Global change includes a multitude of different factors and many of them are known to be able to affect plants and plant communities, or other ecosystem components, on their own. It is undeniable that the knowledge about such single-factor effects, based on experiments focusing on just one or two factors, is fundamental when trying to understand or to predict ecological consequences of global change. Most likely, however, such single-factor effects are only a part of a bigger picture, which is global change as a multifactorial process. In nature, multiple factors often cooccur in space and time, potentially leading to interactive effects which might not be anticipated from the known effects of individual factors.
In early 2020, I joined the ecology lab of Mark van Kleunen, looking for an exciting topic for my master thesis. At that time, Mark and Rutger Wilschut, back then a postdoc in the group, already had an idea. A bit earlier, in 2019, a very interesting and inspiring study by Matthias Rillig and colleagues was published, which, for the first time, explicitly addressed the multifactorial nature of global change itself as a potential driver of ecological consequences, focusing on soil functioning and microbial diversity. So, when Mark and Rutger suggested that I could do a similar project to investigate how global change as multifactorial process could affect plant communities, I knew that I found the topic I was looking for. In the paper, Rillig and colleagues presented a practical approach to circumvent a central hitch of multifactorial experiments, which is the “combinatorial explosion problem”. That is, with increasing numbers of factors, the number of potential factor combinations increases exponentially. So, in our case, for a six-factorial experiment, we would have ended up with at least 26 = 64 treatment combinations, which, despite our fondness for large experiments, would have exceeded our capacities. Instead, we decided to go for the simple but elegant approach presented by Rillig and colleagues, which basically transfers the design of classic biodiversity studies to global change research. Based on this approach, we established a pool of six global change factors that cooccur frequently (warming, soil salinization, eutrophication, fungicide accumulation and microplastic and light pollution) from which we created four factor-richness levels (0, 1, 2, 4, 6).
In July 2020, we set up the experimental plant communities, for which we selected nine widespread herbaceous grassland species which are native to Central Europe. We conducted the experiment in the climate-change facility of the botanical garden of the University of Konstanz during summer 2020, which, similar to most summers in recent years, turned out to be exceptionally warm. Consequently, not just the communities in the artificial warming treatment, but we as well were directly exposed to climate warming, which made the regular visits to water the plants and to apply and adjust the eutrophication and salinity treatment a rather sweaty business, not to mention the randomization of the pots. In turn, visits during night to control the light pollution setup were quite enjoyable. Unfortunately, the summer went by too quickly, and so did the experiment. Yet, after a brief mourning phase, the first results seemed to be quite exciting and helped to get mentally prepared for the upcoming gloomy winter at Lake Constance. The promising initial findings were supported by further analyses, so we could finally start writing our story for publication.
We found that both community productivity and composition (diversity and evenness) were affected by factor richness, i.e. the number of simultaneously acting global change factors. While the productivity was affected positively, the diversity of the communities was reduced by higher factor richness, mainly due to lower evenness. Interestingly, also the mechanisms driving the observed patterns seem to differ. The positive factor-richness effect on community productivity is likely due to sampling effects, as higher factor-richness levels included eutrophication more frequently. In contrast, the negative factor-richness effects on community evenness and diversity could not be explained by such sampling effects, but were driven by the number of co-acting global change factors itself. Hence, our results indicate that the consequences of global change as multifactorial process on plant communities are unlikely to be predictable based on the effects of individual factors alone, as overall effects can be shaped by interactions between co-acting global change factors. In other words, the overall impact of global change most likely is more than just the sum of its parts.
Read the full story at: https://www.nature.com/articles/s41467-022-35473-1.