The manuscript "Data reduction of incoherent scatter plasma line parameters"

presents a study of plasma line detection in the EISCAT UHF radar dataset, 

using two approaches: a supervised (kind of manually adjusted) and an unsupervised

technique (with the detection based on the statistical parameters of the data

itself).

The manuscript is well written and easy to follow. The figures are generally 

clear and of good quality. The conclusions are well supported by the data.

Minor revisions/clarifications could improve the quality of the manuscript, 

which is of high standard already.

As the paper provides an important reference for plasma line analysis, it would 

be good to mention in the introduction plasma lines associated with instabilities

and turbulence. The interested reader will find this information in e.g. the

review by Akbari et al., but it would be good to mention this fact - and maybe

even add references to key publications.   

The discussion of Figure 7 is not very clear in lines 224-227. The description

mentions an increase of the plasma line frequency at "about 11:20:15 UT", and 

then mentions a "gradual increase" from 10:00-11:20 UT. Both are related to the

Sun, one to the "solar EUV radiation" and the other to "sun exposure". In my

opinion, this description should be consolidated. 

An interesting phenomenon is observed in Figure 7, namely the undulation of

the location of detected plasma line. This would indicate that the peak of the

profile is moving up and down in altitude. Some quasiperiodic variations of

the lower "boundary" of the F region are seen also in the top panel of figure 6.

While not directly related to main subject of the discussion, the authors

may want to comment on possible nature of the variations (could these be 

related to gravity waves? or variations in the ionospheric convection? or

yet something else?). Are these variations of the electron concentration 

profile capable of affecting the plasma line enhancement condition?

A note on Fig. 6 is that the date of the production of the figure is not

important, and probably should not be included in the plot. For Figure 7

it is a good idea to have a better setup of the time labels, having only 

two of them, at 10:29 and 11:30 does not help in pointing out specific times

in the plot. The bottom panel indicates the intensity units in eV, is this

correct? Figure 4 gives intensity in K/kHz, while there is also an extensive

discussion of the phase energy" (eq. 16). In Figure 5 the intensity is 

in K, and Figures 10 and 11 present k_B T_p. 

A minor detail: text predominantly refers to "altitudes", while some of the

figures are marked with "height" (while others with "altitude"). This could

probably be harmonized.

The paragraphs in lines 228-236 seem somewhat misplaced. In part they refer

to "similar plots" which are not presented, in contrast to the detailed 

discussion of the even presented. They also refer to Figure 10, not to be 

presented to the reader until later in paper. The consequences of observing

a narrower interval of the phase energy range than covered by the system is

not very clear at this point. A suggestion would be moving this material to

the discussion (line 257 and onwards), rather than having it in Results. 

I suggest considering renaming the last section to something like 

"Summary and Discussion". Conclusions would be expected to be a concise

summary of the findings, the nature of the presentation here is more of 

a discussion (among other aspects, four new figures are presented!). 

The purpose of showing presenting Figures 8 and 9 in the concluding section

is not very clear (paragraph in lines 254-256). Probably they would fit better

in the analysis section?

Line 262-263 "The absence of upshifted plasma lines means that we cannot 

use our data to calculate ionospheric currents." - what this ever the

intension in this work? 

Figures 10 and 11 present k_B T_p in eV. These would correspond to 

temperatures well in excess of those shown in Fig. 5 (left panel). I suspect

that the numbers are not directly comparable, but it would be helpful with a

clarification in the paper on how they should relate to each other.

It would be interesting to discuss how the choice of the dataset affects 

the results. The observations are from several days in a row in January, 

all during daytime. How would the observations look in a different seasons

and/or time of the day?