Bee collections – what they tell about plant-pollinator interactions
Bee collections – plenty of dead, dried bees on a needle in a box – may seem an absurd thing to some people. But they’re actually extremely useful and important for research and conservation. Wait – dead animals important for conservation? Yes, they are. To begin with my personal story: I discovered my love for biodiversity with the bee collection of the University in Concepción, Chile. I realized how many bee species exist and how beautiful and fascinating they are. I trained what I call “my eye for habitus” on this collection. This means to recognize the important features that separate one species from the other. It turns into a “diagnostic eye”, because knowing the usual appearance, it’s easier to recognize if something is wrong. So the hours spent between cupboards full of bees 24 years ago were an important step to arrive where I’m now.
Collections are also the basis of many research questions, or, as the Natural History Museum in London puts it:
The scientific community is using the collection to answer key questions about the past, present and future of the solar system, the geology of our planet and life on Earth.
I already wrote several times why diversity is crucial. There’s no “one fits it all” for pollinating plants. We also talked about pollinator decline. It’s not easy to specify the consequences of it. Nor to measure its dimension. In this area, collections can play an important role.
Bee collections and how they help to understand plant-pollinator interactions
Only recently I read a paper from Burkle and colleagues in Science. It’s a great example of how collections in museums and universities can help to understand ecological contexts. The authors compared spring-flowering plants in a forest in Illinois, USA and their interactions with bees in the 1880s, the early 1970s and nowadays. As a part of their work, they examined if the quality of pollination had changed for a single plant. For this, they identified pollen grains on the bodies of preserved bees – i.e. in collections. These were six mining bee species (Andrena spp.), which are the most important pollinators of the springbeauty (Claytonia virginica), one of the flowers in the studied forest.
Comparing bees that were caught visiting springbeauty, they noticed that the proportion of pollen grains of this plants on the bee bodies declined over the past 40 years. This is important for the quality of the pollination of a plant: if a bee flies to another plant species after visiting the flowers of C. virginica, neither of both gets the pollination service it’s aiming for. Bees in general show flower steadiness or fidelity to a certain extent. They stay with a single species for a while and then change to the next. This apparently in the 1880s and 1970s happened more often then in 2009/2010, the time of the most recent study.
Documented pollinator decline on a local scale
This is only a small part of this interesting work. Burkle and her colleagues studied 26 plant species and their 109 pollinating bees. In addition to the collections, they also had data sets from researchers in the 1880s and 1970 which included phenology of plants and bees. The older records were compared with data from 2009 and 2010, analysing how and why the interactions between flowers and their pollinators changed. The results aren’t very encouraging: only 24% of the interactions observed in the 1880s were still intact. There were also shifts in the network structure, expressed by 121 new interactions that weren’t observed in the older data set.
The most concerning result may be, that species extinction was the main reason for this loss of interactions. Though all 26 plant species still grow in this forest in Illinois, the number of bee species decreased from 109 to 54. Most affected were specialist bees and cuckoo bees. Cavity nesting species, like those you observe in your bee hotels, were lost more often than ground nesting bees. In addition, the formerly continuous woods were now fragmented. This also decreased the interactions between flowers and bees. Together with phenological mismatches all these factors weaken the pollinator-plant community, it gets more vulnerable. Future disturbances will affect the system even worse.
The value of historical data
This paper shows the huge treasure bee collections, or any collection, are. Each specimen gives information where the species occurred, when it was active and maybe also on which plants it foraged. This may not only document pollinator decline, but also give hints on what factors led to it at a local or regional scale. The challenge may be to connect data from entomological and botanical collections, meteorological measurements and land use archives. This is much more than just describing the decline. By understanding the single factors that caused it, it’s easier to develop measures to avoid or mitigate these factors.
In addition, this amazing study (read it, I discussed only a small part of it!) also shows that “old” literature and data are very valuable. It happened already a few times that after a talk somebody asked why I presented data that were already “old”. In general referring to data from about five years ago. Then I always explain that scientific data aren’t technical devices that are outdated after a year. The data an entomologist recorded in 1880 are not a cell phone. They still have value because they give a point of reference. They make us understand how stable or vulnerable plant-pollinator relationships are. Or to cite the authors of this paper:
Several authors have speculated about how changes in biodiversity and phenology might translate into changes in the structure and stability of complex interaction networks. However, there has been a lack of historical data on plant-pollinator networks and phenologies for both plants and insects in the same community.
We need much more work like this.