Data from: Dispersal in dendritic networks: ecological consequences on the spatial distribution of population densities

1. Understanding the consequences of spatial structure on ecological dynamics is a central theme in ecology. Recently, research has recognized the relevance of river and river-analogue network structures, because these systems are not only highly diverse but also rapidly changing due to habitat modifications or species invasions. 2. Much of the previous work on ecological and evolutionary dynamics in metapopulations and metacommunities in dendritic river networks has been either using comparative approaches or was purely theoretical. However, the use of microcosm experiments provides the unique opportunity to study large-scale questions in a causal and experimental framework. 3. We conducted replicated microcosm experiments, in which we manipulated the spatially explicit network configuration of a landscape and addressed how linear versus dendritic connectivity affects population dynamics, specifically the spatial distribution of population densities, and movement behavior of the protist model organism Tetrahymena pyriformis. We tracked population densities and individual-level movement behavior of thousands of individuals over time. 4. At the end of the experiment, we found more variable population densities between patches in dendritic networks compared to linear networks, as predicted by theory. Specifically, in dendritic networks, population densities were higher at nodes that connected to headwaters compared to the headwaters themselves and to more central nodes in the network. These differences follow theoretical predictions and emerged from the different network topologies per se. These differences in population densities emerged despite weakly density-dependent movement. 5. We show that differences in network structure alone can cause characteristic spatial variation in population densities. While such differences have been postulated by theoretical work and are the underlying precondition for differential dispersal evolution in heterogeneous networks, our results may be the first experimental demonstration thereof. Furthermore, these population-level dynamics may affect extinction risks and can upscale to previously shown metacommunity level diversity dynamics. Given that many species in natural river systems exhibit strong spatio-temporal patterns in population densities, our work suggests that abundance patterns should not only be addressed from a local environmental perspective, but may be the outcome of processes that are inherently driven by the respective habitat network structure