165 - Zach Portman - When pan traps might not pan out (for bee surveys). (in English)

Andony Melathopoulos: [00:00:00] Just before Christmas, there was a great article in the New York Times announcing the start of a US National Native Bee Monitoring Network, which you can learn more about at usnativebees.com. And the initiation of the network raises questions about how best to monitor bees on the scale of something like the United States.

To help us understand some of the issues surrounding that endeavor, I invited on the episode, Dr. Zach Portman . Dr. Portman works in the Cariveau lab at the University of Minnesota, is the native bee taxonomist there, and recently he led a paper called the "State of Bee Monitoring in the United States: A Call to Refocus Away from Bowl Traps and Towards More Efficient Methods."

In today's episode, we're going to have Dr. Portman explain some of the key themes in the paper , but also expand a little bit on his thoughts on what we really require when it comes to a survey method moving forward to do native bee survey efficiently, and to really get the most scientific value out for a unit of survey that we do.

So without further ado, Dr. Zach Portman this week on PolliNation.

Okay. Well, welcome Zach. I'm really glad to have you on the show.

Zac Portman: [00:01:24] Thank you. It's good to be here.

Andony Melathopoulos: [00:01:26] Now, to begin with... we're going to talk about these things, these passive traps and pan traps being one of them. Can you describe what a pan trap is, and how people who are doing bee surveys typically use them.

Zac Portman: [00:01:41] Okay. So yeah, for people who aren't familiar pan traps are a tool that people use to collect bees. And what they are essentially, is just colored bowls filled with water, soapy water. And so it's not completely understood how they work, but the thing is that bees fly to these bowls, they think that they're flowers and they try and land in them and they hit the soapy water.

There's no surface tension and they go in and they drown. And people put these out usually in sets of three. So you have different colors, a white, blue, and yellow. And you usually put out 10 sets of three for 30 total for about a day. People leave them out there , the bees fly in, and then they collect them at the end.

Andony Melathopoulos: [00:02:30] Okay. So I guess when you come back after a day or so, you're going to have all these little bees in there and then you can go forward and identify them. And I guess you and your coauthors came up with a paper recently sort of kind of assessing the use of bowl traps. And one of the things I thought was remarkable, and you have a nice graph in there, is that there's been an explosion in the number of studies using these traps to survey for bees.

What makes this survey tool such an attractive method for researchers? Why have so many people cottoned onto it recently?

Zac Portman: [00:03:06] There are a lot of reasons. And also just as a note, before I go forward, these traps had various names; so there are pan traps, bowl traps, bee bowls, muricky traps. And I'll probably be using those names interchangeably because I can never remember which one.

Andony Melathopoulos: [00:03:23] Well, the other thing to mention is that there are other forms of  passive traps.

I know vein traps are another . There's a variety of them, but these have been really prominent, these painted bowls.

Zac Portman: [00:03:37] I think the reason that people really like them is because they are easy, and simple, and they're not very time consuming at least initially. So, you know, you go out, you put them out in a field, you put water in them, and then you just walk away.

And so when you compare that to other methods, like going out with an insect net and catching bees that way, or going out and doing transect surveys where you go and you visually identify bees. You know, those are really labor intensive; there's definitely a learning curve with that.

And different people, especially netting, could potentially have different biases where they target specific bees or they notice specific bees. And so people really like pan traps because they're seen as unbiased; no matter who sets them out, they're always going to kind of do the same thing.

Andony Melathopoulos: [00:04:32] Oh, I see what you're saying. So if somebody went out with a net, for example, you may have somebody who's just got super agile and really knows how to use a net. And they come up with... they survey a certain amount of bees, but you have somebody who's really starting out, and it, those bees are in the landscape, but you've really underrepresented, or you can get different kind of reads on them if the people don't  have the same skill level. Whereas if you put these pan traps out, doesn't matter who puts them out, the rate is sort of determined by placement and all sorts of other things, but not the person putting them out.

Okay.

Zac Portman: [00:05:04] Yeah. And that seems to be what most people site. But yeah, I think it's mostly just easy and convenient; you can go out and lay out pan traps and catch a lot of bees. And so if you're an ecologist and you're really just trying to get as many data points as you can, as quickly as you can, that's a really attractive option.

Andony Melathopoulos: [00:05:25] Okay. So I think if anybody's worked with pan traps and knows the literature a little bit, that these traps are biased towards collecting certain bee taxa . What did you learn when sort of reviewing the literature about the biases of these pan traps and maybe why do these biases matter in being able to detect changes in rare bee species?

Zac Portman: [00:05:50] Yeah, that's, that's a really good question. And it's actually like, there's kind of the simple answer and the complicated answer. And so there've been a lot of studies kind of comparing different methodologies and seeing what bees they capture. And so through those studies, we've seen that, you know, bowl traps collect tons of sweat bees in the family halictidae. Essentially, no matter where they are, they always catch a lot of sweat bees. And then there's certain bee groups like the genus colletes, for example, which very rarely  ends up in bowl traps. So you know that they're out there if you net collect, but then you look at your bowl trap catch and you don't find any.

And then there's other groups where, you catch very few of them with netting, but you see them in huge numbers in bowl traps. And so, there is definitely this kind of taxonomic bias in terms of what bees end up in bowl traps. And one thing that's really worth noting is that there are a lot of studies comparing different methods.

So you say, bowl traps are biased compared to netting, which are biased compared to vein traps. But, what really, I think is important, that's really under-appreciated is that we don't know how they're biased in terms of the underlying bee fauna. Like we don't know which method is the most correct in terms of reflecting what bees are actually there.

So we know bowl traps, for example, catch tons of halictids. But is that because halictids are super attracted to them or because there are just tons of halictids in the environment to begin with? And we really, we really don't have a good sense of that.

Andony Melathopoulos: [00:07:40] I remember also one of the comments in the paper is that many bees in the halictidae are generalists, and that if you were losing habitat, they may not be the most reliable, potentially reliable species to sort of detect the effects of habitat degradation.

Zac Portman: [00:08:01] Yeah, definitely. So there's, there's a couple reasons why you wouldn't want to monitor halictids. And yeah, so the first one is that they are generalists; they are tiny; they are social. You often find them in heavily disturbed habitats, like parking lots and field margins. And they're kind of just ubiquitous in the environment.

So if I had to predict bees that would tend to do well under human disturbance , those little lasioglossum, dialictus would be probably one of my top choices. So if your interest is in, you know, finding rare bees and detecting declines , they would not be my first choice.

Andony Melathopoulos: [00:08:48] Well, keep going Zach.

Zac Portman: [00:08:51] Yeah. So another reason I wouldn't necessarily want to monitor halictids is because so many of them are so incredibly difficult to identify, especially the dialictus, where there's only a handful of people in the United States who can reliably identify them. And a lot of that is just because there isn't a taxonomic foundation.

A lot of these species have never been taxonomically revised. So you end up with like, species especially from the Western US, where they were described maybe in like the 1930s and nobody's treated them since. And so there's just a lot of gaps in that knowledge, and they're probably the most misidentified bee group that I encounter.

Andony Melathopoulos: [00:09:39] You know, that's where I was going with this, is that realizing that bees like dialictus are going to be accumulating, and just not being able to figure them out, leaves you with a very fuzzy picture, I suppose, in terms of it's not a really well resolved group.

And so you end up not being able to really get a diversity estimate, because you stop. So I guess that creates a bottleneck. So, someone like you, a taxonomist, is given a bunch of bees that have been pinned up out of a pan trap, and you're suddenly confronted with bees that will take you a very long time. And that's for a taxonomist, but I imagine what the misidentification is that somebody doesn't have the taxonomic skills, has just done bee taxonomy for a few years. They're going to also kind of grind to a halt as soon as they hit this group.

Zac Portman: [00:10:37] Yeah, absolutely. And that's something you see kind of over and over again in a lot of these pan traps studies, is, you know, it's really easy to collect a lot of bees and people love that, but then you get into the processing, and labeling, and identifying stage, and everything just grinds to a halt.

So they seem like a really, easy, simple method initially, and they suck you in and then you hit the hidden barriers further down the line. And especially with dialictus, there's many people who appreciate how difficult to identify they are. And then there's other people who maybe don't have that knowledge and kind of dive right in. And, there's keys on discoverlife.com where, they claim that you can identify dialictus from them.

But if you run a specimen through one of those keys, you're going to end up with the wrong bee probably 50% of the time. You're just going to misidentify that species. And that's something that's well known among kind of people who are more experienced taxonomists, but for someone who's kind of just jumping in and doesn't have that knowledge, things like to discoverlife keys are just a trap waiting to happen.

Andony Melathopoulos: [00:12:04] Well, I guess the other thing is that pan traps are not something that researchers in the 1930s would have used; it's a very relatively recent development. And I guess that the attraction is, at some level, is that with pan traps you can get in, you can get some kind of measurement of abundance and numbers, but  that these may be very hard to relate to older studies.

Can you talk a little bit about that limitation?

Zac Portman: [00:12:31] Yeah. So there's a couple of things there. With older collectors, generally, professional entomologists or researchers that would go out with a net and collect stuff, maybe they might have a malaise trap where they're catching bees as bycatch.

But in general it was, I think, much more targeted collection. And a lot of those collectors would really seek out rare and interesting species. Whereas now there's a lot more passive traps where, the goal is to sample as many things as possible and so you end up with a lot of the common abundant species.

You get a lot of those. Whereas, an expert who is net catching in the past would probably avoid a lot of those. And then, on top of that, we already know that there are different biases between netting and bowls. So if you take an old data set that was net caught, and then you try and compare it to something that was caught in bowl traps, you're going to be comparing the biases of those two methods and not necessarily any changes in the bee community.

Andony Melathopoulos: [00:13:49] Okay. Well that seems like a very significant limitation overall. So, we've got these various problems... so the pan traps may be biased towards generalist bees that may be super abundant in disturbed areas, but also some of them just are not easily resolvable into species. You've got this problem of just a lot of specimens coming in and overwhelming the collection. I imagine you've got to put them somewhere that you also, someone like yourself, a professional taxonomist, needs to move through in an efficient way.

And you're sort of confronted with these bees that may not give you a real good idea of what's going on with overall biodiversity, and then this problem of relating back to historic studies being kind of hard to do. Given these limitations, what do you propose, and your coauthors propose, as a way forward to survey wild bee populations?

Zac Portman: [00:14:46] Well, before we move on to that question, I think there's one final, huge drawback that really needs to be emphasized. And that's that you don't get information, or good information, on the true underlying abundance. So for example, if you place out a bowl trap and you catch 50 specimens of agapostemon virescens, are you catching 1% of the total that are out there or 10% or 50% or a hundred percent? You just, you don't know what proportion of the bees are being caught. And then there's the added complication of you don't know how that's affected by the environment. So if there's a lot of flowers blooming, potentially those bees are visiting the flowers, and ignoring the bowl traps. Or, they're being pulled in to flowers and then seeing these bowl traps and flying to those. And anecdotally, I've talked to a lot of people where in drought years, they tend to have much higher catch rates in bowl traps, because there are fewer flowers out, but there doesn't seem to be necessarily a straightforward relationship.

Andony Melathopoulos: [00:15:58] And I guess with that point, it's not like you pick up a paper and there's like a count of how many flowers there were, so you don't have any idea when you look at this dataset. What the conditions where you have to sort of... you're stuck.

Zac Portman: [00:16:13] Yeah. And so, in terms of do bowl trap catch rates reflect the actual underlying abundance, there's really not much good evidence one way or the other. The evidence that is out there generally says, no; it doesn't seem to be a good predictor of the underlying abundance.

There's one paper out there by Thomas Wood, where they were looking at in England, they had bowl traps and they also did microsatellite analysis on one or two bumblebee species. So they were able to calculate the number of colonies or estimate the number of bumblebee colonies out there. And when they compared that estimate to the bowl trap catches, they just weren't correlated at all.

Andony Melathopoulos: [00:16:58] So let me get this straight. In that study, they would take the bumblebees and by looking at who's related to who, you could say, "Oh, there was only two bumblebee colonies in this landscape that we're catching, but the bowl traps may have high or low numbers. They weren't really related to the number of nests in that area." So that really does kind of suggest that this is maybe not reflecting underlying reality, giving a distorted picture of it.

Zac Portman: [00:17:27] Yeah. So, especially in the context of are populations going up or down? Bowl traps don't really tell you that. Maybe they could detect really big swings, I think.

But you know, if you're trying to find like, was there a 10% decline or a 20% decline in this species? I don't think bowl traps can tell you that.

Andony Melathopoulos: [00:17:51] Okay, well, let's take a quick break. I wanna just take a moment for our listeners to kind of pause and think about that, and then let's come back to discuss which way forward; how to get around some of these limitations, what makes it a more sensible approach to surveying bees?

And so given these limitations of these passive traps, what do you propose is a way forward for surveying wild bee populations? What would be a sensible way to advance?

Zac Portman: [00:18:18] Yeah. That's kind of the million dollar question that I think a lot of people that are thinking about and struggling with because monitoring things is really, really hard.

And even in other animals where they're big and you can identify them easily, there's still so much uncertainty. And so bees are really hard because they are so diverse, and there's so much we don't know about them, and there's relatively few species that you can identify without killing.

And so, there really aren't a lot of good answers and there aren't a lot of studies that have really done monitoring well.

So that being said, I think there are a couple areas that I am excited about. And I think personally, I think, community science or citizen science is really exciting because I think a lot of people are getting into bees and collecting data on bees and interacting with things like iNaturalist and Bumblebee Watch, and I think that a lot of high quality data is being collected. I think it's just a matter of figuring out how can we use this data to inform monitoring?

Andony Melathopoulos: [00:19:47] Okay. So when you're going through a lot of those photo vouchering approaches, people really don't kind of get stuck in a group of bees where ... it's a much more selective kind of pulling in of specimens, and also it doesn't have some of the limitations of having to sort of sort through all this material. Although I suppose somebody has got to sort through all those images and sort of attempt to put some kind of identification, but at least it gives us a good geographic spread and I'm always impressed with how much material is coming in through those. It's a big pipeline of observations.

Zac Portman: [00:20:30] Yeah. And I think that as more people get into it and as more people gain expertise, and better resources are made, I hope that that will help move that along.

Andony Melathopoulos: [00:20:43] I guess you also mentioned the study with the micro satellites as a way to get a real measure of abundance. That seemed like a really great approach to, how many bumblebee callings are in this landscape. Then some of these other techniques can get at those questions a little bit more precisely.

Zac Portman: [00:21:01] Yeah. And that's something else that we recommended, and that's, I feel like genetic population estimates... they can do that for a lot of other organisms, but it hasn't really been applied to bees  in terms of monitoring them. But I think that technology is really advancing on that front, really quickly, and could hopefully open up a lot of avenues that would allow us to actually look at population abundance in kind of a meaningful way.

So that's the part that I'm optimistic about, but I'm also... that's probably my weakest area of knowledge. So I think there are other people who are looking into that and I'm hopeful that they'll come up with ways to actually use that for monitoring.

Andony Melathopoulos: [00:21:50] You know, and as opposed to abundance, the other area that you raise as kind of a target is range ... getting better ideas of range .  Going for specific bees of interest and delineating getting more observations of their range rather than, how many are there in the landscape?

You talk a little bit about that as an approach?

Zac Portman: [00:22:16] Yes. That one's also tough. I mean, to kind of step back a little bit, I think one of the pitfalls of current monitoring efforts is that they use these passive traps and are really trying to look at everything at once. And when you have, you know, 4,000 species of bee in the United States and usually like four or 500 species in any given state, at least, that's just Eastern States.

You know, if you try and monitor every single one of those species, you're going to get quickly overwhelmed. And so, my hope would be that you can select some species that you can really focus on, go and collect all the data you can on it, and be able to tell something. And there are, there are studies that have done that.

So, there's a big study from 2011 by Cameron et al; that was the big paper showing the widespread continental scale decline of affinis. But for that, they looked at eight species of bumblebee and looked at, I think over 70,000 records of museum specimens, and also going out and collecting specimens, and so in order to really get at these robust changes in declines, it required huge amounts of effort and really drilling down on just a handful of species in order to do it well.

Andony Melathopoulos: [00:23:58] Okay. That that's great, sort of prioritizing. And I guess this is the comment maybe the deception you talked about earlier... it's easy to get lots of data, but then you pay for it later in some ways.

And if you had a more focused question of like here's a group of bees that are related, the bumblebees for example, and kind of drilled in on that question you could actually get at conservation questions much more efficiently for per dollar, or per labor, than just flooding your research with jars of alcohol.

Zac Portman: [00:24:33] I mean, no matter what you do, it's going to be expensive and labor intensive, and for that bumblebee paper, there was a large team of people working for years to get that together.

Andony Melathopoulos: [00:24:53] Okay. But even given that, with a focused question, even though it's going to take a lot of resources, it's going to be resources that result in an answer as opposed to... you collect a whole lot of bees and then it's just that there's no focus and you end up spending a lot of resources, but you actually don't get to any clear answer at the end, I suppose.

Zac Portman: [00:25:17] Yeah, exactly. And I think that's the state that we're in right now, where there's tons of people who are getting into bee monitoring and they're doing a ton of collecting. And then we're ending up with a lot of specimens sitting neglected in drawers, and it's not really being connected or brought together in a meaningful way that can really bring us good conclusions.

Andony Melathopoulos: [00:25:45] Okay. Well, thanks for those suggestions. I think that gives us all a lot to think about. Let's take a quick break. We're going to come back. We ask our guests for book recommendations and tool recommendations. I'm really, really curious what your answers are going to be. We'll be back in just a second.

Okay, we're back. So I'm really curious.  Got these three questions we ask all our guests. Do you have a book recommendation?

Zac Portman: [00:26:11] Yeah.  I have a couple of book recommendations and I think that they kind of tie into how I approach monitoring myself. And when I think of bee monitoring, I think of individual species. How is this species doing?

Like, how is Bombus affinis doing? How is Bombus tericcola doing? You know, how is perdita meconis is doing? I don't think in terms of these big species diversity metrics. And I think that one of the most important things to me is getting a proper and thorough understanding of the natural history and their biology and you know, what host plants they need, and where they nest, cause for so many bees that is a total black box. We know that they exist or they occur in certain places, but in terms of what they need in order to succeed and live their best lives, we don't know.

And so I have a couple of books that I really like that have helped me approach and understand that. The first one is one I really love. And so they're both kind of scientific. I really loved the Bumblebees : Behaviour, Ecology, and Conservation by Dave Goulson. And it's a pretty heavy scientific work, but, you know, if you read that, you'll come out more knowledgeable about bumblebees than almost anybody. So that's probably my number one book recommendation.

Andony Melathopoulos: [00:27:49] What else do you got?

Zac Portman: [00:27:51] And then just one that I like. There's some parts where I wish that it could go into more detail, but The Solitary Bees by Danforth, Minckley, and Neff.

You know, it was a book that I enjoyed and I think will foster in anybody that really deeper appreciation of the intricacies of bees, and how they live, and what they need. And I think that there's so much we don't know about bees. And I would love if more people could go out there and really like be naturalists and help us understand what bees need, and where they nest, and what flowers that use.

Cause that's a huge information gap, even for our common species, like bombus affinis for example. There's only a handful of literature records of where they nest, so it's declining, or it has declined and many people are now scrambling to figure out what does this bee need and how can we help it?

And it used to be one of our most common, abundant bumblebee species, and we know relatively little about it.

Andony Melathopoulos: [00:29:06] That's a great point. There's such an emphasis on getting an identification on a bee, but there's a lot of great discoveries that can be made by amateurs in terms of just learning some basic biology on a bee, watching it, and asking some questions about its life.

Zac Portman: [00:29:23] Yeah, absolutely. Yeah. Even something as simple as, you know, finding a bumblebee nest and taking pictures of the bees going in and out. So, you know where they like to nest; that's really important and there's not a lot of data out there on it.

Andony Melathopoulos: [00:29:37] All right. The next question I have for you is, do you have a go-to tool if you were on a desert island?  What I know they're not going to be a pan trap, but what would be your go-to tool for the kind of work that you do?

Zac Portman: [00:29:53] For me, I would have to say my camera absolutely.

Andony Melathopoulos: [00:30:00] Tell us a little bit about your camera. What's is it, a regular camera, or have you spent a lot of time thinking about ...

Zac Portman: [00:30:05] I have a Sony DSLR camera which I like, cause you can take really nice high quality pictures.

And then also for me, I'll often take a lot of videos. And a big fascination for me is bees on flowers and what they're doing on flowers and the different ways that they gather pollen , which is pretty under-studied and under-appreciated all the different strategies that bees use.

Andony Melathopoulos: [00:30:33] You know, I haven't even thought about that problem before.  I always think, "Oh, bee goes into the flower, gets the pollen." What, what are some strategies that that bees are using that are kind of unique? What would listen or watch for?

Zac Portman: [00:30:46] Yeah. So there was a few kind of the classic two behaviors that a lot of people are looking at these days for bumblebees. So the first one is buzzing where they vibrate their wing muscles to buzz the flower and expelled the pollen that way. And bumblebees and tons of other bees do it. And it's not really well known exactly how pervasive that behavior is. You know, some bees do it, some don't, it's not really clear why.

So that one's really interesting. And then the other main way that bees gather pollen is they just use their forelegs to scrape it up. And usually bite with their mandibles to help loosen the pollen when they do that. But then, in a lot of bees, you get some other really cool behaviors.

So there's a lot of bees that have modified hairs on their faces, and they'll rub their heads against the pollen or to pick it up that way. And then there's other bees... if you've ever watched osmia on various asters, what they'll do is they will rapidly move their abdomen up and down and collect pollen directly that way.

Andony Melathopoulos: [00:32:00] Oh, right. Okay. I, you know that the two things I've always noticed is for example, bumblebees on things that aren't poricidal anthers like roses, they'll still buzz it. And I was like, "huh, honeybees can get the pollen off of those things, but it must speed the whole process up."

Zac Portman: [00:32:18] Yeah. And there's a lot of research being done on that by a lot of different lab groups and yeah, they're still trying to figure out why they do it. There's definitely a learning component involved. And it also has to do with the amount of pollen that's available. I think if there's less pollen available, the more likely they are to buzz it. It's traditionally been thought that buzzing was done on poricidal flowers, but yeah, the reality is that bees will buzz all kinds of different things.

Andony Melathopoulos: [00:32:51] Okay. Great. So that's great. The camera, and I love the idea of developing a search image to try and get something specific captured. These natural history notes with your camera so that somebody can go back and ask the question like, "Oh, how does it collect pollen? And what is it? Is it even collecting pollen? Was it even on that flower collecting pollen?" I think those are, those are great. Okay. Well, we've come to the last question, which is ... how will you answer this? It's, do you have a favorite species? As somebody who who's talked this episode about "let's not do all of them." Is there one that you'd sort of like, or one group I know you've been working on Perdita , does somebody stand out for you?

Zac Portman: [00:33:35] Yeah, and honestly, they're all my favorite. At least all the native bees. And it's hard to rank some above the other. I think one that's yeah, the genus perdita is really special to me.

I think they're some of the most fascinating bees. And one of my favorites to go and watch, and look at, and take pictures of that we have in Minnesota is perdita perpallida. It's this tiny completely yellow pale bee that's a specialist on Dalia and a couple other related plants, and they're really only found in dune systems. They're just delightful to go and watch and find.

Andony Melathopoulos: [00:34:30] You know, this might be a good segue to just a one last thing that we were talking at the break was the importance of taxonomists and specially in doing revisionary work and being able to go through and that a lot of our knowledge of bee taxonomy is old. One of the reasons why people have a difficult time working through some of these bees that are coming out of things like pan traps is because the taxonomy has just not been resolved.

Zac Portman: [00:34:59] Yeah. In the US and in really in the world, there's a really acute, what's called the taxonomic bottleneck where there's, a lot more things that need to be identified and a lot of unknown species that need to be described that just aren't being .  In the United States, a lot of the taxonomy work that forms the foundation for the research we do today, was done in the fifties and the sixties and has never been updated since.

And so, yeah, one thing I think about is like how much other research from the sixties, do you trust uncritically without it being updated? And so for a lot of these bee groups, there is these cryptic species complexes, or there's new species that only a few taxonomists know about.

And there is no one to really update that knowledge because there aren't enough taxonomists, there isn't enough funding for basic taxonomy research, and there's really no funding to create like good high quality identification resources that anybody can use. So that's something I think about a lot.

I don't know if I answered your question.

Andony Melathopoulos: [00:36:22] Yeah. That's what I wanted to hear from you on. And thanks so much for coming on the show. I really hope we can get you back in a future episode.  This was a fascinating conversation.

Zac Portman: [00:36:32] Yeah. Thank you for having me.