Related Experiment Videos
Forecasting secular variation using physics-informed neural networks for IGRF-14
N Shakespeare-Rees1, P W Livermore1, C J Davies1
1School of Earth and Environment, University of Leeds, Woodhouse, Leeds, LS2 9JT UK.
Earth, Planets, and Space : EPS
|May 11, 2026
Summary
The University of Leeds developed a new geomagnetic field model for the International Geomagnetic Reference Field (IGRF) forecast. This model, using Physics-Informed Neural Networks, shows improved accuracy for predicting secular variation (SV) from 2025-2030.
Area of Science:
- Geophysics
- Earth Sciences
- Computational Physics
Background:
- The International Association of Geomagnetism and Aeronomy (IAGA) called for candidate models for the 14th generation of the International Geomagnetic Reference Field (IGRF).
- Accurate forecasting of geomagnetic secular variation (SV) is crucial for various applications.
Purpose of the Study:
- To present the University of Leeds' candidate model for the IGRF-14 forecast period (2025-2030).
- To predict the average geomagnetic secular variation (SV) over a 5-year period.
Main Methods:
- Inversion of the CHAOS-7.18 geomagnetic field model using Physics-Informed Neural Networks (PINNs).
- Generation of two global mesh-free models: one from regional flows and one from a single global flow.
- Advection of the magnetic field assuming steady core flow to construct the 5-year average SV forecast.
Main Results:
- The model derived from regional flows demonstrated reduced Root Mean Square (RMS) misfit compared to CHAOS-7.18.
- Hindcasts for the IGRF-13 period (2020-2025) showed improved performance against other candidate models.
- The regional flow model provided a competitive candidate for the IGRF-14 forecast.
Conclusions:
- The developed Physics-Informed Neural Network approach offers a robust method for geomagnetic field forecasting.
- The regional flow model shows promise for accurate secular variation prediction.
- Further refinements to the methodology could enhance future IGRF submissions.
Related Concept Videos
01:20Maxwell-Boltzmann Distribution: Problem Solving
Individual molecules in a gas move in random directions, but a gas containing numerous molecules has a predictable distribution of molecular speeds, which is known as the Maxwell-Boltzmann distribution, f(v).
This distribution function f(v) is defined by saying that the expected number N (v1,v2) of particles with speeds between v1 and v2 is given by
This distribution function f(v) is defined by saying that the expected number N (v1,v2) of particles with speeds between v1 and v2 is given by
01:11Atomic Nuclei: Larmor Precession Frequency
The earth's gravitational field produces a 'twisting force' perpendicular to the angular momentum of a spinning mass (such as a spinning top) that causes the mass to 'wobble' around the gravitational field axis in a phenomenon called precession. Similarly, the magnetic moment (μ) of a spinning nucleus precesses due to an external magnetic field directed along the z-axis. The precession of the magnetic moment vector about the magnetic field is called Larmor precession, and the angular frequency...
01:05Estimation of the Physical Quantities
On many occasions, physicists, other scientists, and engineers need to make estimates of a particular quantity. These are sometimes referred to as guesstimates, order-of-magnitude approximations, back-of-the-envelope calculations, or Fermi calculations. The physicist Enrico Fermi was famous for his ability to estimate various kinds of data with surprising precision. Estimating does not mean guessing a number or a formula at random. Instead, estimation means using prior experience and sound...
01:20Variation in Acceleration due to Gravity near the Earth's Surface
An object's apparent weight is its weight measured by a spring balance at its location. It is different from its true weight, the force with which the Earth pulls it, because of the Earth's rotation. Mathematically, an object's apparent weight equals its true weight minus the centripetal force that keeps it in a circular motion along with the Earth's surface every 24 hours.
The difference between the true and apparent weights is proportional to the square of the Earth's angular speed. Since the...
The difference between the true and apparent weights is proportional to the square of the Earth's angular speed. Since the...
01:08Determination of Expected Frequency
Suppose one wants to test independence between the two variables of a contingency table. The values in the table constitute the observed frequencies of the dataset. But how does one determine the expected frequency of the dataset? One of the important assumptions is that the two variables are independent, which means the variables do not influence each other. For independent variables, the statistical probability of any event involving both variables is calculated by multiplying the individual...
01:26Influence of Earth's Curvature and Atmospheric Refraction on Leveling
During leveling, the Earth's curvature and atmospheric refraction introduce deviations in the line of sight from a true horizontal reference. When the line of sight is leveled, it remains perpendicular to the plumb line only at a single point. Beyond this, it deviates due to the Earth’s curvature, represented by the correction C. For a sight distance D, the deviation can be derived using the relationship:This relationship shows that the deviation increases quadratically with distance. Over a...