Between 2006 and 2015, about 25 percent fewer bee species appeared in museum and naturalists' observation records compared with the period between 1946 and 1995, according to a study published today (January 22) in One Earth. The results corroborate other recent reports that the fuzzy pollinators are in peril.
To measure insect declines, many studies sample the same location repeatedly over time. “That’s the ideal data,” says Eduardo Zattara, an adjunct researcher at the National Scientific and Technical Research Council (CONICET) in Argentina and a coauthor of the study. But because these data only exist for specific locations and bee groups, examining global bee trends requires a different approach. Zattara and his colleague Marcelo Aizen, a senior scientist at CONICET, used data from the Global Biodiversity Information Facility, a database that collates millions of records from museum specimens, private collections, and community science observations, to look for worldwide trends in bee diversity between 1900 and 2018.
“The idea was that even though those collections are made by very different people for different purposes and many different areas and many different times, perhaps if we can gather [these data] all together, we can average out the differences and [look] for some global trends. . . .We weren’t aiming for precision; we were aiming for scope,” says Zattara.
The Scientist spoke with Zattara to learn more about what these records tell us about bee diversity and how better data sharing could improve scientists’ understanding of bee declines.
The Scientist: What were the main findings from the study?
Eduardo Zattara: What we did was count how many species in total had been reported in these records each year. The idea was that if there has been no change in diversity, basically you would have about the same chance of finding any species in a given year over time. So even though people will change where they’re sampling or what they’re looking for, on average you will end up finding about the same amount of bee species. And so we tested that, and we found that no, that wasn’t the case. Actually, we were seeing fewer species, especially after the 1990s.
Maybe it’s because people are just looking less and there are fewer records. . . . So we counted also how many records we had in the database, and actually we found that because we incorporated not only collections from museums, but also records from [community] science projects, the number of records is actually increasing and it’s still going up. So it wasn’t about a problem with the number of records. . . .
When we talk about loss of species, we’re really talking about species not seen in the records. . . . So imagine that the whole world is kind of like a big sampling program. If you have something that’s rare, every now and then you’ll find it. If it becomes rarer, the chance that you will miss it in a given year is higher. And so when you do that for species that are becoming rare, rarer than before, then there’s this loss in the number of species that you’re counting. And hopefully it’s just that, and there’s not a lot of species that are actually going extinct, but there is a possibility that that’s also happening. So the rarest species might actually be becoming extinct in some places.
TS: What was your reaction to these findings?
EZ: I was surprised by the fact that the data actually showed such a strong trend. Actually, most of the work wasn’t doing the initial analysis but doing all the other stuff to figure out if it’s not some artifact from the data.
There are many different reasons that may cause this artifact. You would think that people who are collecting stuff are going to have this collector behavior where they want to go for the rare things first, right? So that would actually enrich your collection with species rather than rarefy. . . . Another reason could be that taxonomic expertise is being lost. So many more people are finding things and they don’t know what species they are, so that species level ID is lost. Many hard-to-identify species are either left unidentified, and thus filtered out from our analysis, or misidentified as a few more-common species.
When we talk about bees, we’re talking about 20,000 species of bees and not just the honey bee.—Eduardo Zattara, CONICET
If that happens all over the world systematically, that would also cause this kind of trend. So a lot of the work that came after finding this initial result was to see if we can check for those different sources of bias to see if we can find any good enough alternate explanation to the interpretation that actually bees were really in trouble. Most of the potential biases could shift a little bit the trend, but none of them could really explain [the result] as well as an interpretation that probably bees are declining worldwide.
TS: Was there anything surprising about your findings?
EZ: It was really interesting when we looked at which big regions the data were coming from. The lion’s share was coming from either the US or Europe, and there were far fewer data from South America or Asia or Australia or Africa.
The interesting thing was that when we had more data, the trends were clearer and the confidence was higher, so we had narrower confidence intervals. And that to us pointed to another thing that we think is probably the most important part of the paper because it’s the place where we can do the most things about. The idea that if we had more data we would be much better at pinpointing the trends and having more confidence in the trend. And the thing is that the data exists. It’s not like we have to rue the fact that we don’t have the data. There is a lot of data in museums that have not digitized their collections, or they have a catalog that has not been shared with these public databases. So there is a lot of potential in encouraging data holders to share those data and to have funding agencies encourage and fund projects that aim to move data from old shelving units to a digital public database.
TS: Another thing I noticed in the study was it seems like there was a different trend in terms of the number of records of museums specimens compared to human observations. Can you talk a little bit about the significance of this?
EZ: Yes, I think that it has to do with the fact that there are a lot of things that have been changing since the ’90s. A lot of countries have become aware of the value of their own diversity. Back in the days of big museum expeditions, people from a few countries would collect anything from everywhere and bring it back to their own countries and put it in museums. But at some point, many countries realized that that was their biodiversity and that they should be the ones to catalog and keep hold of it. And so they put more restrictions on what you can take out of the country. But oftentimes those countries didn’t have enough infrastructure to quickly catalog and digitize and make available what they have. And so there is an obstacle, or at least an important delay, in these data and data mobilization [and] sharing. I mean, there’s all the right reasons that are correct that each country should have the right and the decision to take care and count their own biodiversity, but the final effect is that you will see a small amount of actual specimens that get moved around.
On the other hand, observations are mostly a record that says, ‘I saw this species of bee in this place in this time,’ and there are no restrictions on that. You can go on vacation, you can go on an expedition. As long as you don’t take physical specimens, it is totally okay for you to publish that data. And to that you should add citizen science, it’s booming all over the world. It’s much easier, a lot more people have access to it, and that creates this exponential growth of records.
TS: There have been a lot of studies recently about insect declines. How does this study tie into these other reports?
EZ: Basically what we are finding is that it’s a subset of the bigger insect decline. Bees are the main group of insect pollinators, so they have this added interest in that we know they’re really important both for wild plants and for crops. We’re learning how much a lot of crop yield depends on bees and not only honey bees, but bees. Oftentimes honey bees cannot replace other species. And also bees are particularly attuned to the environment. So most wild bees are seasonal, they have these specific nesting requirements to build their nests, they come out at a certain time during spring, they need to have certain resources to raise the next generation. If there’s a mismatch between what they evolved to use and when those things are either in store or flowering—because you can have changing phenology [the timing of biological events] driven by climate change that makes the flowers that they depend on flower later or sooner than [the bees] emerge—then you don’t find them. For that reason, we think that they are specifically prone to suffer from a lot of the drivers driving insect decline in general.
I could [tell you] what I think are the main drivers of this decline, but I will just point you to the PNAS series of papers because they basically say we have three main drivers. Changes in land use—that’s a huge topic because it means basically replacing natural habitat with things aren’t natural, from urban sprawl to agriculture, especially monocultures and all of the agrochemicals that come associated with that. There is also introduction of invasive species, which oftentimes can cause impacts beyond that of land use. . . . And then finally, climate change, which causes all kinds of trouble, and it’s one of the main reasons why we are having these shifts in phenology that create this mismatch between bees and the resources they need.
TS: How would you like the results of this study to be used?
EZ: One thing I think is important is encouraging data sharing and data mobility. So hopefully people will realize that dataset that’s very rich and is only on their hard drives might be useful to have it shared along with these other data. Also, this will help people justify to their funding agencies about digitizing their collections and sharing them. So I hope that that’s one impact this paper will have.
The other impact I hope it makes is to put bees on the radar of the public and the decision-makers. . . . When we talk about bees, we’re talking about 20,000 species of bees and not just the honey bee, which is something that’s actually been a major point because a lot of this ‘saving the bee’ is becoming co-opted by the bee industry. Somebody said on Twitter that it’s like being worried about bird diversity and then trying to do something about that by funding chicken farms.
We think that [bees] are specifically prone to suffer from a lot of the drivers driving insect decline in general.—Eduardo Zattara, CONICET
I think it’s important that people should learn a bit more about wild bees. They’re really nice, they’re cute, they’re important. Once they find there’s more than just honey bees and there’s all this diversity, they will start taking care of them. . . . Actions can go anywhere from thinking better about what do you plant in your garden, thinking whether you really want to have that cleanly manicured lawn or you just leave it grow a bit more and leave it to flower.
On the other hand, it’s more about the big things driving declines of bees and pretty much everything else, so it’s what politicians are saying about what they plan to do about global warming and climate change and the economic policies, because those are the big drivers and even though no one can on their own change that, they can do a little something by who they’re voting for.
TS: So do you have a favorite bee species?
EZ: Oh, yes, I do. We have the giant Patagonian bumble bee [Bombus dahlbomii], which is endangered. It used to be very common in Patagonia until the ’80s, ’90s when two species of European bumble bees were introduced in Chile and they became invasive, and the populations [of B. dahlbomii] have plummeted. We think that’s because [the invasive species] brought new diseases that local bumble bees had no defenses for. That’s our focal species these days. The queens are [up to 40 mm] big. You wouldn’t believe that these things can actually fly, but they do.
E.E. Zattara, M.A. Aizen, “Worldwide occurrence records suggest a global decline in bee species richness,” One Earth, doi:10.1016/j.oneear.2020.12.005, 2021.
Editor’s note: The interview was edited for brevity.