I got a couple questions about measuring wind turbulence, one from my labmate Dennis:
What is turbulence and how do you measure it?
and the other from, uh, Mace Windu:
I heard you were going to measure wind somehow. How is that working for you?
I have a sneaking suspicion that both questions may have originated from the same source, as Dennis is probably about the only person I know who would actually try to impersonate a character from justly reviled Prequels. Either that, or Ron Perlman pointed his friend Samuel L my way.
Anyway, my labmate (and fellow blogger of invertebrate field biology) Lindsayalready wrote a post about how she’s measuring turbulence in Moorea (an island just off of Tahiti), and what I’m doing is largely similar over here, equipment wise.
Basically, the idea here is that we’re curious about how windy conditions might affect the ants while they’re gliding. That is, if an ant is falling and trying to glide back to its home tree and a gust of wind happens to blow by at just that moment, does she get blown away with the wind or can she compensate? We’re also interested in how the ants deal with different levels of turbulence, which is roughly a measure of how variable the speed and direction of the wind is at a given point in space.
Now, ideally, what I’d love to be able to do is to measure what the air is doing over the entire volume of space through which the ant is gliding. Unfortunately, that’s kind of hard to do, and so what I’m settling for at the moment is measuring what the wind is doing at one point halfway up the tree while I drop them and film their glides. Now, obviously the further the ant is from the point where I’m measuring the wind speed and direction, the less confidence I have that what I’m measuring is what the ant is experiencing, but we’re going to go ahead and sweep that inconvenient little fact under the rug for the time being.
Those little black arms in the middle are ultrasonic transceivers, which means that they can both emit ultrasound pulses and receive them. To measure wind speed, transceiver A shoots a pulse at transceiver B, and then transceiver B shoots a pulse at A. If the air between them is moving, there will be a difference in the time of flight of the two pulses, and that time difference can be used to calculate how fast the air is moving. Then, by using three pairs of transducers at right angles to one another, you can measure the wind speed and direction in 3D.
Today I set up this anemometer at a height of 15m above the forest floor, while I was dropping ants from a height of 20m. The ants will typically get back to the tree within 10m if they’re dropped about a meter and a half from the tree, so this is a reasonable halfway point. Today was a very gusty day (the data below are the precursors to a huge storm that rolled through today), and the data from about 40 minutes of sampling looks something like this:
The direction of the wind also changed with time, as shown in this plot which shows the azimuth (horizontal angle, like compass direction) and zenith angles (angle away from vertical, where zero implies that the wind is blowing horizontally) of the wind direction and how those angles change with time. The azimuth is shown in blue, and the zenith is shown in green:
I dropped a bunch of ants, some during the lulls between gusts where the wind speed was around 0.5 m/s, and some during the gusts (which sometimes got to be over 3 m/s). Somewhat unsurprisingly, it looks like when you drop an ant into a gust, it’s harder for her to get back to the tree trunk, and if the gust is strong enough she won’t make it back at all. Once I analyze the trajectories, I’ll be able to say how those trajectories are influenced by wind speed.
However, I do want to point out that I picked this tree specifically because it’s more exposed than most trees in the forest, and I was interested in looking at this question. Most of the time, it’s not very windy underneath the rainforest canopy – wind speeds are usually around or under 0.2 m/s, which is barely detectable.