top of page
  • Andy Davis

I study monarchs for a living. I no longer pay attention to the winter colony data. Here's why.

Hi everyone,

The annual estimate of the winter colony size for the eastern monarch population was just released by officials from WWF Mexico, and this has had everyone talking - or at least, in between talking about covid-19! I've gotten more than a few emails and Facebook correspondence about the latest monarch numbers. Here's my take, for those who wish to hear it.

The way I phrased that last sentence was no accident. I am aware that the topic of monarchs, and their conservation, is emotionally charged, and I bet just about everyone reading this has some sort of deep connection with these butterflies. Therefore some people may not want to hear my take. That's their prerogative. I've always said that blogs are like TV channels - if you don't like what you're watching, turn the channel.

I, at least, feel like my opinion on this topic matters, but not just for the fact that I've been studying monarchs for 20 years. It's for a reason you may not expect. Believe it or not, I don't really have that same deep connection with monarchs, even though I study them for a living. That may sound weird, or even callous, but in fact, it is really a beneficial thing, from a scientific perspective. It means that my judgement on the science around them is not clouded by emotions. And most importantly, it means that I can tell you what their data shows without any bias.

OK, so the latest estimate from Mexico shows that the overwintering colonies were on the low side this past winter, which seems to have taken everyone by surprise, given that we all remember just how many monarchs there were last summer. Recall there were monarchs everywhere last summer and people were reporting "more monarchs than in recent memory." Indeed, from looking at the sightings of monarchs on Journey North, it was clear that their breeding range last year was absolutely huge. Because of this, everyone expected to see a correspondingly large winter colony estimate.

Here's the thing, in order to have big colonies, the monarchs need to be able to get to these mountaintop locations safely! Last year, that didn't happen. Keep in mind that the entire journey to Mexico lasts up to 2 months and traverses thousands of miles of sometimes inhospitable landscapes. And sure enough, last year there were large sections of the flyway that were experiencing drought conditions. Drought means fewer nectar sources for refueling the migratory flights. On top of that, there are other dangers that either kill monarchs or hinder their flight, such as car strikes, water-crossings, and increasing levels of the OE parasite, just to name a few. All of these things have been steadily increasing over the years so that the fall migration to Mexico is becoming so fraught with danger that it's surprising that any monarchs reach the destination. So the reason we saw fewer than expected monarchs at the winter colonies was because of the critical nature of the fall migration. I've always said - monarchs don't have a production problem, they have a migration problem.

For some folks reading this, this may go against what you thought you knew, or maybe what you've been told, or read. For years, the story has been that the shrinking winter colonies in Mexico foreshadow the impending demise of the eastern monarch butterfly. It seems like every time the colony size goes down (or up), there is a headline that shouts doom and gloom (or triumph) - "Monarch population shrinks again" This year was no different. The problem is that these headlines, and the organizations that tout them, are using outdated science. They are basing their stories on the information we had to go on a decade or more ago. In those years past, we had a very rudimentary understanding of what these colonies meant. In a nutshell, it was generally thought that the winter colonies in Mexico represented the "entire" collection of eastern monarchs. We figured that since they were all conveniently holding still for us in these trees, then we could simply measure their colony size, which would "represent" the sum of the entire population for any given year. This seemed like such a convenient measure!

But, as we saw this past winter, this colony area measurement does not always work. In some years, the losses during migration are so great that the colonies are much smaller than they could have been. This is why I, and others have argued publicly that these increasingly severe migration losses are what is really causing the declines in colony size.

Incidentally, if you want to see a really, really thorough piece of journalism that describes the ongoing scientific debate over the colony declines, check out the story by Gabe Popkin in Scientific American this month. I thought he did a great job summarizing the history of this issue. Since it's a subscription magazine, I'll put a link to my own pdf copy of the story here. Caution - it is a large file!

OK, so here's the thing you really need to know about the winter colonies, and this is sort of like the silver lining here - the low winter colony sizes we see now don't seem to mean anything, at least anymore. I mean this from a statistical sense. Keep in mind that I'm not saying the data are wrong, or that the colonies are not being measured properly, or anything like that. What I mean is, the size of the colonies from one year to the next does not predict, or tell us anything about how many monarchs we will see the next summer! This is a conclusion that has been reached at least four times in four different studies, using four different datasets now. The most recent one was one that I blogged about last fall (link here). In that study, scientists used monarch abundance data from Ontario and examined what factors predict how many monarchs occur there each summer. They found that the size of the winter colonies in Mexico did not predict summer abundance. Another study that came out last year also looked at records of monarchs in Canada (link here to my blog) and asked if the size of the winter colonies predicts the breeding range size in Canada - it didn't. Two other studies conducted in the last 5 years, conducted by the Zipkin lab, and using still different datasets (from the U.S.), also showed the same thing. I blogged about these too. They also asked if winter colony sizes predict summer monarch abundance, and the answer was no in both studies.

So now we have four different scientific studies which all show the same thing - that the size of the winter colonies has no bearing on how many monarchs are produced each summer, either in the U.S. or Canada. In the world of science, this is about as concrete as you can get for a conclusion.

Incidentally, these same studies indicate that what really matters, or what predicts the size of the summer population, is the environmental conditions during the spring migration. The spring migration seems to be much more important here, which makes sense, since this is the time when the returning monarchs are laying eggs and the entire population is recolonizing their breeding range. If the conditions are good (i.e. no droughts, especially), then the population does well that year. It seems to have nothing to do with how many monarchs we start with (i.e. the winter colony size).

I can add further arguments here for why the colony size doesn't matter. Does anyone remember the "winter crashes" that have occurred in the last two decades? There have been a few, and there was a big one in 2002, where a whopping "80%" of the winter monarchs died from a freak storm that hit the colonies. The world was shocked, and I recall there were doom and gloom headlines everywhere. And even pictures circulating of piles of dead monarchs! But then, look what happened next - later that summer, the breeding population was pretty much back up to full strength. In fact, if you look at the counts of monarchs from any long-term dataset on summer abundance (I have), you have a hard time seeing any dip in numbers for that summer. That monarchs come back from these "crashes" only serves to strengthen the statistical arguments above - it doesn't matter how many monarchs there are in Mexico.

So to summarize the points above, based on the science (and our own eyes), we now know that the winter colonies are not a reliable barometer of the previous summer population. We also now know (exceedingly well) that the winter colony size does not tell us how many monarchs we will see in the next summer population. For these reasons, I no longer pay attention to these numbers.

Lastly, let me reiterate that some folks may not like hearing this cold hard truth about the colonies, their data, and this science. I know that some people have taken the long-term decline in winter colonies as a mantra, or a personal motivator to plant a garden, or make lifestyle changes, save the planet, etc. To those folks, I say by all means, go for it. I'm on the same quest myself. But, I'm also all for using the most up-to-date and available science when we make decisions, and for ensuring that everyone else is aware of the current science. Now, you are too.

That's all for now.


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

bottom of page