertino, University of Padova - Italy and Visiting Researcher  at Princeton University
Tuesday, June 7, 2-3 pm
122 Rogers Hall


Neutral metacommunity models for spatial biodiversity patterns are implemented on river networks acting as ecological corridors at different scales of aggregation. Coarse-graining elevation fields (under the constraint of preserving the basin mean elevation) produce a set of reconfigured drainage networks.

The hydrologic assumption made implies uniform runoff production such that each link has the same habitat capacity. Despite the universal scaling properties shown by river basins regardless of size, climate, vegetation or exposed lithology, we find that species richness at local and regional scales exhibits resolution-dependent behavior. The slopes of the species-area relationships, which are consistent over coarse-graining resolutions, match those found in real landscapes in the case of long-distance dispersal. The rank-abundance patterns are independent on the resolution over a broad range of dispersal length. The results confirm that strong interactions occur between network structure and the dispersal of species, and that, under the assumption of neutral dynamics, these interactions produce resolution-dependent biodiversity patterns that diverge from expectations following from universal geomorphic scaling laws. Both in theoretical and applied ecology studying how patterns change in resolution is relevant for understanding how ecological dynamics work in fragmented landscape, and for sampling and biodiversity management campaigns especially in consideration of climate change. 

Moving from the analysis of the distribution of species on real data (fish and vegetation) of the Mississippi Missouri River System, the clustering of species has been studied on real and artificial networks vs. a 2D non-fragmented landscape with a neutral metacommunity model. The probability of exceedence of the cluster-size is a power-law whose slope is quite the same both for fishes and big trees. The slope of the power-law of the cluster size seems strongly affected only by the dispersal kernel, rather by the shape of the basin or the nature of the species. 

Insights are given in a probabilistic description about the geomorphic structure of river basins, e.g. for the distance between tributaries and the unchanneled paths.  A general theory emerges on the effects of dendritic geometries on the dynamics operating on river basins. Insights provided by such a theory will lend themselves to issues of great practical importance such as integration of riparian systems into large-scale resource management, spatial strategies to minimize loss of biodiversity, climate change effects on species distribution, and effective prevention campaigns against water-borne diseases (like cholera).