New publication: is the longest migration possible?
If controlling for body size (see tabel below), the globe skimmer dragonfly is thought to embark on the longest non-stop migration in the World. A migration first formally hypothesized by R. Charles Anderson, as he witnessed thousands of dragonflies appearing regularly and yearly to the Maldives in the Indian Ocean. He posed the questions: “Do dragonflies migrate over the western Indian Ocean?”, and stipulated a scenerio where globe skimmers rode on seasonal winds from India to Africa, in order to utilize the rains as they move between the continents.
But this was just an hypothesis, a question posed to the scientific commuity.
Comparison of the longest known migrations over open water in relation to body mass i grams (g)
Since then, 2009, this migration has remained just that, hypothetical, and has not been tested or further investigated except to confirm that dragonfies that arrive on the Maldives do in fact originate in India (see Hobson et al 2012).
So the question remains, do they migrate over the Indian Ocean. Or, even, can they?
Is it possible for a globe skimmer to power its on flight over hundreds of kilometers of open ocean? Or if not, would the seasonal winds in the region transport it fast enough for its stored resources not to run out?
In our latest research paper we set out to investigate the physiological and meteorological factors of the globe skimmer Indian Ocean migration.
Globe skimmers aggregating, in Talakaveri, India. Photo by: Shyamal
How fat does a dragonfly have to be?
Many migrating animals store energy in their bodies before their journey, it is a way of bringing food with them so that they have resources to tap into if they cannot access any during their travel. Birds that fly over open water or across barren deserts prepare by fattening up, some species are even capable of doubeling their weight.
Insects are not that different. Monarch butterflies contain much more fat when they are migrating for example. Unfortunately, not much is known about how dragonflies fatten up in preparation of long flights, except one study that found that 27% of the weight of a migrating green darner was fat (May and Matthews 2008).
So, based on that estimation, a standard globe skimmer individual weighing about 330 mg, would contain 89.1 mg of fat.
But at what rate would this standard individual burn this fat while migrating? How long can it fly on that stored energy?
What does it cost to flap over an ocean?
Dragonflies can adopt different flying techniques in order to get around.
When hunting, attacking intruders or escaping predators, dragonflies can flap all four of their wings independently and fly very fast. However, intense flapping is very costly energywise. When there is an updraft or a favourable wind, dragonflies can also glide and soar, a technique that requires very little flapping and energy.
Naturally, having to conserve energy, a migrating dragonfly would aim to glide and soar as often and as long as possible.
If flapping, a dragonfly would burn energy at a rate of 0.19W (joules per second), and if just resting 0.0024W (May 1979). A rough estimation of gliding metabolic rate could hence be 0.002904W, 0.0024W increased by 20%.
Given the 89mg of fat available to a standard individual, a flapping individual could stay airborne for only 4.29 hours, wheras a gliding individual could be in the air for 286.9 hours.
A globe skimmer flapping away, by Basile Morin
How fast can a dragonfly fly?
Powering an escape or an attack, a dragonfly can certainly reach incredible speeds. But at a great cost of energy.
When migrating across the Panama canal, globe skimmers were estimated to travel at a speed of 5m/sec (Srygley, 2003) and across the Bohai Lake, at 5.8 m/sec (Feng et al 2006).
If travelling at 5.8m/sec, it would take a globe skimmer approximately 5 days to cover 2500km. This means that it cannot rely solely on its on flapping flight, as this energy-costly way to fly would only last for about 4 hours.
The conclusion is that the globe skimmer must utilize winds, and glide over the ocean.
The answer is blowing in the wind
Using weather data, or more precisely a Weather Research and Forecasting (WRF) model, we aimed to simulate an ocean crossing by a globe skimmer, investigating whether there indeed were winds available.
Specifically, we simulated:
a departure a day, every day in September-December, from three different heights, from three different sites:
Gwarka lighhouse, Gujarat, northern India
Vasco da Gama, Goa, western India
Malé, The Maldives
a departure a day, every day in May-July, from three different heights, from two sites:
Guardafui lighthouse, Somalia
Shakani, Kenya
The results were very clear in three ways:
There were winds available that could successfully transport a dragonfly across the ocean
The winds were only available at certain times
Some departure sites had more successful simulated migrations than others
Simulated autumn migration events from three different sites, starting at three different heights, every day for 10 days through the months of September, October, November and December. The colours denotes day since departure, ie the first day until the 10th day.
A successful migration
Autumn migration
From the simulated migration events in autumn, it was clear that migrations relying on winds were most successful if they happened in November or December, and if they departed from northern India. In fact, dragonflies attempting to migrate over the Indian Ocean in September or October would likely fail, and so would those flying over / departing from The Maldives.
It took, on average, 127 hours to cross the Indian Ocean in autumn. But the shortest ever crossing event, one setting out from northern India (Gwarka lighhouse) starting at a height of 500m, took only 33 hours.
Spring migration
In spring, for the return migration from Africa back towards India, the wind conditions appeared to be much more favourable. An average successful crossing took only 55 hours, and the fastest one was 32 hours. June and Jully were the best months to depart from Somalia, if you wanted to be more likely to be successfully carried by the winds to Indian soil.
Simulated spring migration events from Somalia, starting at three different heights, every day for 6 days through the months of May-July The colours denotes day since departure, ie the first day until the 6th day. Simulations starting in Kenya were very similar.
So, is the longest migration possible?
The short answer is yes. There are available winds that co-occur at the time globe skimmers are seen moving directionally in swarms, for example dropping down on the Maldives. And a globe skimmer carrying about 27% of its on weight in fat can sustain itself for 4-286 hours, which is within the time it would take it to glide on the winds that can successfully deliver it across the ocean.
However, since the winds are only available at certain times and in certain areas, an individual would have to have the ability to select the right winds. Of all simulated migration events, only about 15% were successful in actually reaching land in autumn (40% were successful in spring). And, importantly, a physiological window of 4-286 hours is very wide, and we do not know where within this estimation the average lie. Maybe most individuals can only combine flapping and gliding to an extent that sustain them for 10 hours. Or maybe the average is 200 hours. We do not know.
Many insects are able to select winds for their migration, and the globe skimmer is regularly and yearly seen on migration throughout India and over the Maldives. There is a risk that most individuals fail to cross the Indian Ocean, but high mortality rates are common amongst migrant animals, and is not evidence against the existence of this migration beeing successful enough to occur.
References
Anderson, R. C. (2009). “Do dragonflies migrate across the western Indian Ocean?” J. Trop. Ecol. 25, 347–358. doi: 10.1017/S0266467409006087
Feng et al (2006) “Nocturnal migration of dragonflies over the Bohai Sea in northern China”. Ecol. Entomol. 31, 511–520. doi: 10.1111/j.1365-2311.2006.00813.x
Hedlund, Lv et al (2021) “Unraveling the World’s Longest Non-stop Migration: The Indian Ocean Crossing of the Globe Skimmer Dragonfly”. Frontiers in Ecology and Evolution, https://doi.org/10.3389/fevo.2021.698128
Hobson et al (2012) “Isotopic evidence that dragonflies (Pantala flavescens) migrating through the Maldives come from the northern Indian subcontinent”. PLoS One 2012;7(12):e52594. doi: 10.1371/journal.pone.0052594
May, M. L., and Matthews, J. (2008). “Migration in Odonata: a case study of Anax junius,” in Dragonflies and Damselflies: Model Organisms for Ecological and Evolutionary Research (1 st Editi), ed. A. Córdoba-Aguilar (Oxford: Oxford University Press).
Srygley, R. B. (2003). “Wind drift compensation in migrating dragonflies Pantala (Odonata: Libellulidae)”. J. Insect Behav. 16, 217–232. doi: 10.1023/A:1023915802067