Can single species responses predict community change under global warming?

Submitted by editor on 25 November 2016.Get the paper!

In previous experiments we grew grassland mesocosms in sunlit controlled growth chambers under different climate conditions. We investigated whether grassland species would maintain their current level of stress resistance when the global climate continues to change. We found effects of warming and elevated CO2 on individual grassland species. For example, we found that drought impacts aggravated by global warming and elevated CO2 only partly compensated for the detrimental effect of warming on drought. As the impact of climate change at single species level is well known, the next step is to study plant responses to environmental changes in more natural conditions. In nature, species interact with many other in complex networks. Therefore interactions such as competition and herbivory need to be considered.

Species show differences in their sensitivity to temperature. Such asymmetries in the thermal responses of interacting species can bring about change in consumer-resource dynamics, with important consequences for the dynamics and persistence of populations and communities. As a consequence, theory predicts that community responses cannot be understood or predicted from single species behaviour. Community-scale experiments are thus needed to understand the effects of climate change on community composition and ecosystem functioning.

In this study, we investigated the effect of warming on a simple community consisting of three species: rosy apple aphid feeding on plantain and a heterospecific neighbouring plant species, perennial ryegrass. The aphid does not feed on perennial ryegrass. The experimental design consisted of monocultures and mixtures of perennial ryegrass and plantain at three temperatures levels: 17 °C, 20 °C and 23 °C. These temperatures were chosen to reflect the potential range of increase in the next century, with the lowest temperature corresponding to the average temperature on a summer day in Belgium. The temperature range allows us to compare our results with previous experiments in the growth chambers.

Fig. 1 Environmentally controlled growth chamber used in the experiment.

Our study shows that warming affected the aphid’s performance directly, but not indirectly through changes in host plant quality. Aphid populations at 20 °C were characterised by shorter generation times, faster growth and larger aphids compared to populations at 17 °C. Despite of this, the biomass losses of plantain did not alter under warming, and, as a consequence, the net interaction strength between plants and herbivores did not change under warming. This finding points to reduced consumption rates at a higher temperature. This is in contrast to theoretical studies which predict that ectothermic herbivores must increase their food intake at higher temperatures to compensate for increased metabolic or nutritional demand. Apparently, the effect of temperature on herbivory rates is highly variable, depending on the herbivore-plant combination. In addition, we also found that plant competition affected the aphids, but trough an interaction with temperature. Our study provides evidence that net interactions between plants and herbivores should not change under warming, despite the direct effects of warming on herbivores when plant-plant interactions are considered. The stability of net interaction strength suggests that the response of a simple community to warming may scale up to understand the effect of warming on more complex communities and ecological networks.

Fig. 2 Healthy and affected plant community at the end of the experiment.

The authors through: Helena Van De Velde

Insights into Oikos papers