- Andy Davis
I reviewed 20 studies that measured monarch abundance over time. They don't show what you think
I hope you are sitting down, because you'll need to be for this news. I have some pretty earth-shattering findings to tell you about concerning the monarch population in eastern North America. Before I begin, I think you should get yourself a cup of coffee, sit down, and strap in - this will be a long post, and deservedly so.
Ok are you sitting down, and ready to have your mind blown?
The study I'm going to discuss today was conducted by yours truly, and it is the result of an entire year of effort. It was just now submitted to a preprint journal - link here to see the paper. This means that the paper itself has not yet gone through official peer-review, but the initial manuscript and supplemental file are now public. This is a weird intermediate stage between being just a draft on my computer versus a full-fledged scientific paper. But as you will see, even at this early stage, the findings are highly relevant to the latest happenings with monarchs. In fact, the timing of everything going on right now is the reason I have made the findings public even at this stage.
So what is happening right now? As most butterfly-lovers know, monarchs are currently being considered for federal protection in the United States, and also in Canada (though fewer people know about the Canadian status). In the U.S., the Fish and Wildlife Service has just wrapped up their report on the status of monarchs and we are awaiting the government decision on whether they will be listed as Threatened. In Canada, their scientific committee (COSEWIC) has already met and recommended the monarch be "uplisted" in that country to Endangered. Monarchs in Canada are already considered threatened. This uplisted recommendation was actually made a few years ago, but the official government decision on this is still pending in Canada. So collectively, both countries are on the verge of declaring the monarch to be in dire trouble, and need of federal protection.
I have talked with officials in both countries about these pending decisions. From my understanding of the reviews of the monarchs' status in both countries, and from reading between the lines, it looks like these pending decisions are based largely on the ongoing declines in the winter colonies of Mexico, and also in California. These declines are certainly very dramatic and, they have been ongoing for many years now. They are also very, very, VERY well publicized. Every day, it seems there is a new news story touting this dramatic decline. It is no wonder then, why both governments would be considering that this species needs federal protection.
Despite this public perception, the science on these declines has never been straightforward, and sometimes even conflicting. Even 10 years ago now, it was clear that the declines we are seeing in Mexico are not consistent with long-term censuses of fall migrants. There was a (now infamous) paper written by Lincoln Brower and colleagues back in 2011 (link here) that described the long-term decline in Mexico. After reading it I then posted a rebuttal paper that showed how over the same time frame there had been no decline in the fall. Over the years since those initial studies, there have been a few others that have also shown how monarch counts conducted during the summer seem to show no declines too. So basically, it's complicated. That is why I have always advocated that to get the best picture of the monarch's long-term status, we (meaning those making these decisions) should be using all available data, not just the winter census from Mexico (or California).
So, earlier this year I set out to put my money where my mouth is. I set out to conduct the largest and most comprehensive review of the status of the eastern North American monarch population to date, in an effort to uncomplicate this picture. I focused exclusively on the east, since this is where the majority of data are available, plus this is the largest and most well-known population. The timing of this review is also important - given the pending decision on whether to federally list monarchs (in the US, or in Canada), I figured we should better make sure we have the story right. And, since these decisions are looming, I figured that this information needs to become public right now, even before full peer-review is complete.
My goal with this review paper was to assemble as much data as I could find on the abundance of monarchs over time, and then put all of this data together to see if I could find out where in the annual cycle the monarchs are in trouble, if at all. I also thought that if there were any consistencies in the data, this would be important to know too. I have spent the past year furthering this goal, and frankly, it has been all-consuming and a much, much bigger project than I initially imagined. To obtain the data, I went about it in various ways. In some cases I directly contacted leaders of citizen science projects and asked if they would share their data on monarchs with me. In others I copied data that had been presented in scientific publications (which is public information). I also gathered certain data on my own from publicly-available online records, such as those in the Journey North program (https://journeynorth.org/monarch/). And of course, I also included the data from the winter colonies, which is public too.
In the end, I assembled a total of 20 (!) different long-term datasets on monarch abundance from across their annual cycle! To be honest, I was even surprised at how many there were. This collection included counts of adult monarchs in the summer, counts of larvae, estimates of breeding range size, counts of migrating monarchs in the fall, and even counts of monarch specimens in museums. And, these data came from all regions of the monarch range in eastern North America, including in Canada. It ended up being a truly astounding collection of data, which mostly had been collected by folks just like you - citizens who have informally contributed random sightings of monarchs, or participated in long-term citizen science programs.
Below is a simple table from my review paper that shows where the majority of the data I used came from. For long-time monarch fans, the project names on this list should sound familiar.
By the way, I bet you've noticed that the description above of the winter colony data looks odd. That's because I also included in this review some "new" data on the size of the winter colonies that had been collected prior to the start of the WWF monitoring program in 1993. A lot of people are even unaware of this, but the colonies (or at least some of them) were measured for many years leading up to the 1990s, by independent researchers, including Lincoln Brower himself! These records go back to 1976, the winter following the discovery of the colony location! The records of these measurements are mostly public, appearing in certain older publications, including a student thesis in Mexico. I included citations of these in my review. I think one of the reasons these early records are not typically included in the public descriptions of the colonies is that they are not easily comparable to the later years. The primary reason is that the early assessments were not of all of the colonies, because of course not all of the colonies had been found in the early years. But, this can be statistically accounted for, which I did in my review. Basically, in my review I wanted to be as thorough as possible, and so I included these early records.
One other thing that should strike you about the table above is that there really is a heck of a lot of monitoring at each stage of the annual cycle - some people forget just how many eyes their are on this one insect! In fact, this might be the world's most well-monitored insect!
Ok, now I'm going to cut to the results of this review. The thing to keep in mind here is that for each dataset I looked to see what it showed (statistically) over the time frame of the study or project, or in other words, did the data show a decline, an increase, or no change in monarch abundance over time.
Here is what I found:
There were 4 datasets that showed long-term declines in monarch abundance. First, the winter colony data did show a decline, even with the addition of the early records. I actually examined these colony data in several ways, and even with these archived records, it seems there is still at least a modest decline at the winter stage. But, we kind of already knew there are declines at the winter stage, so this wasn't too surprising.
Next, there were three datasets from springtime assessments of monarchs or of the spring recolonization size. These were projects that had tracked adult or larval abundance during the spring recolonization phase. The data from each of these projects indicate monarch abundance during spring has been slowly declining over time, which sort of makes sense - if the abundance of monarchs is declining in winter, it should also be declining in the early spring, since it is mostly the same cohort of monarchs in both seasons.
But, here's where the story changes completely. Of the 11 different datasets that track abundance of monarchs in the summer, NONE of them showed declines over time. Not one. There has been no change in the density of larvae throughout their range, no change in the number of adults in the breeding season, no change in the geographic size of the breeding range, no change in the annual number of museum specimens of monarchs, and the list goes on. And, keep in mind that these are surveys conducted at sites throughout the breeding range, including in the American Midwest, which is the central portion of their breeding range. Moreover, these included censuses conducted in Canada, and there were even some datasets from there that showed increases! Collectively, this evidence shows that the size of the summer population of monarchs has not decreased, nor is it currently decreasing.
That's not all. Next, of the 5 different long-term studies where migrating monarchs are counted in the fall, I also found little to no evidence of declines there too. First, there has been no decline in the average size of migratory roosts throughout the fall flyway, based on sightings provided to Journey North since 2003. There has also been no decline in the numbers of migrating monarchs counted at Cape May, NJ, at Peninsula Point, MI, or at Point Pelee, ON. The only exception I found was is in the long-term censuses conducted at Long Point, ON, by the team at bird Studies Canada. There, there has been a very minor decline at one of the two census sites on the Long Point peninsula, but strangely, the other site (on the same peninsula) shows no decline in fall monarchs. When these two sites are averaged together, the decline is not apparent. It is important to point out here that most of these censuses come from northern locations, which means they are surveying monarchs close to the beginning of the migratory journey. All in all, the evidence here indicates that the starting size of the fall migration has not changed over time.
I'm going to paste a figure from the review that visually shows the long-term trends from a subset of the datasets. For each dataset listed below, I included a simple map which shows the approximate geographic area which is covers. Obviously, more weight should be given to those projects that cover more range. But even the small-range studies are important too.
The thing to take away from this figure is the fact that each of the projects which track monarch abundance in the summer shows how the monarchs go through up years and down years over time, but overall there is no downward trend in any of them. The exception might be the winter colony graph above, in which I included the early records. If you simply plot the total colony area for each year, going back to the beginning (i.e. winter of 1976), this is what that graph looks like. When graphed like this, it doesn't look like there is a decline at all! However, keep in mind that when you account for the varying colony measurements over time, there is indeed a modest, statistical, decline over the 45 or so years. It's certainly not a 90% decline though.
I'm not spending much time in this blog talking about these winter colony data, because I actually did not spend a lot of time focusing on the colony data in the review paper. Keep in mind there were 19 other datasets to parse through. And, it seems like the winter colony data don't really matter that much anyway, when you consider all of the results from all of the data. So don't kill yourself looking at that one graph!
The take-home finding from this whole review is that regardless of how the data were collected, of who collected it, or where it was collected, the evidence from all available data we have all points to the same conclusion - the eastern monarch population is not declining, even despite any winter-season declines that might be happening. Let me repeat this in case you didn't read it correctly - the eastern monarch population is "doing fine" in the summer and fall, and is not at all in trouble. I know - this is mind-blowing. Yet another way to put it is that the declines we have been seeing in Mexico over the last 25 years are not representative of what is happening in the US and Canada. If anything, they are misleading.
I feel extremely confident in making these statements above. To me, one of the strengths of this review was the fact that there is such amazing consistency within the various monitoring programs. The fact that a total of 16 different programs and studies all point to the exact same conclusion - the abundance of eastern monarchs in the summer and fall has not declined - is the biggest strength of this paper. And, since each project recorded data differently - from different regions, using different observers, different times of year, etc. - but yet they all show the same thing, this means the results are practically irrefutable.
The other reason I feel confident in this conclusion is because of something I haven't told you yet. There is another new study that has just been completed (and which I referenced in my review), but it has not yet been published. It is a fantastic new genetic assessment of archived monarch specimens from the 1970s, and where researchers looked for evidence that the "genomic diversity" of monarchs has changed over time. Genomic diversity tends to go down in small populations of animals. I'm not going to get into the details here, but the results of this new study also demonstrate that the eastern monarch population has not declined! So now, there is parallel evidence, and, from the monarchs' own DNA! Anyway, I cited this new project since it clearly bears on the findings of this review. Stay tuned for when that new study comes out.
OK, now here is another mind-blowing bit of news to take away from all of this new evidence. Since monarch abundance has not declined in the summer, this can only mean one thing: the milkweed supply for the summer population has not diminished, AT ALL. Wait, what? Yes, I know how this sounds, but think about it - if the milkweed supply had really declined in the last 25 or so years, it would have lead to a corresponding reduction in the summer population. I just don't see that reduction anywhere in any of the datasets, including the MLMP data, which does not show a decline in egg or larval density in the Midwest.
Don't get me wrong, I know that the milkweed in corn and soy fields has diminished in the agricultural regions because of the increasing usage of roundup-ready crops in the last two decades. This is not in question at all. What the monarch data are telling us though, is that the monarch population simply doesn't care. They appear to be doing just fine even without the agricultural milkweed. Thus, it looks like they were never as dependent upon that agricultural milkweed as we all once thought. I know, right? Mind-blowing.
I know this all sounds crazy and it goes against everything you thought you knew about the status of monarchs (and milkweed) in the east. But there is just no way to get around these numbers, and these trends. The eastern monarch population is really not in trouble, despite the losses that have been happening at the winter colonies. Those winter losses do not seem to be affecting the size of the eastern breeding population. In fact, in my paper I actually pointed out a couple of reasons for why these winter "losses" may not even be real!
Now for the western region of the range, things may be a little different. There is certainly less available monitoring data in the west, but so far things do look pretty bad for those monarchs. But, I will say that most of the information we have on western monarchs is based on counts of overwintering monarchs in California. If there is one lesson to be learned from the two new studies here (my review and the genetic study), it is that the winter colonies do not tell the whole story of the population - they do not really reflect what is going on during the rest of the annual cycle. Also, keep in mind too the other recent genetic study that showed that there really isn't a distinct "western" population after all. But anyway, my take is that things do seem bad for western monarchs.
One more thing regarding the west - I should point out that lately I've been hearing from a number of monarch researchers based out there (in Washington, Nevada, and California), and they each tell me exactly the same thing. They have each personally seen "plenty" of milkweed in their region but no monarchs. So at this point, no one really knows why western monarchs are not doing well, but it does seem like it has little to do with a lack of milkweed...coincidence?
OK, whew. this was a huge blog, but deservedly so, for such a huge development. I would encourage you to go back and re-read this blog so that you are fully-aware of all of the details. And most importantly, you should also download and read the actual paper to see the details, methodology, results, etc. It will behoove everyone to do so, so that everyone is fully-informed during the coming discussions over this new science. So, please don't just read the abstract of the paper or (even worse) just the title. It is fully online, so there is no excuse not to read it. Here is the link again.
Be ready for these discussions. There WILL be discussions over this new information. There really needs to be, especially given the importance of this issue - given the pending decisions over the status of the monarch, it is imperative that we get the science right before moving forward with conservation actions.
Direct link to this blog entry: