|For the introduction, I have some suggestions, all published in high-ranked journals:|
* another nice example for atmospheric boundary layer studies of turbulent fluxes using RPAS is
Wildmann N., Rau G.A., and Bange J., 2015: Observations in the early morning boundary layer transition with small RPA. Boundary-Layer Meteorol., 157, 345–373.
* and for precise ABL wind vectors using RPAS:
Wildmann N., Bernard S., and Bange J., 2017: Measuring the local wind field at an escarpment using small remotely-piloted aircraft. Renewable Energy, 103, 613–619.
* and for aerosol in the ABL using RPAS:
Platis A., Altstädter B., Wehner B., Wildmann N., Lampert A., Hermann M., Birmilli W., and Bange J., 2016: An observational case study on the influence of atmospheric boundary-layer dynamics on new particle formation. Boundary-Layer Meteorol., 158, 67–92.
4/23: "hole 1 measures the pressure at the stagnation point of the tip" - still, this is not correct, as explained in my first review. However, at small angles of attack and sideslip, the error caused by this misunderstanding might not be large. I am frustrated to see that this mistake was done in other publications too, obviously. Correct:
"hole 1 gives an estimate of the pressure at the stagnation point of the tip"
Sections 2.3 and 3: And I still do not see the point in using simplified equations that cause uncertainties regarding the measured wind vectors while the precise equations are well known and can by applied easily. The analysis of errors caused by 10 degrees of angles of attack and sideslip is more costly than the application of the correct equations. Anyway, this issue is now disarmed by citing the appropriate literature.
5/20 and Tab 1: SI units instead of 'mbar' would be contemporary
The discussion of the power spectra (section 4.1) and the TKE (4.2) is still weak. The argument that other publications show similar deviations e.g. to sonic anemometers is not satisfactory, since no reason was found for this systematic discrepancy (“be further investigated in the future”). However, the reason could also be faulty sonic measurements of turbulent fluctuations.
Fig. 12, 15 and 18: The meaning of captions 'Radar flight altitude' etc (explanations to the four data curves) are not clear to me.
14/23: 'to remove RPA motion from the wind vectors measured by the 5-hole probe' does not describe the method. It is not a data correction, but a coordinate-system transformation.
* indices like in \sigma_cloud shall not be italic
* almost all Figures: axis labels etc are too small
Summing up: The wind-vector calculation and the analysis of the RPA data is somewhat circuitous. Spectral and TKE comparisons to sonic measurements are not convincing. The comparison of vertical-wind distributions gained from RPA and Radar (ig. 12, 15 and 18) requires plenty of discussion and explanation (pages 11 to 13) and is somewhat puzzling. I wonder if the analysis and the results could have been explained more straight forward and thus more conclusively.
However, it is a measurement-technology journal, and the authors demonstrate what RPAS are good for and how remote sensing of clouds can be accompanied by quite in-expensive in situ measurements. This manuscript shows (probably for the first time) how RPAS can be used also for cloud physics.