We have all seen it: when walking along some watercourse we occasionally note the sudden appearance of eddies in an otherwise smoothly flowing stream. They move along with the flow for a while, then break down into ever smaller swirls and finally disappear.
Such eddies and seemingly “chaotic” flow patterns are to be found wherever there are currents. They appear in atmospheric airstreams, in fuels and gases moving through engines, in air flowing over aircraft wings, and even in the human bloodstream. They arise whenever disruptions in a uniform flow can no longer be suppressed by the “thickness” of the moving medium, its molecular viscosity.
All these processes have an important and far-reaching physical consequence: the original flow loses energy through the formation of turbulent eddies. This energy is stored in the eddies themselves. A cascade of ever smaller eddies further disperses this energy. When they are finally very small (in the millimetre range in the lower atmosphere, in the metre range at elevations of about 90 kilometres), friction converts the energy into heat. In this way, energy is irretrievably removed from the flow and is no longer available to it. In the world of technology, turbulence is the horror of all engineers dealing with hydrodynamics.
In atmospheric physics this process is one of the still incompletely understood components essential for describing the energy balance—the distribution of energy in the atmosphere. We only need to imagine being able to precisely forecast also for lengthy periods the flow processes occurring in the atmosphere.
It is difficult enough for scientists and engineers to create fairly well defined and reproducible environmental conditions in laboratory test rigs, but the task is even more complex in the open atmosphere. For obvious reasons, measurements cannot be reproduced there on principle. Another problem is that turbulent processes take place across large spatial and temporal ranges, whereas the small range where kinetic energy is actually “destroyed”, as mentioned above, cannot in any way be recorded by today’s satellites or by those planned for the near future.
This is where our measurement systems come into play: with the ground-based infrared camera system FAIM (Fast Airglow IMager) it is now possible to record flow processes at altitudes of some 90 kilometres with high spatial and temporal resolution (17 metres horizontally and about two-second time intervals). Use is made of the intensive airglow generated at this altitude by the interaction of particular photochemical processes (that is, involving chemistry and radiation). Atomic oxygen and atomic hydrogen give rise to hydroxyl (OH), which is excited to a higher energy level as it forms (one speaks of rotational and vibrational excitation). As with any physical system, the so-activated OH molecule is induced to return to a lower and thus more stable energy state. When it does so, energy is emitted in the form of electromagnetic radiation, or light. The result is airglow, which can be measured by FAIM with high spatial and temporal precision in the infrared range at around 1.0-1.6 micrometres.
The right part of the figure is a typical image of the intensity of atmospheric airglow over Oberpfaffenhofen in an area of about 30 x 30 square kilometres: the lighter the grey tone, the more intensive the airglow (the bright dots are stars). Now the intensity of atmospheric airglow depends to a large extent on the temperature and the number of hydroxyl particles at the aforementioned elevation. Both are readily modified whenever, for example, a wave-shaped flow passes through the air layer. The hydroxyl particles are then spatially compressed or thinned out and deflected in altitude. The result is a change in the brightness of the airglow corresponding to the periodicity and spatial structure of the passing wave.
If such recordings are examined by so-called two dimensional spectral analysis, information is obtained on how much energy is present at which horizontal range in the observed flows. These “power spectra” can be read like a fingerprint. A power spectrum combining several individual recordings is shown in the left part of the figure.
The y-axis shows the energy content and the x-axis the horizontal range (or, more precisely, the horizontal wavelengths) of the atmospheric structures traveling through the airglow layer. One can roughly recognize three regimes based on their slope, which is indicated by the blue line. Depending on the gradient, the flow in the atmosphere is characterized by steady waves (buoyancy regime, Section I), cascading turbulent eddies (inertial regime, Section II) or viscous friction (viscous regime, Section III). Already some70 years ago the Russian mathematician Andrey Kolmogorov predicted the characteristic slope of these regions as -3, -5/3 and -7 in the power spectrum.
So what is new? Because of the mentioned high 17 metre horizontal resolution of the FAIM system (at 90 kilometres altitude) and high temporal resolution of about two seconds, it was possible for the first time to record the flow transition into the region of viscous attenuation, in other words, the region in which kinetic energy is converted to heat. The results of numeric simulations could now be confirmed at scales that were so far not accessible to measurement.
Figure: at left, power spectrum of a FAIM 2 recording (18 Jan. 2016, 16:20:33 UTC); at right, image of atmospheric airglow over Oberpfaffenhofen at about 90 kilometres altitude.
It could therefore be shown that even turbulent processes continuing into the viscous regime can be detected with remote sensing using the FAIM system. This is another argument in favour of installing FAIM also on a satellite for future global measurement of turbulence.
This work was carried out at the Earth Observation Center as part of a doctoral dissertation in cooperation with the Chair of Remote Sensing of the Atmosphere at the Department of Physics of Augsburg University..