It is a receiver for a time-domain electromagnetic surveying tool. In short, used to take a vertical sounding of the electrical profile of the earth beneath the instrument.

Longer form: different geological materials have different electrical properties. If you lay out a large loop of wire on the ground and put a time-varying electrical signal into that loop, you'll generate a magnetic field with that loop (Faraday's Law).

(Some jargon here which assumes you took calculus at some point in your life.) The shape of the magnetic waveform will be the first derivative of the shape of the electrical waveform you put into that loop. So, for example, if you transmit a sinusoid, you'll get a magnetic sinusoid into the ground phase-shifted by 90-degrees. If you transmit a square wave into the loop, the electric field is only changing at very brief moments -- at the rise and fall of the square wave -- so the derivative is a magnetic pulse at that exact moment. The crispness of that magnetic pulse is related to how fast you can turn on or shut off your electric field, and you have to worry about "bounce" in your switches and all sort of things to try to get this pulse to be as clean as possible. This system uses a square electric signal to create a very sharp magnetic pulse.

The magnetic pulse sets up very strong field lines, but only for a brief moment. When those field lines interact with the geological materials in the earth, you will induce electrons to flow in a direction perpendicular to those field lines. How easy or hard it is to induce this flow will depend entirely on the material properties. This flow of electrons will themselves create a smaller secondary magnetic field which will oppose the primary field created by the giant coil on the surface, and there is a slight time lag between the application of the primary field and the creation of the secondary fields (there is a phase delay each time a field gets converted from electrical to magnetic and back).

On the surface you have a secondary loop, attached to a receiver like the one pictured above. This loop will compare the magnetic field recorded on the loop to the square waveform transmitted on the original loop (using a reference cable, or reference radio signal, or a very sensitive clock on both devices). What you're looking for is the very small magnetic fields that seem to linger in the milliseconds after you shut off the electric field on the primary loop. Using a lot of math, you can work out a model or the electrical conductivity structure of the ground.

An image I shamelessly stole from this paper, to show a typical output: https://www.sciencedirect.com/science/article/abs/pii/S0926985111001108