The author should revise the ms considering the comments of the two reviewers and the following comments.

The explanation of the the two values of H exponent, in the middle of Conclusions and in the last sentence of the abstract, needs substantiation. This pertains to the motion of the auroral electrojet over Tromso and the connection of the time scales to the relevant details with appropriate references should be provided to demonstrate the "excellent" agreement.

The auroral electrojet index AE is referred to at many places and its attributes need correction. For example, AE index in not an average (lines 250 and 330). The "deseasonalization" (line 315) also will not be meaningful for AE index.

The physics of geomagnetic activity is mentioned without specifying which physical processes and the relevant time/scales, e. g. lines 200 and 215. Since this refers to specific results in the paper the connection should be made, with refrences included.

The term "complexity matching" (line 30) is not a widely used term and may not be useful in this context. If the author would like to use it a brief description would clarify its usage.

The author's response and changes in the ms have addressed most of the earlier comments. However the concern on the conclusions (Abstract and Conclusions) still need careful examination so that the results are conveyed properly to the reader.

The discussion of the results in lines 325 -336 is centered on the two values of the exponent, viz. 1.5 (mostly 1.42?) and 1.55. The following are the relevant issues on this:

#1. How significant is the difference in these values - what are the error bars and how are they computed?

#2. Assuming the difference in the values of the exponent is significant, the expression "fast switching of direction in the motion" should be described better. The relevant feature here is the change in the characteristics of the system between persistence and anti-persistence.

#3. The purported explanation (Lines 330-332) in terms of changes in the auroral electrojet is not clear. The scaling exponent is a statistical quantity and ascribing it to specific features, e.g., evening, motion across latitudes, etc., need more detailed study. This discussion needs significant elaboration or should be omitted.

#4. The earlier comment on the need for substantiating and quantifying the "excellent agreement" (line 332) has not been addressed by the author. A comparison with past analyses, e. g., Takalo and Timonen (1994) should be more specific and quantitative to qualify as excellent.