The goal of this project is to study the magmatic system beneath Okmok volcano lund we will be using two passive geophysical imaging methods to do so. Passive seismic data will image the seismic velocity structure and magnetotelluric (MT) data will image the electrical conductivity structure. Here we give a quick overview of the MT method.
The MT technique uses low frequency (0.0001 – 1000 Hz) electric and magnetic fields to probe electrical conductivity from the very near surface to deep into the mantle. The MT electromagnetic source fields below about 10 Hz are generated by a variety of activities occurring in Earth’s ionosphere, the region of the atmosphere that spans from about 90 to several hundred kilometers altitude. In this region, interactions between Earth’s magnetosphere and charged particles arriving with the solar wind, as well as heating from the sun’s incident light, create various current and plasma systems. Variations in the electrical current flowing in these systems generate electromagnetic waves that travel down through the non-conducting lower atmosphere. These waves then diffuse into Earth, where they rapidly attenuate due to the much higher conductivity of Earth’s rocks. The amount of attenuation depends on the frequency of the electromagnetic wave as well as the rock’s conductivity. Generally speaking, high frequency energy only penetrates to shallow depths, while low frequencies can diffuse all the way into the mantle. As shown by Faraday’s law and Ampere’s law, the attenuation of these fields as they diffuse into the Earth induces complementary secondary electric and magnetic fields. By measuring both the electric and magnetic fields on the surface of the Earth over a wide range of frequencies, MT data can constrain electrical conductivity to great depths.
At frequencies above about 10 Hz, the MT source field arises from lightning strikes occurring around the globe. Large strikes create electromagnetic pulses that radiate out laterally, bouncing back and forth in the waveguide created by the base of the ionosphere and the ground and encircling the globe multiple times, a phenomenon known as Schumann resonances. The onshore MT data we collect should have lots of this signal in the 10-1000 Hz band. However, because seawater is fairly conductive (about 3.3 S/m), this high frequency portion of the MT source field gets attenuated rapidly in the ocean and so the seafloor MT data will only record MT signals to about 1 Hz in 1 km deep water, and probably to only 0.1 Hz for the stations in the base of the Aleutian trench (about 6 km deep!). That’s okay though since even with those frequency cut-offs, the seafloor data will be sensitive to structure at the crustal and mantle depths of the magmatic system beneath Okmok.
Here’s a cartoon depicting the MT method, where an array of seafloor electric and magnetic field recorders has been deployed from a ship. The wavy arrows depict the incident MT source fields. Note that in reality the fields decay much faster with exponential depth attenuation and the spatial wavelengths can span from hundreds of meters to hundreds of kilometers (in inverse proportion to the square root of frequency).
Here’s what a few hours of seafloor MT data actually look like. Last year we had a project off the coast of Oregon and Washington where we deployed our array of seafloor EM recorders. I took the raw recordings of the horizontal electric and magnetic field vectors and made this animation showing their polarization and magnitude over a five hour time window. Red arrows show the seafloor electric field and blue arrows show the magnetic field. Both have been bandpass filtered to the 0.005- 0.02 Hz band and were created from the raw time series without any noise removal (so ignore the occasional outlier stations).
Once we have MT ground impedances for several stations, we then throw all that data into a computer program that solves for a conductivity model that fits those observations; specifically, we use a regularized non-linear inversion method to do this.
That’s enough about the basics of the MT method. Now let’s show some previous examples of imaging magma with MT data. Here’s an example where we used seafloor MT data collected across a mid-ocean ridge, specifically, the East Pacific Ridge at 9ºN, in order to image melting in the mantle beneath the spreading center. The green to red colors show high conductivity associated with melt being formed in the upwelling mantle. The blue colors show low conductivity solid mantle rocks that focus the melts to the spreading axis, where it has also been observed erupting.
For this project, we will be studying melt generated at a subduction zone rather than the mid-ocean ridge shown in the image above. Here’s a previous example where some of our colleagues used onshore MT data collected in Washington state to image melts (orange to red colors in the mantle) coming off the subducting Juan de Fuca plate (dark blue linear region) and rising up beneath Mt Rainier (red triangle). From McGary et al., (2014).
The goal for our project is to create a similar image using the long 2D profile of marine and land data, in order to quantify the amount of melting in the upper mantle and the pathway the melt takes to the crust beneath Okmok. Then the dense 3D array of stations we will collect on and around Okmok will be used to image the shallower magma contained within the crust.
Nobody has ever carried out an amphibious MT survey of an arc volcano before, so we are excited about what this data will teach us about the magmatic system.