Paper of the day: Pyke et al. 2016, “Effects of climate change on phenologies and distributions of bumble bees and the plants they visit”

This paper comes from an open-access journal, Ecosphere! Woohoo! You can read the whole paper yourself for free here: Effects of climate change on phenologies and distributions of bumble bees and the plants they visit I encourage you to read both the scientific paper and my summary to get the most complete picture of the story – this paper is more comprehensible than most!

Shameless plug for native bees:

Honey bees get a lot of attention from the media because they help to pollinate the food we eat, but in terms of their conservation, agriculture is really the only reason we would focus on protecting honey bees. They are actually a non-native species from Europe, are not in any danger of dying out as a species, and in the wild, they often outcompete native bee species for floral resources, potentially causing harm to ecosystems overall. The following paper is about native or naturalized bumblebee species and describes one of the problems native bees are facing in nature due to climate change. Read on to learn more!

Background info

In this paper, the authors are focusing on bee and plant phenology, which is the timing of life events like birth, death, and peak activity, and in plants specifically, events like sprouting, budding, flowering, fruiting, and senescing (pre-programmed partial or complete death of plant parts, like when oak trees shed their leaves in the fall). Bees and plants interact with each other in a mutualistic relationship, meaning they help each other out with functions necessary to survival. This occurs primarily through pollination, whereby bees collect nectar and pollen to eat and feed their young, and the plants receive fertilization so they can produce successful offspring. But this mutualism can only occur when the bees and plants are operating on the same timeline – bees need to be buzzing around searching for food at the right time of the season, namely, when flowers are open on plants, for pollination to proceed. With changes in climate, bees and plants may need to shift their phenologies to coincide with optimal weather conditions – but a big question in ecology right now is, will they shift their phenologies in the same way so that they stay matched up with each other? This potential phenological mismatch between bees and plants, or asynchrony, is what Pyke and his colleagues explore in this paper.


  • Organisms are expected to respond to changing climate by shifting geographical ranges and phenology toward remaining in their compatible climate zones
  • Such changes may result in spatial or temporal mismatches between interacting species (asynchrony)
  • This study examines mismatches arising from climate-induced shifts in plants and their pollinators
  • Surveyed bumble bees and the plants they visit in 1974 and 2007 at the Rocky Mountain Biological Laboratory in Colorado (33 years)
  • Tested hypotheses arising from observed climate change

The Rocky Mountain Biological Laboratory (RMBL) has been providing scientists with a natural outdoor laboratory since 1928. Many famous studies have been conducted there, and an author on the paper, David Inouye (below), has spent summers performing experiments on pollination ecology at RMBL since 1971.

RMBL is a great place to study shifts in phenology due to climate change, because changes in temperature occur very rapidly with changes in altitude on the slopes of mountains. In the 33-year intervening period between their two study years, for example, monthly spring and summer temperatures have increased 2 degrees C on average, which means that for plants and bees to experience the same temperature conditions they had in 1974, they must move 317 meters up the mountain by 2007.

One of the authors, David Inouye, at the Rocky Mountain Biological Laboratory in Colorado


  • H1: Species distributions have shifted upwards by about 317 m to match the change in temperature with elevation
  • H2: Phenologies have shifted earlier in the season, but not identically, resulting in asynchrony (mismatching)
  • H3: Bumble bee abundance was lower in 2007 than in 1974 (due to asynchrony)


  • Study encompassed an elevation range of 1000 m and spatial range of 16 km
  • 47 study sites – dominated by grasses and herbaceous plants
  • Surveyed every 8 days (1974) or 6 days (2007)
  • Transects or “circle sites” (covering circular areas on roadsides)
  • ID’d bumble bees observed and flowers visited
  • 12 perennial plant species ID’d as important to bumble bees (represented 74.5% of visits)
  • 8 bumblebee species (represented 97% of bumble bees observed)


Did bees shift up?

  • H1A: Bees shifted up but not necessarily 317 m

Did plants shift up?

  • H1B: Plant species did not show significant change in upwards distribution (expect one species)
Delphinium barbeyi

Did bee phenology shift earlier?

  • H2A: Phenological differences were partially consistent with hypotheses
  • Workers shifted in peak recording rate ~17 days earlier in transect sites

Did flowering phenology shift earlier?

  • H2B: Flowering phenology was significantly earlier in 2007 compared with 1974

Was there asynchrony (mismatching) between bees and plants?

  • H2C: Found expected reduction in synchrony between bees & plants
    • However, the authors assume that the bees and plants were synchronous in the first place in 1974

Was there lower bee abundance in 2007?

  • H3: Found lower bee abundance in 2007 compared with 1974

Summary of results

  • Shifts towards higher elevations for most bumble bee species, but not for most plant species
  • Phenology shifted earlier for plants but not bees
  • Bees and plants were mismatched in 2007
  • Bee abundance was lower in 2007

Discussion points

We discussed this paper in a seminar on the phenological consequences of climate change that I’m auditing this quarter at UC Davis. Here are some of the points our discussion centered around:

  • Only 2 years of data – are their conclusions justified?
  • Why are upward shifts in bees inconsistent with the expectation (317 m)?
  • Their surveys in 2007 ended before bumble bee workers declined, making it difficult to accurately estimate dates for peak recording rates – how might this affect their results?
  • Can’t draw a causal link between reduced synchrony & reduced bee abundance, but how convinced are we that they’re related?
  • Could not support the hypothesis that phenologies coincided seasonally in 1974 but not 2007 – do we think this reduces the power of their results?
  • How can we better incorporate both spatial and temporal changes due to climate change when considering mismatch, especially for mutualisms occurring on steep environmental gradients?

Particularly if you read the paper in full, I encourage you to think carefully about these potential issues with the paper. It took me six years of reading scientific papers to feel comfortable with questioning methods, statistics, results, and claims that authors make, but learning to approach science with a healthy degree of skepticism is an important part of the your development as a critical thinker and the scientific process as a whole and. Always remember that correlation does not imply causation, and that each paper is only a small part of the bigger picture that science will eventually paint on particular issues as evidence builds over time. Certain parts will inevitably be wrong, and it’s our job as researchers to figure out which bits are wrong, and which bits are right, so we can eventually discover the mechanisms and processes that drive natural phenomena over the long term.

Regardless of the potential problems with the paper, I found it most interesting because while phenological mismatch is a hot topic to study in ecology right now, this is the first paper I’ve seen that studies several bumble bee species and several plant species at once, or the whole plant-pollinator community occurring in the study area. It is also the only paper I’ve come across that considers both the temporal and spatial components of mismatch: it’s easier to quantify phenology in ecological studies, so the temporal component has been overstudied in comparison to the spatial component, but both are equally important – think about it, if the bees and plants aren’t in the same physical location, it will be harder for pollination to proceed normally.

Why should you care?

Mismatch between all different kinds of interacting species around the world can have potentially devastating consequences for ecosystem functioning, and in turn, the provisioning of ecosystem services to humans. Nature consists of highly interconnected systems where organisms interact both directly and indirectly with each other in tandem: take the famous example of wolves having cascading effects on the ecosystems in Yellowstone National Park. When wolves were reintroduced, deer began avoiding parts of the park, which allowed plants to grow back. Willow and aspen trees sprang up, and with them came more berries and insects, attracting more bird species to the park. Beavers came back and created dams with wood from the trees, and the dams attracted otters, muskrats, and other reptiles. Wolves also killed coyotes, so the mice and rabbit population grew, attracting weasels, red foxes, badgers, and hawks. The wolves even indirectly changed the rivers — with increased plant growth now that deer were not munching everything in sight, the vegetation decreased erosion and stabilized river banks. Channels narrowed, more pools formed, and the rivers stayed more fixed in their course. In this ecosystem, everything is functioning properly and imposing balance on its individual components. But what if the timing was all off? What if the deer were giving birth before the plants started to grow in the spring time and the babies didn’t have enough food to survive? What if the berries didn’t grow on the trees at the same time as birds were foraging for food? What if the deer moved to a climate they preferred because the climate was warming, but the wolves didn’t move as quickly and the deer started to eat all the vegetation again because their populations weren’t being held in check by predators? These are the kinds of questions ecologists are asking about phenological mismatch due to climate change, and finding the answers will be of paramount importance in understanding how to conserve nature and protect intact ecosystems.


Pyke, G.H. et al. 2016.  “Effects of climate change on phenologies and distributions of bumble bees and the plants they visit.” Ecosphere 7(3). (Published under a Creative Commons license).




One thought on “Paper of the day: Pyke et al. 2016, “Effects of climate change on phenologies and distributions of bumble bees and the plants they visit”

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s