Toward more reliable long‐term indices of geomagnetic activity: Correcting a new inhomogeneity problem in early geomagnetic data

Abstract

For the time before the space era, our knowledge of the centennial evolution of solar wind (SW) and interplanetary magnetic field (IMF) is based on proxies derived from geomagnetic indices. The reliability of these proxies is dependent on the homogeneity of magnetic field data. In this paper, we study the interhourly (IHV) and interdiurnal (IDV1d) variability indices calculated from the data of two British observatories, Eskdalemuir and Lerwick, and compare them to the corresponding indices of the German Niemegk observatory. We find an excess of about 14 ± 4% (5.8 ± 2%) and 27 ± 10% (15 ± 6%) in the IHV (IDV1d) in the indices of Eskdalemuir and Lerwick in 1935–1969. The timing of this excess accurately coincides with instrument changes made in these observatories, strongly supporting the interpretation that the excess is indeed caused by instrument related inhomogeneities in the data of Eskdalemuir and Lerwick. We show that the detected excess notably modifies the long‐term trend of geomagnetic activity and the centennial evolution of IMF strength and solar wind speed estimated using these indices. We note that the detected inhomogeneity problem may not be limited to the data of the two studied observatories but may be quite common to long series of geomagnetic measurements. These results question the reliability of the present measures of the centennial change in solar wind speed and IMF.

Reconstruction of the Sunspot Group Number: The Backbone Method

“…The IDV and IDV(1d) indices were discussed in the last section. Note that recently Holappa and Mursula (2015) have suggested that errors in the geomagnetic data make the To relate OSF to sunspot numbers, in this section we use the model of Solanki et al (2000), based on the continuity equation for OSF:…”

Space climate and space weather over the past 400 years: 1. The power input to the magnetosphere

“…However the insight of Svalgaard and Cliver (2005) is very valuable and Rouillard et al (2009) exploited it to reconstruct both B and V SW , and Lockwood (2014a) used 4 different pairings of indices to derive both with an uncertainty analysis back to 1845. There is always the possibility of calibration drift in the historic geomagnetic data (Svalgaard, 2014;Holappa and Mursula, 2015), some of these have been identified, agreed upon and corrected for (e.g. Lockwood et al, 2014d ) whereas others are still debated, and others noted but their effects not properly accounted for.…”

Tests of Sunspot Number Sequences: 2. Using Geomagnetic and Auroral Data

“…The IDV and IDV(1d) indices were discussed in the last section. We note that recently Holappa and Mursula (2015) have suggested that errors in the geomagnetic data make the reconstructions greatly in error. However, Lockwood, Owens, and Barnard (2016) point out that Holappa and Mursula introduced errors by calibration against unreliable data, which they then exacerbated by using a less sophisticated and robust reconstruction procedure than that used by Lockwood et al To relate OSF to sunspot numbers, in this section we use the model of Solanki, Schüssler, and Fligge (2000), based on the continuity equation for OSF:…”