The trouble ‘measuring’ geomagnetic events

A sunspot five times the width of Earth exploded earlier this week. On September 10, it released an X-1.6 class solar flare. A solar flare is a sudden burst of energy that releases much more light and other radiation in one second than it usually does. This one was also accompanied by a violent release of millions of tons of charged particles and strong magnetic fields called a coronal mass ejection (CME).

Their combined physical influence on Earth’s magnetic field and upper atmosphere is described by a number called the Kp index. It is determined using an array of instruments called magnetometers stationed around the world that measure how Earth’s magnetic field fluctuates over successive three-hour periods. Interestingly, there have been some questions about how the Kp index is defined and measured.

The impacts of solar flares, for example, are well-defined. Although they can get stronger through a continuous spectrum of energies, they’re classified into six groups to make understanding each energy range’s implications easier: A, B, C, M, X and Z. All ranges except X have nine sub-points (e.g. A-1, A-2, A-3, …, A-9), whereas X is known to be able to go up to X-28. Because its scale is logarithmic, an X-1.6 flare is 1.6 times as strong as an X-1 flare.

Each class corresponds to a range of energies in W/m sq. “of 100 to 800 picometre X-rays near Earth, as measured on the GOES spacecraft” (Wikipedia).

On the other hand, the Kp index is calculated indirectly. Each magnetometer around the world measures one of the two components of Earth’s magnetic field: the horizontal. When geomagnetic disruption occurs, the horizontal component fluctuates over certain regions. The fluctuations are picked up. Then, the Kp index is calculated for each magnetometer as the maximum fluctuation at that location compared to that on a normal day.

Finally, an algorithm calculates the average global Kp index according to Kp indices from all the magnetometers. This isn’t as easy as it sounds because each measurement station’s “normal” is different, leading to a constantly changing weighted average being necessary.

Depending on how and where the disruption occurs, alerts are sounded for possible damage to electrical and navigational systems and orbiting satellites – anywhere there is a concentration of electrical or electronic equipment used to provide critical services. The National Oceanic and Atmospheric Administration (NOAA) has a table to describe the extent of damage.


The accompanying Kp numbers describe the extent of the ongoing geomagnetic disruption, with Kp > 5 referred to as a geomagnetic storm. This is what happened last night. The Storm Weather Prediction Center, affiliated with the NOAA, runs a neat live-update system (of the preliminary Kp index) that refreshes every five minutes. You can see where the Kp value peaks in two instances. Another such peaking is expected to happen on Saturday, September 13.


Now, here are the NOAA’s concerns about using a scale to describe geomagnetic disruptions, from a page on their site titled ‘New Scales Help Public, Technicians Understand Space Weather‘.

The simplicity and the usefulness of the scales for some purposes leads to some unavoidable tradeoffs. Using only one physical measure for each scale leads to a description of events less precise than some users might like. Duration and timing, for instance, are not considered in a scale’s assessment of an event; measures like integrated fluxes do not take into account the sometimes important aspects of spectral shape; and the cadence of the data or index may not be well-matched to the time-scale of the physical event, which complicates the climatology. (The average frequency was obtained from the National Geophysical Data Center’s solar-terrestrial database covering the last several solar cycles). As an example, Kp values are derived and reported every 3 hours, although a single geomagnetic storm may last a day or more.

Subsequently, it raises another important issue: about having a scale that goes from 0 to 9 but having few historical incidents of strength 5 or higher. For example, the Richter scale describes the strength of earthquakes, and enough quakes have occurred across the spectrum of strengths for it to be well calibrated.

On the Kp index scale, however, more occurrences of stronger and weaker events are necessary for us to really know where each one sits on the scale because what we know as a Kp 5 today could easily be a Kp 7 or a Kp 3 in the next decade. And as opposed to earthquakes, the potential global nature of geomagnetic disruptions adds to the confusion.