- Andy Davis
How fast does a monarch fly? A close look at the science
With the beginning of a new year upon us, I thought I'd start with a completely random subject for the first blog entry - monarch flight speed. I bet that everyone reading this has heard this question at least once, or maybe has wondered this themselves. It's a question that pops up whenever there is talk of their migration, or of tagging, and I know that there are varying answers out there. In this blog post, I'm going to answer this question by looking at all of the available science on the subject.
For starters, the answer to this question really depends on whether you want to know how fast they fly - like, their actual flight speed, or if you want to know how fast is their migration. These are two different things. The migration consists of alternating flying and stopping to rest. And their flight speed is greatly affected by the wind conditions they are flying in as well. So considering all of this, it is perhaps no wonder that there are so many estimates.
I happen to be well-versed in this issue. In the latest monarch book - Monarchs in a changing world - I and Elizabeth Howard had a chapter in it where we examined some Journey North data, and for that project we summarized all of the prior estimates of monarch flight speed and migration speed. So for this post, I'm going to essentially copy what we had in that paper and then add to it. If you have a copy of the book and want to verify my numbers, our chapter was titled, Tracking the fall migration of eastern monarchs with Journey North roost sightings: new findings about the pace of fall migration." With this title, you can see why this project is relevant to the question at hand.
Let's start with flight speed. The first estimate we have is actually THE first estimate, ever. Years ago (1960), Fred Urquhart published his book "The monarch butterfly", and in it he described how he followed cruising monarchs with his car, and read the car speedometer while doing so. He claimed their flight speed was 18km/hr. If I sound skeptical about this, it's because I am. As you will see, this number is much higher than other estimates we have. And, do you know how hard this is to do anyway? Fred also had made some other dubious claims too, like when he looked at monarch scales under the microscope he saw OE spores, and he thought were tiny scales!
Next we have an estimate from a very obscure study published back in 2001, by a whole family of citizen-scientists, I think (the authors were four people named Moskowitz!), who witnessed a large accumulation of monarchs and dragonflies during one fall migration. The study was published in the Northeastern Naturalist, and was titled "Notes on a large dragonfly and butterfly migration in New Jersey." Apparently, the family was watching the insects as they crossed a parking lot on one exceptionally busy day, and they estimated their flight speed by timing how long it took to cross the lot, and then measuring the length of the lot. They got an average of 7km/hr for the monarchs.
Another estimate was from a paper I published back in 2002 in the American Midland Naturalist, with my colleague Mark Garland, who is now the directer of the Cape May Monarch Monitoring Project. Our paper was a report of some monarch observations we had made while tagging monarchs during the fall of 1999. At the time, Mark was tagging in coastal New Jersey, and I happened to be tagging at a site in coastal Virginia, just south of Mark's site. That year, I captured about half a dozen of Mark's tagged monarchs from NJ, and there was one in particular that was captured the next day. According to what I wrote in the paper, "monarch #182240 was tagged at Cape May on 6 Oct. 1999 at 13:00, and subsequently recovered at 17:00 the next day on Fisherman’s Island (the southern tip of coastal Virginia), approximately 16 daylight hours later." Based on the distance from Mark's site to mine (226km), and assuming the monarch flew only during daylight, we estimated it's speed was 14km/hr. It should be noted though that there was a strong tailwind that day. If anyone is interested, I'll put a link to this paper here.
One more estimate we have also comes from my own work, but this time in a laboratory setting. And this one is pretty cool. Have you ever heard of a butterfly treadmill? Well, in our lab we have one! It's called a "flight mill", because it measures butterfly flight speed. I'll blog about this thing at a later date, but essentially, the way it works is you glue a monarch to a horizontal rod (it doesn't hurt them), which is attached to a center post by a frictionless pivot, and the monarch flies in a circle around the pivot. A computer tracks the speed while it flies in circles. I know, it's crazy, but this is actually a great way to learn about flight behavior in a lab setting, and entomologists use these things a lot. The monarchs actually fly for hours on this thing without stopping. If you want to see this device in action, we uploaded a short video of it on youtube - link. So a few years ago, I worked on a study that tracked flight speed in monarchs in relation to their wing characteristics. At that time we had examined over 100 monarchs on this flight mill and from these data we calculated the overall average flight speed of about 4km/hr.
OK, now for the migration speeds. Keep in mind that these estimates are very rough, because they rely on assumptions of how many hours monarchs spend flying each day. Also, migration speed is a lot different than flight speed because during migration, monarchs fly for a couple days, then stop for a day or two, then fly, etc. So this really should be called the migration pace or rate.
On the MonarchWatch website, there is a rough, back-of-the-envelope calculation for the overall migration rate for the entire fall migration. They figured the time it takes the fall migration wave front to go from Minnesota to the Mexico border (~2250km) is about 30 days, or 75 km/day. If you assume monarchs fly for about 10 hours a day, that translates to 7.5km/hr for the fall migration speed.
Next is another study I worked on years ago with Elizabeth Howard again. It was published in the Journal of the Lepidopterists' Society in 2005. We examined spring sightings of monarchs travelling north, and we used a fancy GIS approach to estimate how fast the spring migration progresses. We came up with 70km per day. If you assume the monarchs fly for about 10 hours each day that translates to 7km/hr.
Finally, for the chapter we had in the new monarch book, Elizabeth and I examined records of fall migration roost sightings from the Journey North program, and we specifically tried to estimate how fast the fall migration progresses, as it proceeds southward. I won't bore you with the math here, but suffice it to say, we found the overall migration pace to be about 32km per day. In other words, with each passing day, new roosts are formed about 32km farther south than in the prior day. However, we did find some clear variation in this estimate. In the northern part of the flyway, the pace is about 14km/day, but as the migration proceeds southward, it picks up to about 42km/day. Again, if you assume monarchs fly for 10 hours per day, the overall average translates to 3.2km/hr.
There was a very fascinating thing we also discovered while doing this project, that is that the fall migration pace appears to be increasing each year. Take a look at the graph below, which shows the overall pace on the y-axis, and year on the x-axis. For the y-axis, numbers nearer to the top of the graph represent a faster southerly pace. While we only had 7 years of data to work with at the time, we were shocked to see that this relationship was statistically significant. We think this has to something to do with climate change. I talked about this in the previous blog - warming temps are pushing the monarch breeding range northward, which increases the distance they have to travel to get to Mexico. Thus it is remotely possible that monarchs are adjusting their migration speed so they can get to Mexico before it gets too cold for them in their breeding range. This is definitely something we should keep track of in the coming years.
So about the migration and flight speeds, let's put all of the numbers together. We have a wide range of estimates that I went over in this post - 18, 7, 14, 4, 7.5, 7, and 3.2. Some of these numbers are based on direct observations of monarchs in flight, some are from looking at the entire migration cohort. I'm not sure we can say that any one of these is better than the others, so I would say let's just average them all. So...the average of all of these numbers is 8.7, which we can just round up to 9. From my perspective (as a scientist) this average should be the most comprehensive and accurate estimate out there, since it is based on all of the science we have on the subject. Keep in mind this is an average, and there is clearly some variation to this estimate. In other words, monarchs are at least capable of flying faster than this, given the right conditions (like that strong tailwind in the tag-recovery story).
The most scientifically-accurate answer then, to the question, "how fast do monarchs fly?' is about 9 km/hr, or 5.5 mph. To put this into perspective, the average person jogs at a speed of 6-8mph.
There you have it. Feel free to share this post with anyone who has ever asked you this question, and keep it ready to pull out in case you ever get asked again.
Direct link to this blog entry: