Comment on “The IDV index: Its derivation and use in inferring long‐term variations of the interplanetary magnetic field strength” by Leif Svalgaard and Edward W. Cliver

[1] Svalgaard and Cliver [2005] (hereinafter referred to as SC05) use the IDV geomagnetic index to infer the variation of the interplanetary magnetic field strength, B, since 1872. They find that ‘‘B increased by 25% from the 1900s to the 1950s’’ (paragraphs 1 and 24) and that this ‘‘is in contrast to the more than doubling of B during the 20th century obtained from an analysis of the aa index by Lockwood et al. [1999]’’ (paragraph 24). We agree with neither statement. We identify a number of errors and biases in the analysis of SC05, each of which acts in the same direction, namely to reduce the true long-term drift. The key result of Lockwood et al. (hereinafter referred to as LEA99) is a doubling, not in B, but rather in the total coronal source flux which is proportional to the radial IMF component, Br. This is an important distinction which we discus here in section 3. We show (in section 2) that SC05’s own analysis gives a drift in decadal averages in B of 38%, not the ‘‘ 25%’’ that they quote, but we also point out (section 4) that SC05’s method for dealing with data gaps gives rise to a (small) bias and that their regression procedure (which is compared to others in section 5) is not robust (sections 6 and 7), in addition to them not taking into account the distinction between B and Br . We here argue that a simple ordinary linear regression procedure, as used by SC05, is inferior to the method employed by LEA99 (section 9) but does nevertheless show that the IDV index is fully consistent with the doubling in the open solar flux found by LEA99 (section 8).