This is an article I've submitted to Argia, the journal of the Dragonfly Society of the Americas.
My wife Betsy and I enjoy dragonflying whenever we can, and when we do we’re always on the lookout for interesting behavior. When a splash-dunk event starts up, for example, we point to the dragonfly as it goes from one splash-dunk to the next, and keep track of the total number of splash-dunks in the event. We also watch it rise from the water after the last splash-dunk, and keep watching until it performs the spin-dry – that dramatic burst of water spraying in all directions as the dragonfly spins head-over-heels at 1,000 rpm (Walker, 2011, 2014b). Sometimes they even do two spin-drys back-to-back.
As the summers roll by, we continue to amass interesting data and observations. The purpose of this article is to present some of the key features of our results, both for this year, and for the period from 2011 to 2015. This updates the results previously published for 2011-2014 (Walker, 2014a).
Splash-Dunk Distribution for 2015
The 2015 season of splash-dunk events was exceptional in a number of ways. First, we saw 124 events during the year, which is a record in itself. The average number of splash-dunks per event was 2.45, which is a bit higher than our cumulative average of 2.31 over the period from 2011-2015. The reason for this higher average value can be seen in Figure 1, where we show the splash-dunk distribution for 2015. Notice the small, “secondary peak” at 3 splash-dunks.
Figure 1 Data for the 124 splash-dunk events observed during the 2015 season.
The 3-splash-dunk peak is a feature we haven’t seen before in our data. It is consistent, however, with the “shoulder” or “plateau” in the data at 3 splash-dunks that has been observed to persist over the years. There clearly seems to be an unexpected preference for 3 splash-dunks in the dragonfly world.
The second notable aspect of the 2015 season was the “swarm” of events at the Beaver Pond, which we refer to as the “Splash-Dunk Derby” (Walker, 2015). During this episode, in which 33 events were seen in a short period of time, the average number of splash-dunks per event was 3.18. This accounts for the higher than usual average value for the year, and the peak at 3 splash-dunks.
Splash-Dunk Distribution for 2011-2015
Over the years since 2011, we’ve observed and recorded data for 576 splash-dunk events. A plot of the resulting distribution for the number of splash-dunks per event is shown in Figure 2. As in past years, there is a clear shoulder at 3 splash-dunks. In addition, the maximum number of splash-dunks for a given event is 8, as it has been for the last several years. The average number of splash-dunks per event is 2.31, as mentioned above.
Figure 2 Cumulative splash-dunk data for 2011-2015. The data encompasses 576 distinct events.
With such a large number of observations, the statistical value of these results is significant. In particular, we see a drop-off by almost a factor of 2 from n splash-dunks to n + 1 splash-dunks – with the notable exception of n = 2 to n = 3, of course. In addition, more events have only a single splash-dunk than any other number, and roughly 82% of all events have 1, 2, or 3 splash-dunks.
Temporal Distribution of Splash-Dunk Events
In addition to the number of splash-dunks in an event, we also record the date. This allows us to plot the temporal distribution of splash-dunk activity. The result, shown in Figure 3, shows a striking peak at the month of September. In fact, more than half of all the 576 splash-dunk events observed from 2011 to 2015 were seen in that month.
Figure 3 Temporal distribution of splash-dunk events observed from 2011 to 2015.
In past years, we’ve seen a number of dragonflies get stuck in the water as they perform a splash-dunk. This generally occurs late in the season, when the air temperatures dip near the lowest value possible for sustained flight (Walker, 2012). They always get stuck on the first splash-dunk they attempt, and once they fail to exit the water immediately they are condemned to remain stuck. At times, we’ve seen as many as 8 or 10 dragonflies stuck and struggling at one time. No dragonflies were observed stuck in the water in 2014, and none were seen again in 2015. The reason for this isn’t clear, but it’s nice not to have to watch the poor fellows struggling mightily to no avail.
Splash-Dunk Species
Most of the splash-dunks we observe are performed by darners. In particular, the Paddle-tailed Darner (Aeshna palmata) seems to be the champion of the splash-dunkers. A pair of male Paddle-tailed Darners are shown resting on my fingers in Figure 4 – I lifted them up from their perches one at a time. The photo was taken at Cranberry Lake in Anacortes, WA, a prime location for splash-dunks and spin-drys.
Figure 4 A pair of Paddle-tailed Darners taking a rest between bouts of splash-dunking and spin-drying.
Paddle-tailed Darners are strong fliers, and they plow headfirst into the water at considerable speed when they splash-dunk, creating a sizable splash in the process. I refer to the splash as a “dragon splash,” because it resembles a typical winged dragon, with a large central splash produced by the body of the dragonfly, and smaller side splashes produced by the wings.
Other darners have also been observed to splash-dunk. They are as follows:
California Darner (Rhionaeschna californica)
Blue-eyed Darner (Rhionaeschna multicolor)
Common Green Darner (Anax junius)
Shadow Darner (Aeshna umbrosa)
We occasionally see dragonfly species other than darners engaging in splash-dunking, though usually in isolated incidents. The following species have been observed to splash-dunk:
Four-spotted Skimmer (Libellula quadrimaculata)
Eight-spotted Skimmer (Libellula forensis)
Western Pondhawk (Erythemis collocata)
Cardinal Meadowhawk (Sympetrum illotum)
Autumn Meadowhawk (Sympetrum vicinum)
Blue Dasher (Pachydiplax longipennis)
The new addition to the list this year is the Cardinal Meadowhawk.
Distribution of Two-Spin-Dry Events
When a dragonfly rises from the water after a series of splash-dunks, we always watch with interest to see the spectacular spin-dry. Sometimes we can even see the body of the dragonfly curled up in a loop as it spins like a bicycle wheel. When the lighting is just right – especially when the dragonfly is backlit – we can often see droplets of water shooting off in all directions. Even in cases where we can’t see the spinning and water droplets directly, we can see the slight “zigzag” in the flight path as the dragonfly drops downward during the spin.
What is particularly interesting is that we sometimes see two spin-drys, one right after the other. Apparently the dragonfly just doesn’t feel dry enough after one spin-dry, so it promptly does another. This occurs in only about 5% of splash-dunk events that we’ve observed since 2013, when we began collecting this data. Our results for 2013 to 2015 are presented in Figure 5.
Figure 5 Temporal distribution of two-spin-dry events from 2013 to 2015.
So far, only 17 two-spin-dry events have been recorded, so the data in Figure 5 should be taken with a grain of salt. As the years go on, we will be able to improve the statistical significance of these results. At the moment, however, it appears that two-spin-dry events are more likely later in the season. It will be interesting to see if this trend persists with additional observations.
Acknowledgements
I would like to thank Betsy Walker for help collecting the data presented here, and for co-leading our splash-dunk/spin-dry dragonfly field trips.
Literature Cited
Walker, J. S. 2011. Spin-Dry Dragonflies. Argia 23(3): 29-31.
Walker, J. S. 2012. Splash-Dunking Gone Bad: The Sticking Frequency. Argia 24(1): 19-22.
Walker, J. S. 2014a. Life at 1,000 RPM. Argia 26(2): 11-13.
Walker, J. S. 2014b. Splash-dunk Analysis For 2011-2014. Argia 26(4): 32-33.
Walker, J. S. 2015. Splash-Dunk Derby at Beaver Pond. Argia 27(4): 22-24.
