Changing world, changing interactions

When I think of how a species might be affected by global change, I tend to focus first on the abiotic: distribution maps predicting species range shifts based on temperature and precipitation come to mind. But obviously it’s more complicated than that—global change is bound to affect biotic factors as well, particularly interactions between species. This makes predicting the consequences of global change more complicated. If species A interacts with species B and species A moves up in elevation and species B starts emerging later in the spring, what happens to the interaction between the two?


Add in all the other possible interactions between species in a community, and you get a complicated network of biotic interactions that are susceptible to change.

Plant-pollinator systems provide a good example of how global change can affect species interactions. We know that plants can exhibit changes in phenology, physiology, and abundance in response to altered temperatures and precipitation regimes, habitat loss, and other factors. Pollinators respond to these changes too, but not always with the same magnitude, which can disrupt plant-pollinator interactions. Often within a given community there are both specialist and generalist pollinators, which leads to a network of plant-pollinator interactions. This leads to the question of whether a network of plant-pollinator interactions will be particularly susceptible to global change, given the sensitivity of plants and pollinators to environmental change, or whether the structure of a network will be resilient to change because of redundancies in plant-pollinator interactions within the network.

Enter Burkle et al. (2013) with a study of disruptions to plant-pollinator interactions over 120 years in Illinois forests. Between the late 1800s and 2010 these forests experienced habitat alteration (mostly conversion from forest to agriculture) and 2⁰C warming in winter and spring. The authors compared data on forb-bee interactions and phenologies collected in the late 1800s to data collected from the same area in 2009-2010.

What they found:

Considerable shifts in overall plant-pollinator network structure. Only 24% of the forb-bee interactions observed in the late 1800s were present in 2009-2010. However, there were novel forb-bee interactions in the contemporary data, meaning only an absolute loss of 46% of interactions over time.

Bee extirpations explain a lot of this. Although all 26 focal forb species remained present in the study area over time, many bee species were extirpated, with specialists more vulnerable to extirpation than generalists.

Fragmentation and phenological mismatches explain the rest. Some lost interactions were accounted for by lack of spatial co-occurrence between forbs and bees, suggesting habitat fragmentation played a role. Other interaction losses were          explained by phenological mismatches (ie peak plant bloom was 9.5 days earlier in the contemporary data but peak bee activity was 11 days earlier).

Changes in the network structure have made it more vulnerable to future changes. Bee extirpations, interaction losses, and interaction shifts have decreased redundancies within the network. In other words, where there used to be many bee species pollinating a single forb species, there are now fewer. The fewer bee species pollinating a plant, the more likely that plant will be to lose its pollinators.

Why this paper is cool:

For one, I like this paper because I’m kind of a sucker for historical data sets. I think the authors elegantly coupled historical data with their own contemporary data to show change in pollinator networks over time. By focusing on the whole network of plant-pollinator interactions and linking it to environmental variables, the authors do a good job of showing how anthropogenic change is affecting species interactions.

Tagged , , . Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *