the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 17 Jun 2025
HoloTrack: In-Situ Holographic Particle Tracking of Cloud Droplets
Abstract. We present HoloTrack, a novel, fully autonomous measurement system designed to capture three-dimensional cloud droplet data and provide detailed insights into droplet dynamics, their spatial distribution and velocity. The HoloTrack system integrates a high-accuracy holographic imaging system with environmental sensors, including pitot tubes for airflow measurements, and a motion-tracking system. Designed for deployment on platforms like the CloudKite and hence compact and autonomous design, HoloTrack is also ideally suited for deployment in laboratory or ground-based environmental research. The system records up to 25 hologram pairs per second, each of which provides two independent measurements of droplet position, size, and shape and measures individual droplet velocities in longitudinal and vertical direction. The holographic system reliably detects particles down to 10 µm, within a sample volume of 17 cm3 of each hologram, which results in 21.5 cm3 sampled particle position and size and 12.3 cm3 sampled velocity for a mean displacement of 0.5 cm within hologram pairs. Reliable sub-volumes for measuring droplets at different yaw angles, to account for the influence of the instrument body are defined. The droplet velocity is measured with errors of less than 1.5 % for mean velocities of 8–10 m/s, but the flexible timing allows adjustment for larger mean displacements which increases accuracy if desired. A series of ground tests and a maiden flight tests validated the system’s capabilities, confirming detection, robustness, automation and its ability to accurately measure droplet dynamics. HoloTrack’s unique combination of holographic particle measurements including capturing their velocities makes it a powerful tool for advancing our understanding of cloud microphysics, including droplet spatial distribution, coalescence, entrainment, and turbulent mixing processes.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(45038 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(45038 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2025-1774', Anonymous Referee #1, 04 Jul 2025
-
AC1: 'Reply on RC1', Birte Thiede, 29 Aug 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-1774/egusphere-2025-1774-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Birte Thiede, 29 Aug 2025
-
RC2: 'Comment on egusphere-2025-1774', Anonymous Referee #2, 09 Jul 2025
This paper thoroughly and rigorously describes an instrument for tracking of cloud droplets measured by a holographic imaging system. The paper should be accepted after addressing the following points:
Major comments:
- The paper is very detailed, even tedious in places. Any streamlining would be appreciated, especially to reduce repetition. In particular, much of the material in the “Discussion” section is not really discussion, but simply repeating what has already been stated clearly in the main text of the paper. The Discussion section should be rewritten to focus on putting results into the bigger scientific context and describing plans for future improvements or applications.
- The big picture is missing, especially for those who may not be familiar with the concept of PIV or particle tracking. In the Introduction the measurement approach should be more clearly explained: two holograms recorded in rapid succession, such that they have significant spatial overlap, and then matching particles observed in both sub-volumes in order to obtain a velocity vector. One part of that vector is due to the mean advection of the flow, the other part is due to turbulence. Perhaps a figure illustrating this, the way you would explain in a presentation to a broad audience, could be useful. On this point: is this really “tracking” or is it velocimetry? With only two frames, it seems like a stretch to refer to this as particle tracking. But I agree it’s also not traditional PIV based on image correlation. Is there standard terminology in the field? If so, it’s best to introduce it and clearly define it.
- In the end it’s not clear whether holograms were obtained from within natural clouds. On line 268 it is stated that in-situ holograms could not be evaluated, but the reason is not clearly stated. If no in situ data are available, it should be stated directly that the tests of the holographic tracking are under laboratory conditions only. This will ensure the scope of the paper is accurately stated.
Detailed comments:
Lines 8-9: It is not clear in the abstract where these volumes come from and what they mean. If you wish to keep the numbers, please explain why the volumes are different, i.e., that the velocimetry depends on two samples separated by very short time, and therefore overlapping in space.Line 19: It’s an oversimplification to state that cloud properties are determined by the microphysics of clouds. One could just as easily claim that microphysics and other cloud properties are determined by cloud dynamics. The truth is that microphysics both follow and help determine other large-scale cloud properties. Please rephrase.
Line 29: It seems strange to cite a review paper on mixed-phase clouds as a reference for holographic cloud imaging. Perhaps there is a more appropriate one?
Line 43: The sentence would read better with a comma after “cloud droplet measurement”.
Last two paragraphs of Section 1: I was confused until the very end, and in fact I thought the paper was about a 2D PIV system. It takes too long to figure out what the paper is about, and even in the last paragraph it’s not really clear how the system works (see related point in “Major Comments” above). For example, “first airborne instrument capable of capturing hologram pair tracking of droplets” is quite obtuse. Please restructure the Introduction to clearly state the problem and to clearly outline the overall approach so readers can have an idea of what the paper is about.
Line 78: Delete “at most”.
Caption for Figure 1: I don’t know what OPC-N3 is referring to… need to define.
Line 111: The concept of reconsruction has not been introduced, so please do that first.
Line 112: Is 8-bit resolution sufficient to resolve the fringe visibility needed for high spatial localization of particles and particle size? See, e.g., discussion of fringe visibility in Fugal et al. 2004 (Applied Optics).
Line 126: State “Section 3.2”.
Line 130: Show numbers to convince that this limit is not typically reached. I believe that it is reached in polluted clouds.
Line 132: Three uses of “sample” “sampled” and “sampling” in one sentence… please rephrase to make it clearer and less repetitive.
Line 139: I don’t understand where the 45 cm constraint comes from… please explain more clearly.
Figure 2 caption: Meaning of “Thorlabs systems” is unclear. Also, meaning of “this simulation code” is unclear.
Line 182: “Despite the simplicity of the described timing.”
Section 2.2.2: It’s not clear why the laser is run at such a high repetition rate. Why not simply fire two pulses with the required delay? Presumably this must not be possible, but the reason is unclear.
Line 194: It’s unclear what simpleRTK2B and U-Blox ZED-F9P refers to.
Line 244: “36 cm to the sample volume and are not look directly into…” is unclear, needs to be reworded.
Line 251: Provide a date and location for the IMPACT campaign.
Line 298: “degree” symbol is not properly coded in latex.
Line 326: “Caused likely with 8-point-averaging filtering for 1D pitot tube” is unclear and grammatically incorrect… need to improve.
Lines 330-331: Units for energy dissipation rate are incorrect.
Lines 335-338: The Stokes number is only one parameter that governs decoupling from the flow. The settling parameter also needs to be considered. I expect that for the larger drop diameter of 50 um the gravitational setting is an important contributor. Please discuss and evaluate.
Line 350: “Recall” is not yet defined, so the sentence is not clear.
Lines 361-364: I don’t understand this, e.g., what is meant by “accurate measurement of precision”. Accuracy and precision have two different meanings.
Figure 7: I don’t recall that d_{gt} has been defined. What does the subscript indicate?
Lines 371-372: Is this a reasonable assumption? Why is it not possible to perform the analysis on an actual hologram pair?
Lines 373-374: s_m is used for both quantities… is this correct?
Line 401: “weighted by the square root”… square root of what? And why?
Figure 9: The “small dark blue square” is difficult to find in the image. Please enhance and/or enlarge to make it more evident.
Line 419: Do you mean “more turbulent data”?
Line 426: Define PMR here so it is clearer what it indicates in Equation 7.
Lines 432-440: I am completely lost. This discussion is quite tedious. Can it be simplified?
Line 447: I don’t see that “both cases” are defined.
Line 449: Is it really constant? It seems to me that it shows a trend.
Line 458: Define turbulence intensity.
Line 471: “Turbulence that exceeds the random error in inter-particle distances” is illogical. Please correct.
Figure 12: Are there results related to u versus w, which could hint at a sedimentation effect? It would be nice to include, in order to have a little bit of science in the article.
Line 484: “by the boundary layer”
Line 485: “expelled from the arms wake accumulate” is not clear… do you mean “from the wake of the arms”?
Line 488: “regions are reach up to” should be “regions reach up to”?
Lines 519-530: As noted above, this is all a repeat of what is in the article… not a Discussion. Similar in multiple other places in this section.
Lines 620-622: What does it mean that “HoloTrack would have sampled”? Is there in-cloud data or not? If so, please show it.
Citation: https://doi.org/10.5194/egusphere-2025-1774-RC2 -
AC2: 'Reply on RC2', Birte Thiede, 29 Aug 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-1774/egusphere-2025-1774-AC2-supplement.pdf
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2025-1774', Anonymous Referee #1, 04 Jul 2025
Review of "HoloTrack: In-Situ Holographic Particle Tracking of Cloud Droplets"
Overview of the Paper: This manuscript presents HoloTrack, a novel, fully autonomous instrument for the in-situ measurement of cloud microphysical properties. The primary innovation of HoloTrack is its ability to perform three-dimensional particle tracking by capturing pairs of holograms at a high frequency (25 pairs/sec). This allows for the direct measurement of individual cloud droplet velocities in addition to their 3D position, size, and shape. The authors provide a comprehensive description of the instrument's mechanical, optical, and electronic design, as well as its automation systems. The paper's strength lies in its thorough performance evaluation, which includes a maiden test flight on the Max Planck CloudKite (MPCK) platform, static tests using a calibrated "CloudTarget," and a series of detailed wind tunnel experiments. These evaluations quantify the instrument's detection efficiency, velocity measurement accuracy and uncertainty, and the aerodynamic influence of the instrument's body on the sample volume under various yaw angles. The work represents a significant technical achievement and provides a powerful new tool for advancing the experimental understanding of cloud microphysics, turbulence, and droplet dynamics.
General Recommendation: The paper is well-written, the instrument is thoughtfully designed, and the performance evaluation is extensive and convincing. It is a substantial contribution to the field of atmospheric measurement technology. The conclusions are well-supported by the presented data. The manuscript is nearly ready for publication.
I recommend this paper for publication after minor revisions. The revisions suggested below are intended to clarify a few points, which will enhance the paper's impact and utility for future users of this technology.
Comments:
- Table 1 (page 10): Two instruments are listed under the heading "Planned but not operational." To enhance clarity, consider naming the subsection: "Future Instrumentation"—to clearly indicate that they were not part of the current test configuration.
- Line 111: For unambiguous interpretation, please state explicitly which side of the windows the measurements refer to (e.g., "interior-facing" or "exterior-facing").
- Line 165: For clarity, consider labeling the holograms as H₁ and H₂ instead of A and B. The use of A/B within the sentence structure may cause confusion.
- Line 170: Please clarify why the second exposure (B) must be exactly 14 ms. Is it not possible to wait 13.9 ms and then use an exposure time of 0.1 ms instead?
- Figure 6: For consistency and easier referencing, use (a), (b), (c) to label the panels.
- Line 275: Please specify the distance between the instrument and the balloon. Additionally, address whether the balloon has any potential influence on the measurements (e.g., wake effects, thermal interference, shadowing)
- Line 298: Typographical correction: replace “near 0°” with the correct form “near 0°”
- Line 316: 1D Pitot Tube Filtering: The text notes that an "8-point-filtering was still set" on the 1D pitot tube, which smoothed the data. To aid reader understanding, please clarify why this filtering was active during the test flight—was it due to a default configuration, an oversight, or intentional for noise reduction?
Citation: https://doi.org/10.5194/egusphere-2025-1774-RC1 -
AC1: 'Reply on RC1', Birte Thiede, 29 Aug 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-1774/egusphere-2025-1774-AC1-supplement.pdf
-
RC2: 'Comment on egusphere-2025-1774', Anonymous Referee #2, 09 Jul 2025
This paper thoroughly and rigorously describes an instrument for tracking of cloud droplets measured by a holographic imaging system. The paper should be accepted after addressing the following points:
Major comments:
- The paper is very detailed, even tedious in places. Any streamlining would be appreciated, especially to reduce repetition. In particular, much of the material in the “Discussion” section is not really discussion, but simply repeating what has already been stated clearly in the main text of the paper. The Discussion section should be rewritten to focus on putting results into the bigger scientific context and describing plans for future improvements or applications.
- The big picture is missing, especially for those who may not be familiar with the concept of PIV or particle tracking. In the Introduction the measurement approach should be more clearly explained: two holograms recorded in rapid succession, such that they have significant spatial overlap, and then matching particles observed in both sub-volumes in order to obtain a velocity vector. One part of that vector is due to the mean advection of the flow, the other part is due to turbulence. Perhaps a figure illustrating this, the way you would explain in a presentation to a broad audience, could be useful. On this point: is this really “tracking” or is it velocimetry? With only two frames, it seems like a stretch to refer to this as particle tracking. But I agree it’s also not traditional PIV based on image correlation. Is there standard terminology in the field? If so, it’s best to introduce it and clearly define it.
- In the end it’s not clear whether holograms were obtained from within natural clouds. On line 268 it is stated that in-situ holograms could not be evaluated, but the reason is not clearly stated. If no in situ data are available, it should be stated directly that the tests of the holographic tracking are under laboratory conditions only. This will ensure the scope of the paper is accurately stated.
Detailed comments:
Lines 8-9: It is not clear in the abstract where these volumes come from and what they mean. If you wish to keep the numbers, please explain why the volumes are different, i.e., that the velocimetry depends on two samples separated by very short time, and therefore overlapping in space.Line 19: It’s an oversimplification to state that cloud properties are determined by the microphysics of clouds. One could just as easily claim that microphysics and other cloud properties are determined by cloud dynamics. The truth is that microphysics both follow and help determine other large-scale cloud properties. Please rephrase.
Line 29: It seems strange to cite a review paper on mixed-phase clouds as a reference for holographic cloud imaging. Perhaps there is a more appropriate one?
Line 43: The sentence would read better with a comma after “cloud droplet measurement”.
Last two paragraphs of Section 1: I was confused until the very end, and in fact I thought the paper was about a 2D PIV system. It takes too long to figure out what the paper is about, and even in the last paragraph it’s not really clear how the system works (see related point in “Major Comments” above). For example, “first airborne instrument capable of capturing hologram pair tracking of droplets” is quite obtuse. Please restructure the Introduction to clearly state the problem and to clearly outline the overall approach so readers can have an idea of what the paper is about.
Line 78: Delete “at most”.
Caption for Figure 1: I don’t know what OPC-N3 is referring to… need to define.
Line 111: The concept of reconsruction has not been introduced, so please do that first.
Line 112: Is 8-bit resolution sufficient to resolve the fringe visibility needed for high spatial localization of particles and particle size? See, e.g., discussion of fringe visibility in Fugal et al. 2004 (Applied Optics).
Line 126: State “Section 3.2”.
Line 130: Show numbers to convince that this limit is not typically reached. I believe that it is reached in polluted clouds.
Line 132: Three uses of “sample” “sampled” and “sampling” in one sentence… please rephrase to make it clearer and less repetitive.
Line 139: I don’t understand where the 45 cm constraint comes from… please explain more clearly.
Figure 2 caption: Meaning of “Thorlabs systems” is unclear. Also, meaning of “this simulation code” is unclear.
Line 182: “Despite the simplicity of the described timing.”
Section 2.2.2: It’s not clear why the laser is run at such a high repetition rate. Why not simply fire two pulses with the required delay? Presumably this must not be possible, but the reason is unclear.
Line 194: It’s unclear what simpleRTK2B and U-Blox ZED-F9P refers to.
Line 244: “36 cm to the sample volume and are not look directly into…” is unclear, needs to be reworded.
Line 251: Provide a date and location for the IMPACT campaign.
Line 298: “degree” symbol is not properly coded in latex.
Line 326: “Caused likely with 8-point-averaging filtering for 1D pitot tube” is unclear and grammatically incorrect… need to improve.
Lines 330-331: Units for energy dissipation rate are incorrect.
Lines 335-338: The Stokes number is only one parameter that governs decoupling from the flow. The settling parameter also needs to be considered. I expect that for the larger drop diameter of 50 um the gravitational setting is an important contributor. Please discuss and evaluate.
Line 350: “Recall” is not yet defined, so the sentence is not clear.
Lines 361-364: I don’t understand this, e.g., what is meant by “accurate measurement of precision”. Accuracy and precision have two different meanings.
Figure 7: I don’t recall that d_{gt} has been defined. What does the subscript indicate?
Lines 371-372: Is this a reasonable assumption? Why is it not possible to perform the analysis on an actual hologram pair?
Lines 373-374: s_m is used for both quantities… is this correct?
Line 401: “weighted by the square root”… square root of what? And why?
Figure 9: The “small dark blue square” is difficult to find in the image. Please enhance and/or enlarge to make it more evident.
Line 419: Do you mean “more turbulent data”?
Line 426: Define PMR here so it is clearer what it indicates in Equation 7.
Lines 432-440: I am completely lost. This discussion is quite tedious. Can it be simplified?
Line 447: I don’t see that “both cases” are defined.
Line 449: Is it really constant? It seems to me that it shows a trend.
Line 458: Define turbulence intensity.
Line 471: “Turbulence that exceeds the random error in inter-particle distances” is illogical. Please correct.
Figure 12: Are there results related to u versus w, which could hint at a sedimentation effect? It would be nice to include, in order to have a little bit of science in the article.
Line 484: “by the boundary layer”
Line 485: “expelled from the arms wake accumulate” is not clear… do you mean “from the wake of the arms”?
Line 488: “regions are reach up to” should be “regions reach up to”?
Lines 519-530: As noted above, this is all a repeat of what is in the article… not a Discussion. Similar in multiple other places in this section.
Lines 620-622: What does it mean that “HoloTrack would have sampled”? Is there in-cloud data or not? If so, please show it.
Citation: https://doi.org/10.5194/egusphere-2025-1774-RC2 -
AC2: 'Reply on RC2', Birte Thiede, 29 Aug 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-1774/egusphere-2025-1774-AC2-supplement.pdf
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 719 | 126 | 20 | 865 | 13 | 28 |
- HTML: 719
- PDF: 126
- XML: 20
- Total: 865
- BibTeX: 13
- EndNote: 28
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Birte Thiede
Freja Nordsiek
Yewon Kim
Eberhard Bodenschatz
Gholamhossein Bagheri
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(45038 KB) - Metadata XML
Review of "HoloTrack: In-Situ Holographic Particle Tracking of Cloud Droplets"
Overview of the Paper: This manuscript presents HoloTrack, a novel, fully autonomous instrument for the in-situ measurement of cloud microphysical properties. The primary innovation of HoloTrack is its ability to perform three-dimensional particle tracking by capturing pairs of holograms at a high frequency (25 pairs/sec). This allows for the direct measurement of individual cloud droplet velocities in addition to their 3D position, size, and shape. The authors provide a comprehensive description of the instrument's mechanical, optical, and electronic design, as well as its automation systems. The paper's strength lies in its thorough performance evaluation, which includes a maiden test flight on the Max Planck CloudKite (MPCK) platform, static tests using a calibrated "CloudTarget," and a series of detailed wind tunnel experiments. These evaluations quantify the instrument's detection efficiency, velocity measurement accuracy and uncertainty, and the aerodynamic influence of the instrument's body on the sample volume under various yaw angles. The work represents a significant technical achievement and provides a powerful new tool for advancing the experimental understanding of cloud microphysics, turbulence, and droplet dynamics.
General Recommendation: The paper is well-written, the instrument is thoughtfully designed, and the performance evaluation is extensive and convincing. It is a substantial contribution to the field of atmospheric measurement technology. The conclusions are well-supported by the presented data. The manuscript is nearly ready for publication.
I recommend this paper for publication after minor revisions. The revisions suggested below are intended to clarify a few points, which will enhance the paper's impact and utility for future users of this technology.
Comments: