• Andy Davis

Monarchs have a growing parasite problem, and it's not from natural causes

Hello everyone,

This might be the most sobering blog post I've ever written about monarchs, and so I hope you pay close attention. I also hope that after reading this, you share it with every single person you know who cares a whit about monarchs, because this information should be shouted from the rooftops. Today I'm going to talk about a brand new study that is a game-changer for the world of monarch conservation. And, once I explain this paper to you, I guarantee it will change your entire view of the role we humans are playing in saving the monarchs. Think this is overstating things? Not in the slightest. Keep reading...

Before getting too far into the weeds, let me get us all on the same page about the parasite I'm talking about. For many monarch enthusiasts, this should be review (if it isn't, then shame on you). Ahem...Monarch butterflies, Danaus plexippus, along with their relatives, like the queen (Danaus gilippus) and other Danaids, are sometimes infected with a naturally-occurring protozoan (meaning single-celled) parasite called Ophryocystis elektroscirrha, or simply OE for short. This parasite has been found in all monarch populations around the world, at varying levels, and, it has been around for probably as long as the monarch has - for thousands of years, probably. We think it co-evolved with the monarchs and their relatives, because the parasite appears to be perfectly adapted to living on these butterflies. We don't think it lives anywhere else, just on these butterflies. For the longest time, we thought that this parasite was ONLY found in Danaid butterflies, but this actually turned out not to be true just recently (see a prior blog).

The OE parasite forms microscopic, football shaped, spores that hang out on the outer surface of the monarch abdomens, sandwiched between the butterfly scales (see the picture below). When the adults land on, or fly close to, a milkweed plant, such as when a female lays eggs, these spores fall off onto the plant (think of it like glitter). When monarch eggs hatch, and the caterpillars start eating the plant, they also ingest these spores. Once inside the caterpillar, the spores "hatch" and become the reproductive stage of the parasite. They reproduce (a lot!) within the caterpillar and in the pupal stage, and then when the butterfly emerges from the chrysalis, it has millions of spores on its abdomen, ready to begin the cycle again. And I'm not kidding when I say millions - we've counted.

The other thing that people should know about this parasite, is that it sometimes causes the butterfly to be deformed when it emerges, but then it sometimes does not. In fact, in a lot of cases, the infected butterfly looks completely normal to the naked eye. This is why the only way to truly know if a butterfly is infected with OE, is to take a sample of their abdomen scales using scotch tape (which is harmless), and then look at the tape under a microscope. Anyone who tells you there is any other way is lying or uninformed.

OK, so now let's get to this new study. Full disclosure - the study I'm referring to is one that I am a coauthor on, so I had a front row seat for this whole effort, which actually started many years ago. The other authors are very easily the leading experts in the world on this very parasite. They were my wife, Sonia Altizer, a world expert on OE, Dr. Jaap de Roode, a professor from Emory University, who trained in Sonia's lab as a postdoc 17 years ago, and is now too a world expert on OE, and finally, Dr. Ania Majewska, who was a grad student in Sonia's lab, and is now a postdoc in Jaap's lab! So like I said, this esteemed group is easily the combined world experts on the OE parasite. Ania should get a lot of credit for this paper, as she was the lead author, and did a lot of the data-crunching.

The new paper has been peer reviewed and accepted, and the abstract at least is now publicly-available online in a very prestigious scientific journal - the Journal of Animal Ecology. Here is a link to the accepted paper, though I'm not sure that the full pdf paper is downloadable for free. However, without breaking any laws, I should be able to legally tell you that if you wish to get a copy of the actual paper, it is as simple as emailing one of the authors and asking for it! But if you don't do that, not to worry, I think that I'll be able to convey the main points and takeaways from the paper here.

The idea for this paper actually was hatched many years ago. I can recall us talking about doing something like this in various meetings and conferences, and we actually had started it a couple of times over the years. We had always had the same very basic question though - since the OE parasite has probably been with monarchs for a long time, what is the long-term prevalence of the parasite in the North American population? And we had other questions too - Does the rate of infection differ between the west and east? And, are the long-term trends between east and west the same? We already knew that the various populations of monarchs around the world tend to differ in their infection prevalence, because of some earlier work by Sonia and Lincoln Brower - they had shown that it is very tied to the length of the migration of a population. Monarch populations that have a long-distance migration tend to have low infection levels, while those that have shorter, or no migrations, tend to have lots of parasites in the population. This is because the parasite does the most damage when the monarchs try to migrate - think of it like trying to run a marathon while infected with the flu - those monarchs that attempt the journey tend to die along the way. In this way, the migration in North America actually helps to keep the entire population healthy, by weeding out (most of) the sick ones each fall.

So the data we used to address this question came from a wide variety of sources. It actually wasn't citizen science data at all. Some folks may know that Sonia heads up the MonarchHealth program, which is where citizens sample adult monarchs with tape, and send these samples to her lab for analysis. This is a great program for science and education, and more can be found here - www.monarchparasites.org. However, we did not use these data for this project. Instead, we used parasite data that had already been collected by scientists (mostly Sonia, Jaap, and others). Over the years, this team has made many collections of wild monarchs, from various locations around North America. Each time they make a collection, they record the parasite prevalence of the monarchs. All of these data have gone into a giant spreadsheet over the years, which is what formed the basis of this study. In addition, a number of other scientists have also collected monarchs for their own projects, and we also used their data too - for this we either asked the scientists directly, or we copied the information they presented in their published studies. And finally, we also had data on parasites of monarchs that had been stored in the late Lincoln Brower's freezer! For decades, Lincoln had been collecting monarchs, and he had stored some of them. These go back to the late 60s, before we even knew where the winter colonies were! At various points in time, researchers have "raided" this famous freezer and taken tape samples from the specimens to know the OE levels. We had this data too. Collectively, we had amassed tape samples from nearly 60,000 monarchs! Below is a map from the paper which shows the various locations where monarchs were sampled.

Forgive me if I sound like I'm geeking out about this dataset, because I think I am. If you think about it, this represents the longest-running and most comprehensive dataset on parasite prevalence of monarch butterflies in the world! Each point in the map above represents a scientist who collected (wild) adult monarchs and then sampled their abdomens with a tape. The data literally span 50 years - five decades - from the 1960s to 2019. And we have data on various stages of the life cycle, including the breeding stage, the migration stage, and of course the overwintering stage. Also pointed out on the map is a site called St. Marks, which is a place where the Emory crew has been going for years to collect and sample migrating monarchs. So in other words, we had a boatload of data to work with.

Next, let's talk about what we did in the paper. Our initial goal was to simply statistically compare the overall parasite prevalence in the east and the west, to see if they were different. These next two graphs provide the answer to that question. These graphs show the average prevalence level (proportion that are infected) across all collections in the east or west. Note that we categorized the monarch infections into two groups - heavily infected, and "contaminated" based on how many spores were counted in each tape sample. Some monarchs have just a few spores on their abdomens, and we don't think these are true infections, but just that the monarch "acquired" some spores from somewhere else, such as if it mates with an infected partner, or maybe rubs up against an infected monarch in a roost.

As you can see, western monarchs have two-to-three times the parasite prevalence in their "subpopulation" as the east. At least 20% of them are heavily infected, and over 60% of western monarchs are carrying some spores! Or in other words, 3 out of every 5 monarchs in the west are carrying spores. That part was surprising, though to be fair, we did expect that there would be more parasites in the west overall - recall the migration thing. The western monarch migration is only a third of the distance of the eastern, so really, their migration does not do as good a job of "cleansing" the population each fall.

Note that this graph above also shows the "overall" infection level for the east, which is close to 6% on the left graph - BUT - I want to stress that this average takes into account all data from the entire gamut, including data from the summer, fall and winter stages. This is important because in the winter the prevalence is always low, because of the aforementioned migration-culling. And, the winter prevalence is actually not as important as the infection prevalence during the summer, as I'll explain later. So this 6% number is not the real number to look at here.

Now let's talk about the long-term trends, and here is where things get really, really dicey. First the west. Below is a graph showing the proportion of monarch that were heavily infected (true infections) per year, over the time scale that we had. Note that the western samples did not go back 50 years, but only to the 1980s.

Based on this graph, we determined that the OE parasite in the west has gone up and down and then up again. There actually was not a lot we could say about this graph, in terms of explaining this trend.

But for the east, things were different. And here, let me also point out that in some ways it is really not fair to be even considering the west in a separate category as the east, since the western subpopulation is genetically identical to the east (see this prior blog), and in terms of numbers, the west makes up a trivial slice of the overall North American monarch population, which is mostly in the east. In fact, the "eastern" monarchs make up 99% of the entire North American monarch population. Sorry, western folks. So anyway, for these reasons I think the results for the east should carry much more weight than those for the west.

Below, I'm going to put a graph that is a modification of the one in the paper. I put this graph together, using the same data, but simply made it easier to read. I also added one extra year of data, from 2020, which wasn't available at the time we did this study. So...this graph shows the long-term trend in the prevalence of OE in the eastern subpopulation since the 1960s, but only for the summer and fall (i.e. not the winter). I'll explain why later.

I hope you stared at this graph for a long time, because it's pretty sobering. This graph shows that for decades, the prevalence of infection with this parasite (in the summer) was less than 1% - that is, that about 1 in every 200 monarchs were infected. That's probably the way it had been for eons, where the parasite eked out a living infecting a few monarchs here and there. Nowadays though, 1 in 10 are infected during the summer on average, and sometimes it is more than that. Anyway, based on our analyses of these data, we determined that there has been a clear, statistically significant increase in OE prevalence since the early 2000s. This is across the entire eastern breeding range. Note that we did not include any samples of monarchs from southern Florida, where prevalence is nearly 100% - don't get me started on that.

OK, now are you ready for the really, really sobering news? In our study, we combined information on the breeding density of monarchs each year, with the subsequent size of the winter colonies, to try to determine the overall impact of the OE infection rate to the population size. We found a striking correlation between the annual summer infection rate and the relative size of the winter colonies. In other words, we found that in years when the summer prevalence was high, there was a lower-than-expected winter colony size, and vice-versa. In other words, higher summer prevalence means fewer monarchs reach Mexico (or the California colonies)! This is why I showed the graph of the summer and fall prevalence above, because it seems like this is the time period when this infection really matters the most. Not only that, we also did some math to try to figure out how many monarchs are being killed each year due to this infection spike. We determined that a national prevalence of 10% infection during the summer results in the loss of "tens of millions" of monarchs each fall during their long-distance migration. We don't know the exact number but we do know it is really, really high. Or you can do the math yourself - in any given summer there are 300-500 million monarchs throughout the country. And consider that these days, the winter colonies contain around 100 million monarchs, and then you can see the magnitude of this problem.

OK, now take a breath.

So you might be wondering what caused this sharp increase in the early 2000s? Well, we actually tried to find out too. What I haven't told you yet is that we also had performed a series of analyses where we tried to correlate these temporal trends with environmental variables, like precipitation and temperature, plus with other things like season length, and even monarch summer density (using data from the North American Butterfly Association). In a nutshell, we did find some evidence that prevalence tended to be higher in years when monarch density was higher, but this only explained the minor ups an downs from the graph above. It did not explain the dramatic increase in the early 2000s. In fact, we found absolutely no natural, ecological explanation for why prevalence increased so much.

There is one thing that happened around exactly the same time though, and it is not a happy thought. We did not go into this much in the paper but we did suggest that the cause must be anthropogenic (since there was no ecological cause). If you think about it, the late 1990s and early 2000s is roughly the same time when the entire monarch conservation movement picked up steam in the United States. That's right. I'll demonstrate this below as well. Now, keep in mind I said "roughly" here because it is impossible to know exactly the starting year(s) for this movement, and, because it really has been more like a slow-moving freight train. But, if you actually dissect some of the early literature and the research papers on monarchs (especially conferences), you'll see what I mean.

I took a close look at some of this early literature to try to find out what was happening with the world of monarchs throughout this 50 year timeframe. Of course we've known about the winter colonies since 1975. If you look at the scientific meetings that have taken place over the years since then, you can see what the overall consensus was at the time, and where the conservation energy was focused. Back in 1981, there was an international scientific meeting in Mexico, and also in 1986 there was a meeting in California. In both of these meetings, the conservation focus was nearly 100% on protecting the winter colonies, both in Mexico and in California. There was no discussion of anything north of Mexico. This is well-documented in the 1993 book, Biology and Conservation of the Monarch Butterfly. Next, there was another scientific meeting in Mexico in 1997, and this meeting is documented in a book - The North American Conference on the Monarch Butterfly (published 1999). In that meeting, you can see there was still a lot of focus on protecting the winter colonies, and you can also see the beginnings of some discussion of monarch biology in the U.S. For example, there was a presentation about the early days of Journey North which began in the late-90s, and another documenting the early stages of monitoring the larval monarchs in the U.S. There was also one presentation by someone from the National Wildlife Federation, who indicated that they wanted to begin incorporating monarch habitat and pollinator gardens into their wildlife garden certification program.

Looking further ahead, the next international conference on the monarch was in 2001 in Kansas (hosted by MonarchWatch, which was also in its early years). The presentations from that meeting are detailed in the 2004 book, The Monarch Butterfly: Biology and Conservation. The last book on my office shelf is the 2015 volume, Monarchs in a Changing World: Biology and Conservation of an Iconic Butterfly, which details the presentations from the 2012 scientific meeting in Minnesota. There are some very insightful reports and presentations in both the 2004 and 2015 books. Reading through these studies, it is clear that during these years, there was definitely more focus on research from the U.S. and Canada, and less emphasis was placed on protection of the Mexican forests. The reports in these two books also make it clear that the late-90s was a time where there was an uptick of volunteers for these early programs, which signifies a growing national interest in monarchs. Finally, the 2012 book has a report on a growing problem in the United States, the planting of tropical milkweed, which had been ongoing since the early 2000s.

To further illustrate my point, below I present a copy of the chart above, but in this version I have indicated the dates of significant events in the history of monarch conservation in the United States and Canada. You can judge for yourself if these events are relevant to the OE increase.

As you can see, there were a lot of significant events all starting to coalesce during the late 1990s and early 2000s. Collectively, these events all make it clear that this is the timeframe when the entire world seemingly turned its attention to the monarch. With this attention came a lot of grassroots efforts to monitor their numbers, join citizen science programs, donate to causes and pollinator programs, and a lot of genuinely good things. But, this attention also brought a lot of not-so-good things too, which I'll point to next.

Now I'm sure there's going to be a lot of finger-pointing after this paper circulates, and I know that this is going to cause a lot of arguments. Maybe it should, since this all needs to be aired and sorted through. But, the OE numbers simply don't lie, nor do our analyses. If it makes anyone feel better, I don't think there is any one thing that is causing the OE increase, but probably multiple things at once. From my view of the world, I see three anthropogenic practices that have all arisen during the save-the-monarchs movement, and each of these probably has some influence on this, because of what we know about its influence on OE numbers. They are, in no particular order: 1) commercial monarch breeding operations, where thousands of monarchs are raised, sold, and shipped around the country for release. These monarchs are often infected with OE (we've checked). And, these operations are always touting that their monarchs are helping to boost the population 2) homeowner rearing of monarchs for release. If people rear more than just a few monarchs, their kitchen-rearing operations have the potential to result in major OE spread. 3) planting of tropical milkweed. In places where it doesn't freeze, this milkweed has been proven to enhance local OE spread. Even though these places are usually more southern, migrating monarchs still pass through them on the way south, and/or north, and, they probably interact with the highly-infected resident monarchs there.

I'm winding down now, and I know this is a lot to take in. Here is a quick summary:

- Monarchs have a growing disease problem. This disease is wiping out tens of millions of migrating monarchs, leading to smaller than normal winter colonies.

- We know this increase is not being driven by any natural environmental factor, which argues the cause is anthropogenic.

- The timing of the increase coincides with the rise in collective efforts to save the monarchs in the last 15-20 years, including mass-rearing of monarchs and planting of tropical milkweed.

So as not to end on a low note, I think there is a strong lesson to be learned from this new paper - that is, if you want the winter colonies to be bigger, the answer is to reduce the summer OE level. This will probably have a much bigger effect than restoring habitat, even. And if you think about it, the other good news here is that anthropogenic activities can be managed, or regulated. The three main actions I listed here - commercial rearing homeowner rearing, and planting tropical milkweed, all can be curtailed, but only if we so choose to take action.

OK, I'm going to leave it here. Please share this.


Direct link to this blog entry:




The science of monarch butterflies

A blog about monarchs, written by a monarch scientist, for people who love monarchs