@article{BauerVosKlauschiesetal.2014,
  author    = {Bauer, Barbara and Vos, Matthijs and Klauschies, Toni and Gaedke, Ursula},
  title     = {Diversity, functional similarity, and top-down control drive synchronization and the reliability of ecosystem function},
  series = {The American naturalist : a bi-monthly journal devoted to the advancement and correlation of the biological sciences},
  volume    = {183},
  journal   = {The American naturalist : a bi-monthly journal devoted to the advancement and correlation of the biological sciences},
  number    = {3},
  publisher = {Univ. of Chicago Press},
  address   = {Chicago},
  issn      = {0003-0147},
  doi       = {10.1086/674906},
  pages     = {394 -- 409},
  year      = {2014},
  abstract  = {The concept that diversity promotes reliability of ecosystem function depends on the pattern that community-level biomass shows lower temporal variability than species-level biomasses. However, this pattern is not universal, as it relies on compensatory or independent species dynamics. When in contrast within--trophic level synchronization occurs, variability of community biomass will approach population-level variability. Current knowledge fails to integrate how species richness, functional distance between species, and the relative importance of predation and competition combine to drive synchronization at different trophic levels. Here we clarify these mechanisms. Intense competition promotes compensatory dynamics in prey, but predators may at the same time increasingly synchronize, under increasing species richness and functional similarity. In contrast, predators and prey both show perfect synchronization under strong top-down control, which is promoted by a combination of low functional distance and high net growth potential of predators. Under such conditions, community-level biomass variability peaks, with major negative consequences for reliability of ecosystem function.},
  language  = {en}
}
@article{MartinJagerNisbetetal.2014,
  author    = {Martin, Benjamin and Jager, Tjalling and Nisbet, Roger M. and Preuss, Thomas G. and Grimm, Volker},
  title     = {Limitations of extrapolating toxic effects on reproduction to the population level},
  series = {Ecological applications : a publication of the Ecological Society of America},
  volume    = {24},
  journal   = {Ecological applications : a publication of the Ecological Society of America},
  number    = {8},
  publisher = {Wiley},
  address   = {Washington},
  issn      = {1051-0761},
  doi       = {10.1890/14-0656.1},
  pages     = {1972 -- 1983},
  year      = {2014},
  abstract  = {For the ecological risk assessment of toxic chemicals, standardized tests on individuals are often used as proxies for population-level effects. Here, we address the utility of one commonly used metric, reproductive output, as a proxy for population-level effects. Because reproduction integrates the outcome of many interacting processes (e.g., feeding, growth, allocation of energy to reproduction), the observed toxic effects in a reproduction test could be due to stress on one of many processes. Although this makes reproduction a robust endpoint for detecting stress, it may mask important population-level consequences if the different physiological processes stress affects are associated with different feedback mechanisms at the population level. We therefore evaluated how an observed reduction in reproduction found in a standard reproduction test translates to effects at the population level if it is caused by hypothetical toxicants affecting different physiological processes (physiological modes of action; PMoA). For this we used two consumer-resource models: the Yodzis-Innes (YI) model, which is mathematically tractable, but requires strong assumptions of energetic equivalence among individuals as they progress through ontogeny, and an individual-based implementation of dynamic energy budget theory (DEB-IBM), which relaxes these assumptions at the expense of tractability. We identified two important feedback mechanisms controlling the link between individual- and population-level stress in the YI model. These mechanisms turned out to also be important for interpreting some of the individual-based model results; for two PMoAs, they determined the population response to stress in both models. In contrast, others stress types involved more complex feedbacks, because they asymmetrically stressed the production efficiency of reproduction and somatic growth. The feedbacks associated with different PMoAs drastically altered the link between individual- and population-level effects. For example, hypothetical stressors with different PMoAs that had equal effects on reproduction had effects ranging from a negligible decline in biomass to population extinction. Thus, reproduction tests alone are of little use for extrapolating toxicity to the population level, but we showed that the ecological relevance of standard tests could easily be improved if growth is measured along with reproduction.},
  language  = {en}
}