I just read an interesting new paper about monarch migration habitat, and the challenges with defining it. I thought since this was fresh in my mind, I would tell you a little about this paper today, plus some other work I've done in this area, and even sprinkle in an anecdote. All of these interesting bits should hopefully lend themselves to an interesting blog entry for anyone interested in monarch conservation (see what I did there?)!
The paper has just been published in an odd journal, or at least, not one where you'd expect a monarch study to be published. The journal, which is fully online, is simply called "Land", and it appears to publish research on large-scale, landscape-level analyses. This is where scientists look at changes and variation across large scales (like continental scales), using aerial photographs, or satellite images. The new monarch paper definitely fits this theme, and it is titled, "Challenges for Monitoring the Extent and Land Use/Cover Changes in Monarch Butterflies’ Migratory Habitat across the United States and Mexico". It was authored by a group led by Rafael Moreno-Sanchez, from the University of Colorado. Here is a link.
I'll warn you, this is a thick, meaty paper filled with jargon and text that obviously is written for people who do spatial analyses all day. I only skimmed the bits where they were specifically discussing the monarchs. As the title suggests, it was essentially an in-depth discussion of the challenges that scientists have in studying and defining the habitat that monarchs use during their fall migrations. I can certainly attest to this challenge - I had spearheaded a similar study back in 2012, which was heavily cited in this new one. I'll get to mine later.
The authors appeared to be well-aware that there is a great need right now to figure out how best to help the eastern monarch population. They indicated that we know that the fall migration is a key stage of the life cycle, so it is important that we ensure that monarchs have what they need during this critical phase, and ensure that they are not losing it. But, here's the problem - what do they need, anyway? I know, I know, they need nectar, yes, but exactly where? And what type of nectar? Where is the best nectar? And what about roosts? We know they don't migrate at night so they need places to roost. But where are the roosts? Are some roost sites more important than others? If they can't find a good roost site in an area, what happens? You can see how many questions we have on this topic, and no one really has the answers to them yet. This is mostly because there isn't much research on this topic - it's an inherently hard thing to study, because of the huge spatial scale (eastern monarchs migrate across three countries!).
I'll pass on a story here that highlights just how important the answers to these questions are, and the difficulties that arise when you don't know these answers. Last year I was contacted by the Ontario government to help settle a legal dispute up there. Apparently, in a local town there was a property that had some trees that monarchs sometimes used during their migration for roost sites, and the property owner was going to cut them down to make way for a development. It was more complicated than that but I'm glossing over some bits here for brevity. Anyway, they wanted to know from me what this would do to the monarch migration. This is a great question! I offered my opinion, which was that as long as they replaced the trees with something similar, then the monarchs should be ok. If the trees were completely gone, then this could be more of an issue. But in reality, this is just an educated guess. We really don't have any data on whether the trees monarchs use are critical for migration success or not. This is the reason why we need much more research on this topic.
OK back to the paper. The authors describe how hard it is to define the actual habitat that monarchs are using during the migration when using large-scale landscape data (satellite imagery) across the flyway. This is the kind of data these folks work with. These data can usually be downloaded from government sources, so that a scientist can place a point on a map (like a monarch sighting), and then you can figure out what landscapes are surrounding that point, like grasslands, or forests, or urban, etc. In doing so, you could then figure out if the critter in question is preferring any one habitat type. The authors point out that we need to have similar spatial data between the U.S. and Mexico, and that's not easy. There is also a lack of biological information about the monarchs at this stage, as I mentioned earlier. I can count on one hand the number of scientific studies where researchers tried to assess migration habitat. There were more issues raised in this paper, but you get the point. The authors indicate that because of all of these difficulties, it is challenging to define migration habitat, and to determine if it is changing over time.
There was something that occurred to me while reading all of this. The authors may not even have been aware of this too. Not only is it hard to pinpoint the critical migration habitat from a logistical standpoint, but an even bigger issue is that the monarchs seem to use everything during their migration! Consider this - we've all seen migrating monarchs coming to our gardens in the fall. Which plants do they use? It's not just one, right? How about at a landscape-level now - which backyard gardens do they prefer? Which neighborhoods? Is there any "preference" at all? And this issue goes beyond flowers and gardens. Monarchs also use trees during the fall. Here is where I could tell you about the project I spearheaded a while ago.
That paper was published in the International Journal of Zoology, which is online (link here), and it was an attempt to define the roosting habitat of migrating monarchs. My colleagues and I figured that one way to define "migration habitat" was to look at places where we know migrant monarchs settle each night - roosts, like in the photo above. So, my team and I used sightings of migratory roosts (from Journey North), along with satellite imagery data to try to figure out if monarchs were "choosing" any particular landscape types to roost in during the fall migration. In a nutshell, we plunked a whole bunch of roost sightings down on a map of the United States, and then asked the computer, what are the landscapes around those points? Is it mostly forested areas, grasslands, urban, wetlands, etc.? We also asked (statistically) if the habitats around these "monarch-chosen" "sites differed from some randomly-chosen sites that the computer selected. We did this for different parts of the fall flyway, beginning in Minnesota and ending in southern Texas (we know the migration continues past Texas). We also tried to figure out if there were any trees that were preferred for roosts, like if one tree species was used most often. Below is a map from the paper that shows the roost sightings (black circles) plus the randomly-chosen sightings (triangles). Note that at the time we only had a few years of roost sightings to work with, or about 300 sightings. Still, it was enough to work with.
In the map above, we arbitrarily defined 5 different flyway regions so that we could compare roost habitat in each "stage" of the migration, and to see if it changes along the way.
I'll skip over some of the analyses here and just give you the end results. In the end, we found little evidence that the monarchs were choosing any particular tree type to roost in - they used every tree you can think of! Below is a table showing the ten most common tree types that were reported for monarch roosts. You'll see from this list that there were a couple tree types that were used more often, but, those trees may also just be more numerous on the landscape. And keep in mind that this was just the top ten tree types. There were many more listed as being used by monarchs for roosting. This fact alone should say something about the monarch "preference" for specific roost trees - none!
And similar to the tree thing, they also did not really choose any particular habitats (landscapes) to roost in for the majority of the flyway, except for when they get to Texas. At that point, they did appear to be choosing a specific landscape - they were much more likely to roost in trees in either agricultural areas or grassland areas in Texas. We weren't sure why. It's possible that they are being more selective when they get to Texas, because there are more actively looking for landscapes with nectar, which these landscapes have. This makes sense because we know from Lincoln's early work that they really, really get fat when they are in Texas (I blogged about that paper). Or it may have to do with the location of their flyway through Texas, which happens to fly over some open areas. Notice from the map above that our randomly-chosen sites were dispersed evenly throughout Texas, while the roost sites were not.
We concluded in that study that monarchs are extremely adaptable during the migration, and are capable of roosting in pretty much any tree, and in a wide variety of landscapes. Thus, they appear to be habitat generalists. If you think about it, this almost seems like a necessity for a creature that needs to travel across a continent and traverse a wide variety of landscapes along the way. In other words, they need to be able to use lots of different habitats and landscapes. If this species was only capable of using one type of landscape to survive, then this migration probably wouldn't even have evolved.
If you think about it, this is pretty much the worst result possible if we want to figure out what their migration habitat is! If they can use all landscapes and all trees, then which ones are the most important for conservation?
So let's recap - we know monarchs use nectar from a lot of different flowers and different nectar sources during the fall migration. We know they use a lot of different trees during the migration. There is also little evidence their roosts are located in one particular landscape over another, except for a mild preference for open landscapes in Texas. So where does this leave us? What is migration habitat? Everything?
I'll leave you with this final thought - in the coming year or so this issue is going to become really complicated, along the lines of my interactions with the Ontario people. I'm referring to the upcoming USFWS decision to list, or not list, the monarch on the endangered species list! Think about this, if monarchs become listed, that will mean the "habitats" they use will then become federally protected. This should include habitats at all stages of their lives, including their migration. See where this is weird? What habitats? Anywhere there's a flower? A tree? What do we protect?