Small-scale spatial pattern and dynamics of experimental plant communities

Plant-plant interference is inherently local and seed dispersal generally limited. Both
processes generate spatial and genetic structure within plant populations and communities that
need to be better understood in order to predict dynamic community changes due for example
to biodiversity loss or global change. There is increasingly strong theoretical evidence that
spatial pattern is an essential factor controlling the species dynamics of many communities. In
particular, one conclusion from spatial models is that intraspecific aggregation promotes
coexistence by slowing down competitive exclusion. Whereas local interactions contribute to
interspecific segregation, limited seed dispersal leads to aggregation at two hierarchical
levels: i) species within communities and ii) genetically related individuals (e.g. siblings)
within populations. However, especially for plant communities there is a need for
experimental tests of the predictions generated from spatial models.
The principal goal of this thesis was to narrow the gap between theoretical and
empirical investigations on the role of spatial pattern in plant communities and population
dynamics. I focused on the effects of spatial pattern on the dynamics of experimental plant
communities at the level of species as well as at the level of genotypes within species. In
particular, I (i) manipulated the spatial pattern, i.e. the relative frequency of intra- vs.
interspecific contacts and (ii) contrasted the performance of genetically related (half-sibs) vs.
non-related individuals. The basic goal of the experiments was to investigate whether
different spatial patterns (random vs. aggregated) and relatedness of neighbors had any effects
on population dynamics within experimental plant communities.
The experiments provided interesting results and showed essential aspects of the role
of intraspecific aggregation and sibling interference in regulating the dynamics of populations
within experimental plant communities. I showed that weak competitors increased their
fitness (e.g. biomass and seed production) when grown in neighborhoods of conspecifics
compared to neighborhoods of heterospecifics, at least in the short run. The data further
suggested that the advantages of intraspecific aggregation for weaker competitors might be
independent of the species identity and that all other species are best avoided.
An additional aggregation at the level of genotypes (e.g. seed families) suggested speciesspecific
effects linked with seed size. For instance, I found negative sibling competition
effects for the small-seeded species (Capsella), while rather positive effects for the largeseeded
species (Stachys). Negative effects of sibling competition were also observed among
relatives of sunflower seed families. By contrast, genetically similar individuals of the dimorphic species Senecio jacobaea increased their fitness (e.g. biomass) compared to
genetically dissimilar individuals. However, also this species suggested seed traits specific
relatedness effects (e.g. dispersal ability). Positive relatedness effects were more evident by
seeds expected to aggregate more locally (without pappus) than by seeds expected to disperse
wider (with pappus). Generally, I observed lower size variation (measured as coefficients of
variation) among related compared to non-related individuals. This might be a consequence of
more genetic uniformity and / or kin selection among relatives compared to non-relatives.
Although, I could not provide strong evidence for sibling competition or kin selection, I
believe that relatedness among plants, especially for species with highly localized dispersal,
should play a considerable role in the regulation of local population dynamics. Similar to the
species level, there must be subtle trade-offs (e.g. between neighbour relatedness and density)
that determine the complicated local dynamics of plant communities. However, the question
under which circumstances and to which extent relatedness effects are species-specific
remains open and deserves further investigation.
At the level of species, effects of intraspecific aggregation on the dynamics of
experimental plant communities were clear and consistent throughout my experiments. By
contrast, at the level of genotypes, they were less clear and to some extent contrasting. This
emphasized the importance for further investigations on population dynamics at levels below
that of species.
From an applied point of view, findings of this thesis might help to give better information for
management practices (e.g. restoring species rich communities). For example, by varying
spatial pattern (random vs. intraspecifically aggregated) of selected species in wildflowers
strips or fallows, the dominance of undesired species (e.g. Dipsacus sp.) and the exclusion of
weaker species can be delayed.