In physical exploration including magnetic prospecting, it is a basic requirement to explore in a regular network of standards. However, due to the various obstacles encountered in the survey, measurement is generally not performed in a regular network. Therefore, it is very important to study inversion algorithms with high noise reduction ability and increase the accuracy of sampling of the measured data at irregular intervals.

Ri Hyon Sok, a researcher at the Faculty of Earth Science and Technology, proposed a method of using the Fourier transform-based robust estimation method for processing magnetic prospecting data measured at disproportionate intervals, based on its high sampling accuracy and high noise enhancement capability, and demonstrated the advantage of the proposed method though model experiments and application.

The results of model calculations and field applications show that the magnetic prospecting analysis by the Fourier transform-based robust estimation method is superior to the existing magnetic susceptibility tracking imaging.

For more information, please refer to his paper “Study on the Application of Fourier Transform-Based Robust Optimization for Magnetic Inversion” in “Proceedings of KUTIC-2025”.

Mt. Paektu volcano is the largest intraplate stratovolcano located at the border of the Democratic People’s Republic of Korea (DPRK) and China. It is of great interest to volcanologists worldwide due to 3 factors: 1) its geographical location (nearly 1 400km far away from the subduction region of the Pacific plate); 2) one of the largest eruptions during the past 2 000 years and 3) a 2002-2005 unrest episode showing its potential for future activity.

One key focus of study on volcanoes is the depth and geometry of a magmatic system. Especially, resistivity is sensitive to the presence of fluids and magma, and can provide fundamental information on a variety of volcanic activities.

Originally, some researchers suggested that MT data revealed a conductivity anomaly at approximately 20km depth beneath the volcano. Later, some other ones conducted a 3-D inversion for a MT dataset along about 120km profile, proposed a resistivity model beneath the volcano and asserted that three relatively high conductive zones exist at the depths of 3-5km, 10-16km and 40-60km.

However, much of the geophysical work in the international literature is largely based on the data from China.

To delineate the resistivity structure beneath the DPR Korea side of the volcano, Kim Kang Sop, an institute head at the Faculty of Earth Science and Technology, collected magnetotelluric (MT) measurements in the frequency range of 0.0017-320Hz at about 60 sites to produce a 3D resistivity model, showing the presence of 3 conductors within the top 20km of crust.

He employed a 5-channel MT receiver, three inductive magnetic sensors and two electrical dipoles with Pb-PbCl₂ electrodes manufactured at Kim Chaek University of Technology (KUT).

He conducted a 3D inversion using the ModEM3D code.

His research results are consistent with previous MT and seismic studies.

You can find the details in his paper “Magnetotelluric Constraints on the Magmatic System Beneath Mt. Paektu Volcano” in “Proceedings of KUTIC-2025”.

Euler deconvolution (ED) has been one of the popularized interpretation techniques in magnetic prospecting, which is based on the well-known Euler’s equation and the concept of structural index (SI). It makes it flexible to implement calculation of parameters for typical types of geological bodies such as fault, magnetic boundary and dyke, and it has been widely applied to practical problems.

In the case of high-precision ground magnetic surveys, for several reasons such as topography, river flow, cliff, etc., raw field data are hardly given on a rectangular grid but collected along crooked profiles in general. Such data should be interpolated into a rectangular grid with availability of the ED technique; this may lead to a distortion of original data and erroneous analysis results.

Kim Il Hyok, a section head at the Faculty of Earth Science and Technology, proposed a new 2D & 3D Euler deconvolution scheme of magnetic data irregularly spaced on an uneven topographic surface.

The scheme first inverts the given irregular ground magnetic data to obtain subsurface equivalent sources, from which horizontal and vertical derivatives are calculated directly on the given data points. Then, it, based on the ED equation, simultaneously estimates the structural index (SI) and the window radius. Finally, the depth and position of a magnetic body are calculated based on the estimated structural index and the window radius.

He verified the validity of the proposed method via synthetic and field datasets. The method gives correct structural index and location even under moderate Gaussian noise, which demonstrated its accuracy and noise resistance.

You can find the details in his paper “Euler Deconvolution of Irregular Ground Magnetic Data” in “Proceedings of KUTIC-2025”.