- Andy Davis
New study published: despite winter colony declines, monarchs are thriving in North America (really)
(picture above was taken by my mother!)
Yes, the title of this blog post tells you everything I'm about to tell you today. And strangely, this will be good news for some folks, not so good news for others. I'll explain why later on. Today's blog is dedicated to a newly-published study in a very high-profile journal, and it is sure to set the monarch world on fire. I know it has already been highlighted by various news outlets, including the New York Times, and National Geographic. This is a study where yours truly was part of a team of scientists who set about to assess the entire monarch population in North America, to determine if it really is in trouble. And yes, we found what the blog title says - no, they are not in trouble. Nor are they close to extinction, and, they should never even have been considered for endangered species protection! Crazy, right? So, sit back and get ready to have your mind blown. Fair warning, this is a long post, since there is so much to cover.
Let's back up a bit first. For long-time readers of this blog, you'll likely see that this post has a familiar ring to it. That's because the project I'm talking about is actually the third iteration of a long-running effort on my part and some of my colleagues, and I've described some of these earlier versions before. For starters, recall that in December 2020, I had completed and submitted a year-long effort to review all available evidence on the status of the monarch population in North America. My assessment from those data was that the population is faring surprisingly well (see the blog). That paper was ultimately rejected by the journal, because reviewers did not think it was rigorous enough (they wanted more statistics, not a review). Fast forward to 6 months later, and this time I had collaborated with a team of entomologists to submit a second iteration of this effort to a journal, and in this version, we had used very fancy (and rigorous) statistics to fully analyze these same datasets. The conclusion of that paper was the same - monarchs are doing surprisingly well, despite declines in winter. My blog on that one is here. That paper too was rejected, and that time, the reviewers didn't think we had examined the right data for the job.
I guess the third time is a charm, because our latest version is an absolute masterpiece, combining cutting-edge analytics with a massive dataset on monarch abundance, and not to mention that we brought in a statistical ringer, Dr. Tim Meehan, who works with Audubon Society Christmas bird count data for his day job. And yes, this time the paper was accepted after rigorous peer review (three separate reviewers too!). And as you can guess, this third version fixed all of the issues that were raised in prior iterations. In fact, combining these reviews with the prior iterations, I think you could say that this paper went through an enormous amount of peer-review to get to this point.
Our paper was published in the journal, Global Change Biology, and here is a link. This paper should be open access, so you should be able to download a copy and read it in its entirety. You'll see that I am the last author in this version, and my colleague, Mike Crossley, is now the lead author. Mike was a postdoc here at UGA when this project began, but is now an entomology professor at the University of Delaware! We felt that he put in more time on this version (especially with the stats), and so he deserved most of the credit. You'll also notice that one of the co-authors is Jeff Glassberg, who is the head of the North American Butterfly Association (NABA). It was the data from this group that was the basis of this paper. In this version, we focused solely on this one dataset, because it was so massive, and allowed us to examine how the abundance of monarchs has or has not changed over time, while also accounting for variation in observer effort, and even simultaneously examining other contributing factors, which I'll describe later.
I've talked about the NABA butterfly data before, and most knowledgeable monarch lovers know of this organization, but just to get us all on the same page, this group has its volunteers survey for butterflies at sites throughout the U.S. and southern Canada. I'll put a screenshot below of the map of monitoring sites taken directly from their website (naba.org/counts). You can see there are a lot of sites! Not all of these were included in our study, but most were.
The NABA volunteers do their surveys each summer, usually in July, and volunteers survey all adult butterflies they see within a really, really big circle, that is 15 miles in diameter. I used two reallys there because sometimes people get the erroneous impression that these survey sites are just someone's backyard, or a single butterfly garden. In fact they're really, really, really big areas, filled with all kinds of landscapes and habitats, including cities, parks, rural roads, farm fields, etc. See the second screenshot below of one count circle from the middle of Kansas. Jeff tells me that the original idea with these circles was to have people sample as many habitats as possible and even have duplicate habitats sampled, in case one gets destroyed or something over the years. Anyway, these volunteers go out each year to these circles and count all butterflies they see in them - every species - including monarchs. This part is important. They are not at all focused on just monarchs. That means they are completely objective.
So, my team had access to these butterfly survey data and we pulled all of the observations of monarchs out - all 130,000 of them (!), over 25 years. There were monarchs seen in over 400 different sites across the two countries. Note in the map above that there are circles all over North America, including many in the northeast, lesser numbers in the west, and importantly, lots in the American Midwest, which is the center of the monarch breeding range. This is a key thing to keep in mind, because this is a heavily agricultural region, with lots of fields and crops, which used to be great places for milkweed. Note that there are many circles in these regions, so that if there had been changes in the number of monarchs there, we would be able to see that by looking at the data from these sites.
I know I'm droning on and on about the circles and data, but this part is crucial to the story here, and it has long been a point of argument with some of us monarch scientists. There are some scientists who feel that the NABA data do not accurately represent the monarch population status, for reasons I'll explain later. We obviously didn't think this, and in fact, this new study proves their argument wrong, which I'll explain later too.
OK, so I told you about the monarch observations, and so next we wanted to know if the number of monarch sightings has changed over the years, across the entire breeding range - yes, all of it. We wanted to know if the entire population in North America has undergone any changes. We didn't even separate the continent into east and west (because they're all the same species). Note that this is something that no other study has done before - to examine their numbers at such a massive, continental, scale. This is not something that is easily done, however, and this is where our stats gurus came in - Mike and Tim. At the same time we were looking for changes in abundance, we had to account for some inherent problems in these data, like when there are more people doing a survey, there are more butterflies seen. And, the number of observers has changed over the years too. Not to worry though, Mike and Tim were able to statistically account for these issues.
In addition, we also were conscious of the importance of weather (temperature and precipitation), and how this can change each year, thus leading to changes in butterfly numbers. This weather effect has been found before in other studies of monarch survey data. So, in our statistical models of monarch abundance changes, we included these variables, to ask if either of these (temperature or precipitation) affects the abundance of monarchs at any of the sites across the breeding range, and/or, if the annual variation in these has affected monarch numbers. In other words, we were simultaneously addressing a very important question - how does climate change affect summer monarch numbers!
Finally, we also knew that there have been changes to the agricultural regions over the years, specifically the rise in roundup-ready crops, which has led to the reduction in milkweed in corn and soy fields in the Midwest. This is not something that is in dispute at all - there really has been a loss of milkweeds in corn and soy fields over the years since roundup-ready crops have been adopted. Genetically modified roundup-ready crops came onto the market in the late 1990s, and these crops allow farmers to spray roundup directly onto their fields, to remove all weeds, including milkweeds (but not the crop plants). So in our study, we also incorporated other data on how much roundup (glyphosate) has been used each year within (or near) each circle. We figured that this number should be negatively correlated with monarch abundance, since fewer milkweeds in a region would lead to fewer monarchs.
So in the end we had a very massive and state-of-the-art statistical analysis that essentially asked the following questions - 1) are the numbers of breeding monarchs declining in North America, 2) does the amount of glyphosate used near a survey site impact monarch abundance, and 3) does temperature or precipitation at a survey site affect monarch abundance?
Keep in mind that for this post, I have purposely simplified the description of the stats here, just to make it clear to everyone. In reality the analytics were extremely complex.
Anyway, here is what we found.
I'll start with the biggest finding. When we looked at the long-term trends in the 400+ monitoring sites, we found there were some places where breeding monarchs were declining, but there were some where monarchs were increasing over time - see the map below. Importantly, when we lumped them all together (i.e. the entire population), we found there was no overall decline in numbers of monarchs seen in the past 25 years, going back to the mid 1990s. In fact, there was an overall positive trend of 1.3% per year. Over 25 years, that's about a 30% increase!
Here is a map of the sites where monarchs were counted, which is color-coded to show where monarchs were declining (orange/yellow) or increasing (blue). The darker the color, the larger the change. The highlighted squares are where the trend was most pronounced.
Looking at this map, you can see that there were some spatial patterns that emerged. First, there were general increases in most areas to the far northern part of the breeding range (mostly Canada, northern Minnesota), and also increases in the southeast and southern Gulf states. In Florida, the monarchs are increasing significantly. There was an inexplicable region of decline in the Massachusetts and New York region. In the west, there was a region in middle California that has seen long-term declines, and in fact, this is the same area where another long-running butterfly survey (conducted by Art Shapiro) has also found long-term declines. This consistency in findings between our study and the Shapiro data, gives us confidence in these results. But look closely at the sites north and northeast of California - the data was sparse in that region, but all sites showed increases in abundance!
If you squint at the map above, you can also see that there was a region around the middle part of the range (the Midwest) where there has been more than a few declines. This brings me to the next big finding. There was a statistically significant negative effect of glyphosate useage on monarch abundance, and this effect was most pronounced in parts of the agricultural Midwest, as you can see above. This is the same pattern that has been found in other studies, and it is consistent with the idea that the loss of milkweeds in corn and soy fields has led to declines in monarch abundance near these fields. I'll put another screenshot below of a map that shows this glyphosate effect, and you can see the strongest negative impact (pink squares) is in the upper part of the Midwest. Note too that all colors here are negative, indicating that this glyphosate effect was overall negative. In other words, in survey sites where higher levels of roundup was used, there tended to be fewer monarchs counted.
The roundup issue deserves further comment here for another reason, which I alluded to earlier. There has been a long-running argument over these NABA data, and whether they accurately reflect the breeding population size, and it all has to do with this chemical. The argument goes that because farmers in the Midwest have adopted roundup-ready crops and removed all milkweed in their fields, this led the monarchs to shift their distribution away from farm fields and into the NABA survey sites over the years, which gives the illusion that the numbers of monarchs in those circles has not changed. This sounds logical, though there never really was any direct evidence of this "monarch shift". And in this study, this argument is finally proven false anyway. We found there really was a decline of monarchs in the count circles that were in regions of high roundup use. So how can increasing roundup use over time have led to a net increase in monarchs in these circles?
Still on this roundup issue, I also want to point out something else that makes the argument from my colleagues ridiculous, and you can see it here too. The argument goes that the monarchs have moved from the surrounding farm fields and into the count circles "because the count circles contain the only remaining milkweeds left for monarchs". Go back and look very close at the landscape of the circle I showed above. See all of the farmland that is within the circle? The landscape within the circle is basically the same as the landscape outside the circle. And, this is essentially the same story for all circles within the Midwest. In other words, it is ridiculous to think that the circles somehow contain all of the remaining milkweed. Again, I think this argument is flawed because it fails to consider just how vast these survey circles are.
Anyway, that's my take on this roundup issue. In my book, this case is closed.
Moving on from roundup, here's where our findings get even more interesting. As we suspected, the effect of temperature was also important for determining whether monarchs were increasing or decreasing, but this effect was positive in some places, and negative in others. In particular, this temperature effect appeared to have a positive impact on monarch abundance in the Midwest, which is opposite the glyphosate effect in this region. Below is the temperature effect map, and note the Midwest, where there are lots of very positive squares. A similar pattern (temperature affects monarch abundance) was actually found in another recent study (led by Erin Zylstra in 2021), and so this consistency also gives confidence in our results.
There were some effects of precipitation too, which I won't get into here.
To summarize the main findings then, we found that increasing usage of roundup does (or did) lead to reductions in monarch abundance, especially in the Midwest, but at the same time, increases in temperatures over the years have also positively impacted monarch abundance in this same region, which has led to a situation where there has been no overall change in monarch abundance, especially when you consider the entire breeding range. I guess you could also say that global warming has aided the monarchs in the Midwest! I know this is odd to say this, but I'm just telling you what the data show. Or, you could also say it thusly - the growing use of roundup on farm fields over the last 25 years did lead to a slight reduction in monarch abundance, though this effect has been overstated. It did not lead to the destruction of the monarch population.
OK, so I know what you're thinking - how is this possible? We've been hearing for years that their habitat is disappearing, that the entire population is near collapse, or close to extinction! Well, I've actually given this a lot of thought since working on this project, and there is a lot to unpack on that issue - like how the entire world (including you) became convinced of this, despite the reality. There is so much to discuss there, I'm actually considering devoting my very next blog to this topic, so stay tuned.
For now, I think it is safe to say that this entire notion - that monarchs in North America are in trouble - is wrong. There clearly are long-term declines in winter colony sizes, both in California, and in Mexico, however, these declines appear to have given too much emphasis, because they do not appear to be affecting the collective breeding population of monarchs. Again, for long-time readers of this blog, this is a familiar theme, and I even posted a while back that I myself do not even pay attention to the winter colony reports anymore.
There are a variety of reasons for why the winter colonies are declining, despite a thriving breeding population. This is not something we specifically addressed in this paper, but I have talked about this at length in this blog site. What I believe is happening is that monarchs are experiencing problems during the fall migration, which leads to fewer monarchs actually reaching the winter colony sites. This gives the illusion that the population has declined, but in fact it is more of a migration decline. In support of this, recall the recent study showing how the monarch parasite (OE) has risen within the population, and this parasite reduces monarch flight potential and migration success. In fact, we know for sure that higher OE levels actually lead to lower winter colony sizes! And don't even get me started on how many millions of monarchs are killed on roadways each fall.
Buuutt... despite these migration declines and smaller wintering colonies, the monarchs are still doing well every summer - really well, in fact. Did you know that monarchs are one of the most often-observed butterflies in the NABA surveys? In fact, based on our examination of the NABA data, monarchs appear to be doing better than 320 other species! From all accounts it looks like they are able to continue to successfully repopulate their spring and summer ranges, despite the small winter colonies. This is likely because of their tremendous reproductive potential - a single female can lay hundreds of eggs. In other words, it does not seem to matter that much if their winter colonies are small. Really, for all we know, these colonies now could be close to the original size that they were before humans cleared the landscape of this continent within the last century.
So what does all of this mean for the official conservation status of monarchs in the United States and Canada? The USFWS declared in 2020 that monarchs should be considered as threatened on the endangered species list, though they held off doing so for now. And, I know that their counterpart in Canada is also close to listing monarchs as federally endangered. This is already the case in two provinces, and also in California. These assessments are wrong. Period. They were all based on the assumption that the trends we have been seeing at the winter colonies, both in Mexico and in California, are representative of the entire species' population. We now know they are not. Even in California, this is the case - look again at the main map above showing the trends, and you'll see that the breeding cohort in the Pacific Northwest (the "breeding range" of western monarchs) is not declining. What I'm saying is that the size of the winter colonies appears to have no bearing on the size of the breeding population. Or another way to put it is, that the winter colony sizes actually do a piss-poor job of telling us how the population is doing. Yes, those exact words. You could even say that the declines at the winter colonies are a red herring - they are misleading. I think you get my point.
OK, so now let's talk about you, the blog reader, and what you should take away from all of this. I know this is a lot to take in, and I may have thrown your world asunder with this information. If you are a long-time or even a new reader of this blog, then that probably means you are fond of monarchs, and you yourself have probably spent considerable time (and money) working to "save" them, by planting milkweeds, maintaining a butterfly garden, or doing something similar. Are all of those efforts wasted because of what I just told you? Maybe, maybe not. There is some good that can come from these things, and certainly a lot of other insect species can benefit from your butterfly garden. Though, there are some negative things that can come from all of this too - remember the latest study on the OE parasite (see the blog here). Recall that the dramatic rise in OE prevalence has directly coincided with the mass efforts to save monarchs, including rearing monarchs in captivity and planting non-native milkweed. So yes, some bad things can come from people "just trying to help" monarchs.
You've also been hearing for years now that the milkweed is all gone, and our data also demonstrate this this too is not quite true. Think about it - given that the overall numbers of breeding monarchs is not going down, then this inherently means that the breeding population currently has all of the resources it needs to be stable. In other words, it has enough breeding habitat, and importantly, enough milkweed. Yes, there has been some declines of milkweed in farm fields, but, it looks like the importance of this issue has been dramatically overstated. In other words, monarchs are not declining because of a lack of milkweed. There is no lack of milkweed, really. And recall the study from just a month or so ago, where researchers determined there were BILLIONS of native milkweeds in the western part of the continent (see the blog here).
Finally, this brings me to what I had mentioned at the outset, about how some people will find this news to be upsetting, for some strange reason. This is a reaction that I've gotten before when telling folks about these same data, or when similar conclusions were drawn from other data, and frankly, it is very, very weird. In essence, people seem to get mad when hearing this GOOD news about monarchs - that they aren't really in trouble. I'm going to touch on this too in my next blog, but I think this reaction is because people are enjoying "saving" monarchs too much, and then when they hear that monarchs don't really need to be saved after all, that takes the wind out of their sails. But again, isn't it a good thing that they seem to be doing well? It's weird. Anyway, please don't send me hate mail for telling you that monarchs are doing fine.
OK, I'm finally winding down so I'm going to put a quick summary below, in case anyone wants to just share these highlights. Sorry to make you read to the end to get the highlights!
Take-aways of this new study:
The abundance of breeding monarchs in North America has not declined over the last 25 years. This means that the long-term declines of monarchs at wintering colonies - both in Mexico and in California, are not indicative of how the summer population is faring.
If the abundance of breeding monarchs has not changed over time, this inherently means the monarchs are not suffering from a lack of summer resources - they currently have everything they need, including milkweed, to produce their summer generation(s).
This conclusion also means that monarchs should not be considered for endangered species listing either in the U.S. or in Canada.
Here is my last and biggest take-home message for you, which is something that considers both the findings from this study, plus the ones from the last OE parasite study. This is just my (informed) opinion, but based on these data I would say that monarchs do not really need our help, they just need to be left alone.
OK folks, thanks for reading. Please do share this article widely.
Direct link to this blog entry: