top of page
  • Andy Davis

Is the monarch spring migration becoming later? A close look at Journey North data

Greetings readers,

Did anyone notice how late the spring migration seemed this year? From watching the online monarch sightings trickle in this spring, it seemed that they took forever to reach the top of their breeding range this year (but they did reach it). In this week's blog entry, I'm going to take you through some graphs that I made recently using the spring sightings database from Journey North, which turned out to be very surprising.

​​As most of the seasoned readers here know, the best way we have to track the monarch spring migration is through the Journey North ( citizen science program. This is a program that asks people to report sightings of monarchs, which then get posted to an online map. This program, which is run by my friend, Elizabeth Howard, has been operating for about 2 decades now. While primarily an educational tool, the data from Journey North has been used in a lot of scientific endeavors - I myself have collaborated with Elizabeth on at least 10 scientific projects (that I can remember), where we analyzed the data from this program to answer a wide variety of scientific questions about the monarch migration. Plus, with each year that goes by, the scientific value of this program continues to grow, because now we can use the data to ask long-term questions, as I'll do in this blog entry. In short - it's a scientific gold mine.

In the early days of the program - like way back when the internet was just getting started (mid-90s), the goal of the program was simply to have people report the date and location where they see their first adult monarch each spring. Since then, Elizabeth has been adding more and more new sighting categories that people can report, so that nowadays, you can report almost anything, like when you first see an egg, a larva, first milkweed shoots, and then in the fall, you can report sightings of migratory roosts, peak migration day, and the list goes on. This is all good stuff.

Now back to the initial issue - about the lateness of this year. Like my readers, I was watching Journey North this spring very intently, because of the well-publicized winter storm that affected the overwintering population earlier this year. This storm event likely reduced the number of potential re-migrants considerably, although I have always contended that this is not all bad - see my previous post about this. Plus, given the vast reproductive potential of the monarch, they can usually bounce back from events like this, and while it is early yet, it looks like they have done so this year too.

So from what I could tell, the monarchs seemed to be pretty late this year, that is, there were a lot of people who were watching for them from the get-go this spring, but they didn't see their 'first' one until late in the migration. So I went to the Journey North records to see if this was indeed the case - that is, were the monarchs later this year than normal? I have a copy of all Journey North data from the eastern monarch population (I removed all western sightings) going back to 1997. From these data, I honed in on the dates and latitudes of each sighting. I then compiled two charts, below, which list, for each year, the 'average' sighting date. I know, you're thinking 'but dates aren't numbers that you can average,' Actually they are - you can convert any date of the year into a single number that represents the number of days since January 1. In other words, each calendar day is one number between 1 and 365. So once I did this conversion for all of the monarch sightings I could then calculate the average number for each year. This average number can then be converted back to a date.

So there are two graphs below - the first shows the average dates for all first sightings that are north of 40 degrees latitude. To give you some context for where the latitudes are on the continent I pasted a blank map from Journey North below as well. You can see from the map below that this 40 degree latitude I chose is more or less a convenient cutoff line representing the 'northern' half of the range.

Now, here's the graph I just mentioned. Look at the last 4 years especially. Crazy, right? I was surprised to see this as well.

So this graph did answer the original question - monarchs did appear to be later than normal this year, but what was more surprising was that they have been late for the past 4 years. In other words, based on this graph I believe it is safe to say that monarchs have arrived to the northern half of their breeding range later than normal for the past 4 years. In fact, they are about 20 days later than normal. Incidentally, keep in mind that the dates on the y axis probably won't match up with your own records from your backyard. Remember these are the averages of all sightings above 40 degrees latitude, which is many thousands of records and locations.

So what's going on? What changed between 2012 and 2013? It's possible that these late sightings are a manifestation of the low overwintering numbers in the past few years, since fewer monarchs would mean it takes people longer to see their first monarch. I'm not totally sold on this because if that were the whole story, you'd expect the 2015 average to reflect the larger wintering colony that year.

It's possible that monarchs are just migrating later in the spring. In other words, they may be late in leaving the wintering colonies to begin with. So to answer that question, I looked at the Journey North data again but this time only looked at sightings below 31 degrees latitude. From looking at the map above, you can see that this will select only the cohort of returning migrants. The dates of these sightings should therefore reflect the timing of arrival to the United States, which should roughly correspond to the timing of the Mexico departure. So when I graphed the average sighting dates for these monarchs only, I got the chart below.

From looking at this graph it looks like there is a slight trend for later arrival to the US in recent years, but not much. The average dates for the past 4 years are about 12 days later than the average for all other years. However, it looks like this year (2016), the average arrival date is not that different from the norm. So I'm not sure we could conclude from this that monarchs are departing the overwintering sites later in recent years.

So from all of this, it seems that we can say (pretty confidently) that monarchs are taking longer to reach their northern breeding range than they used to. Why this is is not clear. Now, before people start freaking out about this, I'll point out that this is not necessarily a bad thing. We humans tend to associate lateness with something bad, but in the natural world, this is not always the case. Some migratory animals can shift their arrival schedules around either forward or backward, depending on the conditions around them. The other thing I'll point out is that regardless of their timing, keep in mind that monarchs still successfully recolonized their entire breeding range this year, as they usually do. Take a look at the latest map from Journey North. This map, which I downloaded today, shows the locations of monarchs in the month of August for this year. I think we can all agree that they have successfully recolonized the full extent of their breeding range this year.

So I'm going to leave it here for now. The take-home messages for today are this:

- Based on objective interpretation of Journey North records, there a recent trend for later spring migrations

- The reason for this is not clear, but it is not necessarily bad, since they still successfully recolonized their full breeding range.

In this case you could say - better late than never!


Direct link to this blog article:



The science of monarch butterflies

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

bottom of page