the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Towards sensible heat flux measurements with fast-response fine-wire platinum resistance thermometers on small multicopter uncrewed aerial systems
Abstract. This study demonstrates the feasibility of measuring temperature variance and sensible heat flux with self-calibrated fine-wire platinum resistance thermometers (FWPRT) on multicopter drones. The sensors are especially designed for light-weight, fast response-times and to be carried on miniature drones for turbulence measurements. A significant improvement was found in vertical profiling of temperature gradients compared to slower solid-state sensors, demonstrating reduced hysteresis between ascent and descent phases and accurate representation of strong gradients. More than 100 single flights with the sensors attached to drones of the SWUF-3D fleet were carried out in vicinity to a meteorological mast array at the WiValdi wind energy research park in Northern Germany. The comparison to sonic anemometers shows that mean temperature and temperature variance can be accurately measured within the background flow variability. The same applies for sensible heat flux, which was measured for the first time with multicopter UAS and the eddy covariance method. Sensible heat flux is a crucial parameter to understand the energy balance of the atmospheric boundary layer. An uncertainty of 50 W m-2 was determined with the constraint that only low wind speed conditions could be used to allow vertical wind speed measurements with the current algorithm. The results indicate that the temperature sensors are suited for sensible heat flux measurements, but further improvements are necessary with regard to vertical wind speed estimates to decrease the overall uncertainty.
- Preprint
(4713 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2025-241', Anonymous Referee #1, 02 Apr 2025
The manuscript "Towards sensible heat flux measurements with fast-response fine-wire platinum resistance thermometers on small multicopter uncrewed aerial systems" by Norman Wildmann and Laszlo Györy is well within the scope of AMT, presenting a novel method to measure the sensible heat flux by combining fast-response temperature measurements from a platinum fine wire temperature sensor and vertical velocity estimates retrieved from the UAS flight state data. While the authors carry out an extensive validation of their new measurement approach against a well-established reference system and also take the expected uncertainty of the experiment into consideration, the methods presented and used in this manuscript may not always be the optimal choice. Furthermore, the level of detail in the documentation of the methods and the scientific background is not always high enough. To put the manuscript in the correct scientific context it also lacks a few references. I therefore recommend major revisions before considering the manuscript for publication in AMT.
Major Comments:- Although the purpose of this manuscript is a proof-of-concept, detailed analysis is only provided for a narrow selection of data, without clearly stating why a data selection has been made. Table 1 indicates 3 calibration flights with 5 UAV each. Time series and spectra are only shown for 1 flight and 4 UAV. I acknowledge the fact that the authors are not hiding flight 69, which did not show great agreement, but at the same time, it is not even mentioned why the third calibration flight is not analyzed. All three flights should have been analyzed in the same way. On the other hand, second-order statistics, i.e., virtual temperature variance and sensible heat fluxes, are then presented for all flights, which gives a robust foundation for more solid conclusions. For completeness, I would, however, request to also include similar analyses as shown in Figure 9 and 10 for the vertical velocity variances.
- The validation against the sonic anemometer data and the cross-validation of two masts, to quantify the uncertainty in the experimental setup due to heterogeneities, appears to be carried out applying different methods. These methods should be clearly documented and designed to allow for the best and most fair comparability between the different data sets. This includes cross-correlation of the different time series to eliminate the time lag resulting from spatial separation (in reasonable time windows), using the same mast data for the mast-mast validation as for the UAV-mast validation. This implies that error statistics are computed for the same set of data and not for 172 samples vs 75 samples, which are based on different methods. Don't compare the buoyancy flux from the sonic anemometer to the sensible heat flux from the UAS (this may be just an error in the terminology used). A more thorough data processing as suggested, in particular the time-lag correction, may contribute to a lower error estimate for the UAV turbulence parameters as well as for the experimental setup.
- The sonic anemometer data is sampled at a higher frequency (not clearly stated) than the UAV data and Figure 8 indicates that the smallest resolved scales are relevant. Since these scales cannot be resolved with the UAS the sonic data should be downsampled to the same frequency and use matching time windows (corrected for the time lag) as the UAS data for a direct comparison.
- The focus of this manuscript is on high-resolution temperature measurements and sensible heat fluxes; however, little attention is given to the measurement of the vertical velocity. A corresponding manuscript has already been published demonstrating the ability of the system for the retrieval of vertical velocity fluctuations. However, it is not sufficient to only refer to it without summarizing the important details of the methods used to determine the vertical velocity, the uncertainties and limitations of this method, and potential improvements or changes to the method presented in Wildmann et al. (2022).
- The article largely relies on previous work by one of the authors, but it should refer to the peer-reviewed article Wildmann et al. (2022) and not the corresponding preprint/discussion paper.
- The sensor placement in a rather narrow housing may hinder vertical motion and potentially trap air. The design of this housing needs to be clearly described and some sort of validation, e.g., CFD simulation, should be presented to verify potential flow distortion effects of the housing under different wind conditions. In particular, in free convection cases, the ambient air velocity vector can be almost vertical. It is not clear whether the housing allows for proper airflow under such conditions, so there may be a minimum wind speed limit for this method.
- The method presented in the manuscript has great potential, but a clear statement on the motivation of measuring atmospheric turbulence with one or a fleet of small UAS is lacking.
- The chosen approach of using a fine wire temperature sensor and vertical velocity estimates based on the UAV state should be put into better perspective and not only compared against the fixed-wing multi-hole probe approach. Relevant publications (Fuertes et al. 2019; Greene et al. 2020; Ghirardelli, et al. 2024).
- Overall, quite a lot of data has been taken into account and the results appear robust, but the atmospheric conditions during which the experiments were carried out were rather limited. I request to a) document these conditions better, including information on atmospheric stability (no stably stratified cases or negative heat fluxes were sampled), wind speed and direction (already in Table 1), footprint and soil appearance (significant latent heat fluxes), solar radiation and cloud cover and b) take these conditions into account when interpreting the results, drawing conclusions, and potentially also for the outlook.
- I lack a clear statement on the limitations related to the currently limited frequency resolution. The relative contribution of small-scale turbulence to the total flux may become more important under certain conditions. What does this mean for the applicability of this method, e.g. for stable boundary layer cases?
- Some results are not presented consistently. Fig 6, 9, and 10 use different marker colours to distinguish different height levels, but the error statistics are presented as a bulk. The discussion of these figures sometimes lacks any discussion of differences between the measurement heights.
- Some Figures are hard to read and need to be improved, including a more detailed description of what is shown.
Specific comments:Abstract:ol {margin-bottom:0in;margin-top:0in;}ul {margin-bottom:0in;margin-top:0in;}li {margin-top:.0in;margin-bottom:8pt;}- L8-9: I slightly disagree with the statement that this is the first time sensible heat fluxes have been measured using multicopter UAS (e.g., Fuertes et al. 2019; Greene et al. 2020; Ghirardelli, et al. 2024). I agree that this is the first time it has been done with the method presented in this manuscript or more generally, applying the EC method from small UAS.
- L11: Since you show that the method works for wind speeds up to 8 m/s, I recommend stating this threshold specifically instead of using the term "low wind speed conditions".
Introduction:ol {margin-bottom:0in;margin-top:0in;}ul {margin-bottom:0in;margin-top:0in;}li {margin-top:.0in;margin-bottom:8pt;}- Although it is stated that this study focuses on small multicopter UAS, it still deserves some attention that EC measurements can be performed from large UAS using standard sensors.
- I suggest first providing some background on the EC method and other relevant methods for estimating turbulence from UAS. The parameters (and their resolution) needed to measure turbulence using the EC method should be clearly stated, as well as the different sensors and methods for retrieving these parameters, including a critical evaluation of the different approaches for estimating turbulence, particularly the sensible heat flux and the difference between the buoyancy and sensible heat flux. It should also be made very clear that the described method does not rely on a dedicated wind sensor.
- L29-30: It is not clear that this refers to resolving mechanical turbulence by making use of the UAV's INS data.
Section 2:ol {margin-bottom:0in;margin-top:0in;}ul {margin-bottom:0in;margin-top:0in;}li {margin-top:.0in;margin-bottom:8pt;}- This section should also include important details on the UAV and the different versions used.
- L57: Maybe a typo in "MAX38165" - could be "MAX31865".
- L59: Does the conversion to temperature also depend on the length and diameter of the wire? I assume this is one of the main reasons why this setup still requires calibration.
- L61: Also indicate the Arduino board in Figure 1. The purpose of this microcontroller board is not clear to me. The I2C and SPI could be directly wired to the flight controller and logged with some custom script.
- L65-67: A more detailed sketch of the housing would be relevant. It looks very narrow, and it is not clear where the air inlets and outlets are. Since the purpose of the presented setup is to measure sensible heat fluxes, and you restrict yourself to rather low wind speeds, it would be very important to obstruct the vertical flow as little as possible. The housing looks like it is only allowing for horizontal flow. As shown in Wildmann et al. (2013), in most cases, radiation shields are not necessary for wires with very low thermal inertia, so why do you still use a radiation shield? In case the housing does not allow for vertical flow passing through it should be mentioned that strong convection may not be measured correctly and you should provide a minimum wind speed or angle of attack.
- L70: What is the minimum wind speed required for the wind vane mode to work properly?
- L71: Replace "sensitive elements" with "sensing elements".
Section 3.1:ol {margin-bottom:0in;margin-top:0in;}ul {margin-bottom:0in;margin-top:0in;}li {margin-top:.0in;margin-bottom:8pt;}- L80: "The analyses in this study are based..."
- L84-85: Provide a more detailed description of the MMA pattern. From Figure 2 and Table 1, I can only guess that you aimed to sample at each of the 4 positions marked on the map at 25 m and 90 m. There are probably practical reasons why you could not always operate 8 UAV at the same time, but in flight 41, you indicate 9 UAVs. Where did the last one fly?
Section 3.2:ol {margin-bottom:0in;margin-top:0in;}ul {margin-bottom:0in;margin-top:0in;}li {margin-top:.0in;margin-bottom:8pt;}- L92-94: I am confused by this statement as it is not clear whether you used the old or new version of the UAV in this study. Can you also list the flight up to 300m in Table 1? Section 2 does not include detailed information on the UAV, neither on the new nor old version.
- Table 1:
- I think the table should be cross-referenced somewhere in section 3.
- Provide a description of the different columns in the caption. Consider using "MMA" instead of "mast array".
- I assume the meteorological conditions are from one of the masts. Which one and which observation level? How are they averaged?
- Consider sorting the UAS by their IDs.
Section 4.1:ol {margin-bottom:0in;margin-top:0in;}ul {margin-bottom:0in;margin-top:0in;}li {margin-top:.0in;margin-bottom:8pt;}- Table 2:
- Are the names related to a specific model version? If yes, the differences should be described; otherwise, this column is rather irrelevant and could be dropped.
- How is delta Ts computed?
- Please provide information on the sampling rate/frequency of the sonic anemometers.
- L105: Sonic temperature is very close to but not equivalent to virtual temperature (also in L121).
- L105: According to the manual (https://www.thiesclima.com/db/dnl/4.383x.xx.xxx_US-Anemometer-3D_e.pdf), the cross-wind correction can be enabled in the settings (TC = 1). Are you sure that the data is not already cross-wind corrected?
- L109-117: Make a clear difference between sonic temperature and temperature.
- L106-107: Do the Thies 3D sonic anemometers provide temperature output along orthogonal axes u, v, w or along the non-orthogonal axes given by the transducer paths? I found a corresponding statement in the manual, but it doesn't make any sense to me to also rotate the temperature data into an orthogonal coordinate system.
- L111: Also provide the sensor details on the inflow mast to make it comparable to the instrumentation on the other masts.
- Figure 3:
- Add a list explaining the error statistics provided in the figure.
- Indicate temperature in the axes labels, e.g., Ts_MMA and Ts_IEC.
- What does IEC stand for?
- To which mast do the parameters indicated as text refer? It would be better to provide these error statistics for each mast.
- Why is the data from the central mast and the other heights not shown here?
Section 4.2:ol {margin-bottom:0in;margin-top:0in;}ul {margin-bottom:0in;margin-top:0in;}li {margin-top:.0in;margin-bottom:8pt;}- T_s and T_v are not equal, but a more direct comparison to T_s could be achieved by simply modifying Eq5, by changing 0.61 to 0.51 you compute T_s instead of T_v.
- E4: What is r_v? Also, mixing ratio (m_h)?
Section 4.3:ol {margin-bottom:0in;margin-top:0in;}ul {margin-bottom:0in;margin-top:0in;}li {margin-top:.0in;margin-bottom:8pt;}- The difference between the sensible heat flux and the buoyancy flux (~) cannot be ignored and requires some discussion, e.g., estimating the difference for relevant conditions.
- It should also be made clear that the covariance of w and T_s is close to proportional to the buoyancy flux.
- L137: m_h is already defined.
- L143: This is rather an input for the outlook: Could longer measurement periods be achieved by "stitching" time series from different UAVs together, i.e., one drone replacing another after 10 minutes or so?
- Eq8: It should be mentioned that R0, A and B are the coefficients you determine in the calibration experiment. For clarity, it may also be beneficial to use lowercase letters for these coefficients.
- Fig4: It would be great to also show the parameters (R0, A, B) determined in this calibration experiment, e.g., in the form of a table or directly in the legend (move the legend to the side of the figure). It would also be very interesting to see whether there is any drift. Have these calibration experiments been repeated at some point for some of the sensors, or is it maybe possible to do this now? Alternatively, the stability could be determined based on a comparison to the HYT sensors, preferably over a longer time period than the 4 days covered in this manuscript. This would, however, require that the HYT sensors have acceptable long-term stability.
- L155-165: This section should be linked better to Eq 8. You start with a second-order polynomial fit and determine that B tends to 0, so you can use a simpler linear model. A bias is accounted for by adjusting R0, and the slope coefficient is A.
- L165: Instead of providing an average abs RMSE, it would be beneficial to provide uncertainty estimates for both the 0D and 1D calibrated sensors. Following the purpose of this manuscript, it would be of high value to demonstrate the value of a higher calibration effort. This, in combination with tracing the individual sensors, may also be valuable information for the interpretation of the results.
Section 5:ol {margin-bottom:0in;margin-top:0in;}ul {margin-bottom:0in;margin-top:0in;}li {margin-top:.0in;margin-bottom:8pt;}- The header should be "Results and discussion".
Section 5.1:ol {margin-bottom:0in;margin-top:0in;}ul {margin-bottom:0in;margin-top:0in;}li {margin-top:.0in;margin-bottom:8pt;}- L168: For clarity, you should use the term mechanical turbulence.
Section 5.2:ol {margin-bottom:0in;margin-top:0in;}ul {margin-bottom:0in;margin-top:0in;}li {margin-top:.0in;margin-bottom:8pt;}- The presentation of the results should include more background information on the atmospheric conditions during these 3 calibration flights, the setup of the experiment, and what is shown in the figures. A few suggestions and questions that should be answered here:
- Repeat the date and time of the flights in the text, refer to Table 1.
- From Table 1, I get that all UAVs were flown at 99m, so why are there data points for 25m in Fig6b?
- The three flights were completed within a period of less than 1.5h. How did the atmospheric conditions change during this period? How about solar radiation and atmospheric stability?
- How do you get to around 30 data points when using 3 flights with 4 UAVs?
- Can you add the information on which sensor was used and how they were calibrated? It would be interesting to see whether it is possible to detect the effect of the higher calibration effort for some of them (the temperature range covered here may be too narrow).
- Fig 6a: The figure is not clear enough, since it is not possible to distinguish individual lines. It is not even possible to see whether it is 2 or 4 UAV temperature curves or whether one of them has a larger variability than the other. Consider using a stacked figure layout or at least apply offsets of e.g. 0.25K increments to the different lines and use different colours.
- Fig 6b: Label the axis correctly (T_sonic, T_FWPRT), correct the units (K), use different colours/symbols for the different UAS/FWPRT. The caption should indicate the label (b) and state the closest mast is south.
- L185: Do you mean below 0.1 K?
Section 5.3:ol {margin-bottom:0in;margin-top:0in;}ul {margin-bottom:0in;margin-top:0in;}li {margin-top:.0in;margin-bottom:8pt;}- L193: I agree that fast sensors are preferable for soundings, but they also have to be accurate over a wide temperature range, have good stability and robustness.
- Fig7: Change labels to FWPRT. PT100 is misleading.
- It would be relevant to provide more details on the response time correction applied, e.g., the time constant used, the function applied, and filtering to avoid the amplification of noise. How does the applied time constant compare to the one stated by the manufacturer?
- The large relative difference between ascent and descent at the top of the profile for the FWPRT in Fig7b deserves a more detailed interpretation. How long was the UAV hovering at this altitude before starting its descent? Could this be an artefact resulting from a wrong altitude measured by the barometer due to reduced thrust when transitioning from ascent to descent?
Section 5.4.1:ol {margin-bottom:0in;margin-top:0in;}ul {margin-bottom:0in;margin-top:0in;}li {margin-top:.0in;margin-bottom:8pt;}- The section header is a bit misleading since the section also covers the comparison against the sonic anemometers. I recommend focusing on the validation of the calibration flights and including all three of them.
- L222-223: The interpretation of flight #69 is unfortunately a bit shallow. From Table 1, I would expect similar conditions between all three calibration flights. From TI for #69, I would expect the vertical velocity variance measured by the sonic to be in the order of 0.77. A value of 0.3 indicates highly non-ergodic turbulence or some other problem, potentially flow distortion by the mast. Why is flight 71 not shown?
- Figure 8: I have problems seeing whether it is three or four lines for the SWUF_T. Can you use different colours and labels also indicating the date and sensor ID for the same type of sensor?
Section 5.4.2:ol {margin-bottom:0in;margin-top:0in;}ul {margin-bottom:0in;margin-top:0in;}li {margin-top:.0in;margin-bottom:8pt;}- L232: For clarity, you should add "at the corresponding height level" or similar.
- L234: R2 is not the correlation coefficient (correct further up but also wrong further down).
- L236-237: In Figure 6a, it looks like there is a clear time lag between the UAV-based and sonic-based temperature signals, even at a TI value of 22.2. Even if it does not eliminate the error induced by this spatiotemporal separation, there is a good chance to improve your agreement by determining the time lag from cross-correlation analyses. This could also have the nice side effect that you could compare time series directly and compute correlation coefficients for the instantaneous time series. The sonic data would, of course, have to be downsampled to the same frequency as the UAS.
- Although the error statistics in Fig 9 don't distinguish between the two heights shown (I think they should), Figure 9b indicates that there is a larger scatter at 99m compared to 25m, which is somewhat surprising. This should also be discussed.
Section 5.4.3:ol {margin-bottom:0in;margin-top:0in;}ul {margin-bottom:0in;margin-top:0in;}li {margin-top:.0in;margin-bottom:8pt;}- L245: Use a different notation to express numerical ranges, e.g., "2 m to 5 m".
- L249: This statement should include a reference.
- L250-251: As mentioned, the 10min sampling period is rather short and may thus be subject to large uncertainties due to poor statistical representation of larger eddies. Non-stationary is often, but not necessarily, related to too long averaging periods.
- L255-260: Both correlations, sonic-sonic and UAV-sonic, could be improved when correcting for the time lag. Given the fact that the masts are separated by 100m and the UAVs distance to the reference sonic is around 50m, the comparison would also become fairer. The number of samples in Fig. 10a and b is very different. Why is this the case and how does this affect the error statistics? For a fair comparison, you should re-compute the error statistics for the same periods as shown in Figure 10b. You should also downsample the sonic anemometer data, since the scales >4 Hz still carry some energy, but they cannot be resolved with the UAS. Why do the sonic data in b) not appear in a) with the same value? For example I don't see a data point with a flux exceeding 400 W m-2 in a) but there is one in b). If the data processing is slightly different, this is important to mention.
- L256: You mention only 99m but also show 25m in the corresponding figure.
- L259: Add information on how the filtered and unfiltered data are displayed and specify whether the error statistics correspond to filtered or unfiltered data. Apply the same filtering to the sonic data when computing error statistics and also indicate the corresponding data points with transparent markers.
- L260: Correct the units.
- L261: Repetition: "Figure 10b shows...".
- L265-L266: In the first place, this only shows that horizontal wind speed has an impact on the uncertainty. Vertical wind speed variance may scale with wind speed and can be suggested to be the main factor behind this uncertainty, but this is not what is shown here.
- L269-271: Why are only UAS 15, 25, 21, 32 mentioned here when you operated so many more UAS during these flights? Are these the drones at 25m?
- Figure 11: What exactly is indicated by the shades and error bars? I suggest recomputing the mast data using a running average, e.g., with a 10min averaging window to provide smoother curves. The spiky data is a result of block averaging often resulting in sub-optimal start and stop times. This block averaging is likely to contribute to the second observation you highlight - substantial variability due to small offsets in time and space. I expect the comparison to become more intuitive when using running averages.
- L272-282: Indicate the subfigure you refer to.
- L276-279: Does the solar radiation or cloud cover suggest strong differential heating? The footprint of the mast and the drone at 25m are typically not strongly influenced by the surface right below. In the case studied here, winds from the west of about 8 m/s suggest a different footprint.
Conclusion:ol {margin-bottom:0in;margin-top:0in;}ul {margin-bottom:0in;margin-top:0in;}li {margin-top:.0in;margin-bottom:8pt;}- L287: 5 Hz or 4 Hz as stated in L227?
- L298-290: Stationary and homogeneous flow conditions.
- L293: Specify turbulent sensible heat fluxes or buoyancy fluxes.
- L298: Future work should also focus on the validation of this method in a wider range of atmospheric conditions, e.g., stable conditions, very weak turbulence, and free convection.
Citations:ol {margin-bottom:0in;margin-top:0in;}ul {margin-bottom:0in;margin-top:0in;}li {margin-top:.0in;margin-bottom:8pt;}- Check citation style (textual or parenthetical citations).
Figures:ol {margin-bottom:0in;margin-top:0in;}ul {margin-bottom:0in;margin-top:0in;}li {margin-top:.0in;margin-bottom:8pt;}- All subfigures should have labels if referred to as a, b, c.
- The captions should include all relevant details to understand the figures or tables.
- If displaying different measurement levels or different sensors in different colours, the error statistics should also be presented separately for each category.
- Labels should be consistent. I see PT100, SWUF_T, UAS.
References:ol {margin-bottom:0in;margin-top:0in;}ul {margin-bottom:0in;margin-top:0in;}li {margin-top:.0in;margin-bottom:8pt;}- B. R. Greene, S. T. Kral, P. B. Chilson, and J. Reuder, “Gradient-Based Turbulence Estimates from Multicopter Profiles in the Arctic Stable Boundary Layer,” Boundary-Layer Meteorology, Mar. 2022, doi: 10.1007/s10546-022-00693-x.
- M. Ghirardelli, S. T. Kral, E. Cheynet, and J. Reuder, “SAMURAI-S: Sonic Anemometer on a MUlti-Rotor drone for Atmospheric turbulence Investigation in a Sling load configuration,” EGUsphere, pp. 1–28, Sep. 2024, doi: 10.5194/egusphere-2024-1548.
- F. C. Fuertes, L. Wilhelm, and F. Porté-Agel, “Multirotor UAV-Based Platform for the Measurement of Atmospheric Turbulence: Validation and Signature Detection of Tip Vortices of Wind Turbine Blades,” J Atmos Ocean Tech, vol. 36, no. 6, pp. 941–955, Jun. 2019, doi: 10.1175/jtech-d-17-0220.1.
- N. Wildmann and T. Wetz, “Towards vertical wind and turbulent flux estimation with multicopter uncrewed aircraft systems,” Atmospheric Measurement Techniques, vol. 15, no. 18, pp. 5465–5477, Sep. 2022, doi: 10.5194/amt-15-5465-2022.
Citation: https://doi.org/10.5194/egusphere-2025-241-RC1 -
AC1: 'Reply on RC1', Norman Wildmann, 04 Jun 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-241/egusphere-2025-241-AC1-supplement.pdf
-
RC2: 'Comment on egusphere-2025-241', Anonymous Referee #2, 09 Jun 2025
Towards sensible heat flux measurements with fast-response fine-wire platinum resistance thermometers on small multicopter uncrewed aerial systems
Wildmann & GyöryThis manuscript is well written and provides a topic that is relevant to the AMT readership. Fast response temperature measurements using UAS are typically achieved using fixed-wing aircraft. It is valuable to see more studies that demonstrate the ability to use rotary-wing UAS for this purpose. The authors are well qualified and the experimental techniques are suitable to demonstrate the objectives of the study: that sensible heat flux measurements can be achieved with the assistance of rotary-wing UAS. I did find it difficult at points to follow some of the data processing steps, but this can be rectified in a revision. I feel that the paper would benefit from a major revision.
The paper would benefit by having more background on eddy covariance measurements
The paper references papers from previous work by the authors, which is appropriate, but the reader should not need to read those papers to follow the flow of the proposed study. More information from the previous studies (as related to the present work) should be provided as a summary.
This is related to the previous point but more information on the actual UAS would be useful.
The authors should provide more information on the sensitivity of the wind vane mode to wind speed.
What is the practicality of performing such measurements on a routine basis? Is this just a proof of concept? Is there an operational motivation of the study?
Are the FWPRT sensors available commercially? How prone are they to damage? Is this a limiting factor?
Maybe I missed it, but it is not clear to me how the vertical wind data (from the towers?) are paired with the temperature measurements from the copters?
I did not see information on the sampling rate of the sonic anemometer
Which instrument was used to measure humidity on the UAS?
To me it seems that demonstration of the FWPRT measurements against the tower and the onboard solid state thermometer would be sufficient for a study. The inclusion of measurements of flux adds extra layers of complication, which are not necessarily adequately resolved in the paper.
Citation: https://doi.org/10.5194/egusphere-2025-241-RC2
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
213 | 46 | 13 | 272 | 14 | 12 |
- HTML: 213
- PDF: 46
- XML: 13
- Total: 272
- BibTeX: 14
- EndNote: 12
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1