EMusic Listening Guide (ENG)
RIDING THE EDDY CURRENTS:
A LISTENING GUIDE TO EMusic
In order to catch the meaning of EMusic, I have prepared a short listening guide, associated to the Aquila track I have posted some days ago: https://youtu.be/0DD1l2ZppF4
Just a few words about TEM method: Time Domain EM methods use a direct electric current, flowing into an insulated transmitting loop and generating a static primary magnetic field. The abrupt switch off of the transmitter current causes a rapid decay to zero of the primary magnetic field, so that secondary electric currents (eddy currents) are induced in the Earth. According to the Lenz’s Law, the eddy currents act to oppose the decrease in the primary magnetic field and spread out like horizontal “smoke-rings” in the subsurface. The fast decay of these fields causes the birth of a secondary magnetic field that is measured by means of an induction coil. The voltage decay (i.e. the “transient”) is measured as a function of time; depth of investigation increases with acquisition time, as at later times the current will proceed deeper into the ground. The inversion of voltage data collected at different acquisition times produces a 1D model of the subsurface, in terms of resistivity.
The Aquila track was composed by means of a TEM sounding collected just out of the Aquila walls, by means of a Geonics Protem 57 equipment. The initial sound has the only purpose to mark the impulse provided by the transmitter, hence the pitch is not related to any effective voltage. Then, after the turn-off time, every pitch is located exactly at the start of each time gate of the Protem, after having applied a time expansion of 100,000 times, so that to make distinguishable the different sounds (the latest useful gate is set at about 6 msec). Then, we kept the pitches roughly until the end of the gate window; indeed we would improve this arrangement in a following audio test, taking into account the effective window width and the progressive attenuation of the signal.
We can listen the descending pitches (that are directly correlated to the voltage values), as the signal decays more and more, with increasing spacing between the notes, as the gates are set according an exponential scale. This makes immediate the concept of vertical resolution in TEM sounding, that decreases with time, i.e. with exploration depths. At the same time, it is clear how the EM instrument works. For creative purposes, we applied a feedback effect to the notes, so that to create a sort of echo.
The attached figure shows, on the left, the transient with the musical notes associated with each time gate. On the right, the same notes were distributed at the corresponding depths, as drawn from the 1D modeling.
The attached table shows the times of start, center and end of each gate, the voltage response, the corresponding musical note and the approximate depth of investigation.
|gate center||gate start||gate end||voltage||notes||depth|
The first signal coming from the subsurface (the A# at 35 seconds) is referred to a depth of about 45 m and this explains in a simple way the concept of shallow resolution: if we would resolve the shallower layers, we should collect earlier the signal. Hence, we can link the notes to the geological formations: the first part of the track refers to a thick complex of silts and sands (40 ohm-m), then at about 1 minute and 40 seconds, we reach a clay layer (3-5 ohm-m) resting at 90 m depth. Over 8 minutes we enter the conglomerate bedrock (90 ohm-m), at 230 m depth. It’s interesting to note the different times spent to cross the more resistive overburden (90 m in just 100 seconds), in comparison with the clays (140 m in about 6 minutes), so that to provide an immediate perception of the velocity of diffusion of the eddy currents, due to resistivity variations.
Thus, in this case, we are able to really follow the eddy currents’ travel into the subsurface and to get a real perception of the response collected by the receiver coil with the elapse of time.
This way of handle EM data is just one possible option (that I named “station mode”), that makes use of a single TEM sounding. If we use the whole available time gates, in the case e.g. of the Geonics Protem equipment) and apply a time expansion of 1 million time, we could arrange a very long musical track (in theory of about 19 hours !) that could be used as an audio installation, like an “ambient music” performance.
A second way to work on EM sonification is the “flight mode”, in which we transform Airborne EM data, so that every sounding is played as a block chord (all the time gates are played at the same time), whereas the elapse of time is given by the actual temporal scale (i.e. the chords are set along the track at the effective acquisition time, e.g. every 1 second).