Complexity signatures in the geomagnetic &amp;lt;i&amp;gt;H&amp;lt;/i&amp;gt; component recorded by the Tromsø magnetometer (70° N, 19° E) over the last quarter of a century

C M Hall

doi:10.5194/npg-21-1051-2014

Abstract

Abstract. Solar disturbances, depending on the orientation of the interplanetary magnetic field, typically result in perturbations of the geomagnetic field as observed by magnetometers on the ground. Here, the geomagnetic field's horizontal component, as measured by the ground-based observatory-standard magnetometer at Tromsø (70° N, 19° E), is examined for signatures of complexity. Twenty-five year-long 10 s resolution data sets are analysed for fluctuations with timescales of less than 1 day. Quantile–quantile plots are employed first, revealing that the fluctuations are better represented by Cauchy rather than Gaussian distributions. Thereafter, both spectral density and detrended fluctuation analysis methods are used to estimate values of the generalized Hurst exponent, α. The results are then compared with independent findings. Inspection and comparison of the spectral and detrended fluctuation analyses reveal that timescales between 1 h and 1 day are characterized by fractional Brownian motion with a generalized Hurst exponent of ~1.4, whereas including timescales as short as 1 min suggests fractional Brownian motion with a generalized Hurst exponent of ~1.6.

Highlights

Introduction and methodologyUnderstanding the coupling mechanisms between various processes and phenomena in the solar–terrestrial system remains a considerable challenge
The geomagnetic field characteristics represented by a local time series measured, on average, beneath the auroral oval at 70◦ N, 19◦ E will be examined
To summarize the above findings, all analyses, irrespective of scale, indicate fractional Brownian motion (fBm). Both spectral analysis (SA) used in the regime 12–1 h and detrended fluctuation analysis (DFA) indicate α ≈ 1.42

Summary

Introduction and methodology

Understanding the coupling mechanisms between various processes and phenomena in the solar–terrestrial system remains a considerable challenge. The majority of studies to obtain complexity signatures from time series aims at evaluating the Hurst exponent, H , as invented by Hurst (1951) as a quantification of the scaling nature, or self-affinity, of the stochastic component of the data. Variances are calculated for each subseries and are averaged to obtain a mean F (n) After repeating this for a range of subseries lengths (usually all possible n), the function F (n) is plotted vs n in log–log space (as was done in the spectral analysis case) to hopefully identify a regime exhibiting a scaling exponent α:.

Underlying data and analysis

Discussion

Conclusions