Analysis of geomagnetic hourly ranges

Abstract

In an attempt to develop better forecasts of geomagnetic activity, hourly ranges of geomagnetic data are analyzed with a focus on how the data are distributed. A lognormal distribution is found to be able to characterize the magnetic data for all observatories up to moderate disturbances with each distribution controlled by the mean of the logarithm of the hourly range. In the subauroral zone, the distribution deviates from the lognormal, which is interpreted as motion of the auroral electrojet toward the equator. For most observatories, a substantial deviation from the lognormal distribution was noted at the higher values and is best modeled with a power law extrapolation, which gives estimates of the extreme values that may occur at observatories which contribute to the disturbance storm time (Dst) index and in Canada.

The latitudinal variation of geoelectromagnetic disturbances during large (Dst≤−100 nT) geomagnetic storms

“…There are reasonable arguments to be made that GMDs should follow a lognormal distribution, based on the multiscale physical processes [ Limpert , ], and authors have convincingly shown that certain geomagnetic and GMD‐associated quantities can be well modeled using lognormal distributions. For example, Danskin and Lotz [] have shown that hourly ranges of Δ B are lognormally distributed; Love et al [] demonstrated that D s t is itself well modeled by a lognormal distribution; and Pulkkinen et al [] estimated 100 year geoelectric field magnitudes using extrapolation from a lognormal distribution.…”

Influences of various magnetospheric and ionospheric current systems on geomagnetically induced currents around the world

“…This supports previous research that conductor networks in auroral regions are at greatest risk of generating large GIC than networks in the rest of the world. For example, Danskin and Lotz [] show that auroral regions are more prone to extreme events and Thomson et al . [] also refer to the latitudinal dependence of extreme GIC.…”

“…This supports previous research that conductor networks in auroral regions are at greatest risk of generating large GIC than networks in the rest of the world. For example, Danskin and Lotz [2015] show that auroral regions are more prone to extreme events and Thomson et al [2011] The GIC would likely also change in relative proportion to the time rate of change of the B fields and the currents (via the E fields). When a sudden commencement occurs, marking the start of a geomagnetic storm, the sudden change in the horizontal B field would create spikes in the perpendicular horizontal E field that will send corresponding impulses of GIC through a conductor path.…”

Regression‐based forecast model of induced geoelectric field

“…The extremity of these events is placed in context by considering the probability of finding an event of equal or greater intensity. Danskin and Lotz [] calculated the complimentary cumulative distribution function of HRH at HER over a 15 year interval. According to the distribution of hourly ranges at HER we estimate that the probability of finding HRH ≥67 nT is about 1.5 × 10 −3 (or 14 times per year) and about 2 × 10 −4 (about twice per year) for finding HRH ≥125 nT.…”