Estimating raindrop size distributions using microwave link measurements: potential and limitations

van Leth, Thomas C.; Leijnse, Hidde; Overeem, Aart; Uijlenhoet, Remko

doi:https://doi.org/10.5194/amt-13-1797-2020

Articles | Volume 13, issue 4

https://doi.org/10.5194/amt-13-1797-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/amt-13-1797-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 13, issue 4

Research article

|

08 Apr 2020

Research article |

| 08 Apr 2020

Estimating raindrop size distributions using microwave link measurements: potential and limitations

Thomas C. van Leth, Hidde Leijnse, Aart Overeem, and Remko Uijlenhoet

Download

Final revised paper (published on 08 Apr 2020)
Preprint (discussion started on 17 Apr 2019)

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'Review', Anonymous Referee #1, 14 May 2019
- AC1: 'Author response', Thomas C. van Leth, 20 Aug 2019
RC2: 'review_report', Anonymous Referee #2, 14 May 2019
- AC1: 'Author response', Thomas C. van Leth, 20 Aug 2019
RC3: 'Interactive comment to the manuscript Estimating raindrop size distributions using microwave link measurements', Anonymous Referee #3, 16 May 2019
- RC4: 'Reference list to the RC3', Anonymous Referee #3, 16 May 2019
  - AC1: 'Author response', Thomas C. van Leth, 20 Aug 2019

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by Thomas C. van Leth on behalf of the Authors (15 Oct 2019) Author's response Manuscript

ED: Referee Nomination & Report Request started (08 Nov 2019) by Saverio Mori

RR by Anonymous Referee #3 (21 Nov 2019)

Suggestions for revision or reasons for rejection

General comment:
The manuscript concerns a retrieval of drop size distribution (DSD) from attenuation of microwave links. Its focus is primarily on numerical validation of a DSD retrieval concept. This rose major concerns of all three reviewers to the initial submission as it was unclear to which extent were the proposed methods applicable for real microwave link observations affected by different sources of errors. The authors took an effort and substantially revised the manuscript. They provided additional numerical analyses investigating effect of signal quantization (precision) on the DSD retrieval and also changed several figures (time series plots to scatter plots) to better illustrate the results.
The additional analyses show that accuracy of microwave link observations in real networks will greatly affect reliability of the results and substantially limit the feasibility of a DSD retrieval in practice. The authors discuss the feasibility issue in the Discussion section, however, the Conclusions and Abstract section lacks clear and unambiguous statements about feasibility. E.g. the Conclusion section starts with the statement: “Using both simulations and actual link data we have shown that a DSD retrieval on the basis of multiple microwave link variables can be successful and highly accurate, but only when precise high-resolution records of received power are available.” Yes, the authors showed, that a DSD retrieval can be highly accurate on ideal data, but definitely did not demonstrate this high accuracy on real data. And of course, if precise high resolution records would be available, the method would be highly accurate also on real data, but this is not the case. I think that partly negative results are not a reason preventing the manuscript to be accepted for a publication, however, the authors should be much more careful not to give an impression that retrieving accurate DSD estimates from real microwave links is highly feasible. This concerns especially the Conclusion section and Abstract section. Furthermore, the manuscript would clearly benefit form more specific conclusions with respect to limitations of the method in practice. For example, the figure 5 together with the table 4 indicates pretty clearly what is the minimal required accuracy of observed attenuation and the range at which it is feasible to estimate the ‘mu’ parameter. This might be relatively easily linked with minimal rainfall intensity and minimal microwave link length (which affects the sensitivity to rainfall) required for a DSD retrieval.
Finally, it is difficult to interpret if the obtained results are in fact good or bad. The two-parameter method relies on calibrating relation between parameters of gamma distribution Lambda and mu using disdrometer data (or DSD typical for a local climate), i.e. typical values of Lambda and mu have to be known. It is therefore not clear, what is the real information gain, when using attenuation data compared to the baseline scenario when average (or most common) DSD pdf is assumed to be same for all the events. The results for such baseline scenario (or similar) should be, therefore, included into evaluation. The comparison to ‘’baseline scenario’ might also explain, why three-parameter method (which works independently on calibrated relation between ‘Lambda’ and ‘mu’) perform substantially worse than two-parameter method.
I recommend the manuscript for a publication after minor revisions concerning mostly presentation and interpretation of the results.

Specific comments:
Fig. 7, 8, and 9: Cannot see orange dots. Maybe transparency of points would help. In addition, consider showing correlation coefficients for all three methods.
P18L1-3: Is it so, that tree-parameter method can work also without disdrometer data used for calibrating relation between gamma distribution parameters Lambda and mu? If yes, this might be worth to note, because it is important benefit of this method compared to two-parameter method.
P22L5-6: Consider noting that resulting DSD is not shown.
P26L23-28: The evaluation of baseline scenario with most common Lambda and mu parameters used for all the events would enable more robust reasoning clearly separating effect of DSD observations from the information gain obtained from microwave link attenuation.
P27L11-13: The phrasing gives misleading impression that highly accurate retrieval of DSD was demonstrated also on real data, which is not correct.

Hide

RR by Anonymous Referee #1 (03 Dec 2019)

RR by Anonymous Referee #2 (20 Dec 2019)

RR by Anonymous Referee #4 (06 Jan 2020)

ED: Publish subject to technical corrections (25 Jan 2020) by Saverio Mori

AR by Remko Uijlenhoet on behalf of the Authors (03 Mar 2020) Author's response Manuscript

Short summary

We present a method of using collocated microwave link instruments to estimate the average size distribution of raindrops along a path of several kilometers. Our method is validated using simulated fields as well as five laser disdrometers installed along a path. We also present preliminary results from an experimental setup measuring at 26 and 38 GHz along a 2.2 km path. We show that a retrieval on the basis of microwave links can be highly accurate, provided the base power level is stable.