the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 09 Dec 2025
Light-weight Observatory for sOuNdIng clouds and aeorSol, LOONIS: a balloon lifted platform for troposphere aerosol research
Abstract. High-altitude aerosol research is crucial but faces significant cost and logistical hurdles that limit our ability to capture the highly variable vertical distribution of atmospheric trace substances. This paper introduces the Light-weight Observatory for sOuNdIng clouds and aerosol (LOONIS), a versatile, cost-effective, balloon-borne platform that provides an approach to address these challenges. LOONIS integrates a suite of lightweight instruments, such as optical particle counters, which provides real-time in situ detection of aerosol particle number and their microphysical properties, and impactors for collecting particles due to their inertia for subsequent offline physico-chemical analyses. Deployed during two measurement campaigns
in Germany during August 2023 and June 2024, LOONIS provided insights into vertical aerosol distribution, capturing aerosol activation processes within saturated atmospheric layers.
The platform demonstrated enhanced accuracy of particle concentration data from the UCASS instrument through the integration of a Thermal Flow Sensor (TFS). The deployment and resulting dataset underscore LOONIS’s capability as a tool for improving our understanding of atmospheric processes and potentially reducing the knowledge gap in atmospheric aerosol processes.
Competing interests: At least one of the (co-)authors is a member of the editorial board of Atmospheric Measurement Techniques.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.- Preprint
(10944 KB) - Metadata XML
-
Supplement
(1863 KB) - BibTeX
- EndNote
Status: open (until 24 Feb 2026)
- RC1: 'Comment on egusphere-2025-5568', Anonymous Referee #1, 09 Jan 2026 reply
Data sets
Light-weight Observatory for sOuNdIng clouds and aeorSol, 14th of June 2024. Luis Valero Tuya https://doi.org/10.5281/zenodo.17397010
Video supplement
IPAMZ81(14/06/2024) footage Konrad Kandler https://doi.org/10.5281/zenodo.17589746
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 220 | 138 | 10 | 368 | 44 | 60 | 60 |
- HTML: 220
- PDF: 138
- XML: 10
- Total: 368
- Supplement: 44
- BibTeX: 60
- EndNote: 60
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Valero et al. present in this manuscript a newly developed balloon-sounding system for atmospheric-state, aerosol, cloud-droplet, and trace-gas observations from the ground to 30 km altitude. A custom payload structure containing instruments, power supply, data acquisition, and telemetry is carried by single-use weather balloons with the potential to recover the payload after balloon collapse. The system was launched 28 times during two campaigns in Germany. A case study provides a single profile of observed parameters, focusing on observations through a low-level cloud below the first two kilometers. In summary, the presented work provides insights into possibilities for vertical atmospheric observations using a low-cost approach.
Although the approach is valuable, the presented work is not suitable for publication with AMT as it is. The manuscript is not in a mature state, as would have been appropriate for the journal. Most importantly, some sections are written in poor English, with incomplete or doubled sentences that make it difficult for the reader to grasp the essence of entire paragraphs. The analysis lacks coherence and appears inconsistent, with generally weak treatment of the data and several critical elements seemingly omitted. In general, the scientific scope of the presented study remains unclear, and many open questions remain for the reader. What are the targeted atmospheric conditions and the scientific question intended to tackle? How is the payload actually recovered? When it is modular, why only show one setup? Why focus only on 2 km coverage for one profile when 28 flights potentially reached up to 30 km in altitude? Was the development successful and feasibility proven or not?
The methods and conclusions drawn indicate that the authors may need to further develop their understanding of certain fundamental principles of aerosol science. Freely available literature and information on the use of the optical particle size spectrometer were likely not considered or evaluated in sufficient detail. For instance, arbitrary aerosol refractive indices were assumed (no explanation for their use was given), without accounting for corrections for the calibration materials used. No attempt was made to correctly quantify the measured variables or to evaluate their plausibility. The authors state that the presented data only serve as a qualitative description of atmospheric processes. Some sensors did not perform well because they were not suitable for the application, which is probably true of other sensors as well, but the authors did not consider this. Inlet sampling issues for the aerosol sensors were not sufficiently considered.
The authorship statement suggests that the corresponding author was primarily responsible for manuscript preparation, with limited apparent involvement in other aspects of the work. This may partially explain the current state of the manuscript, yet the presence of senior co-authors implies a shared responsibility to ensure that the paper meets an appropriate scientific and editorial standard. The challenges encountered in reviewing this submission, including the substantial time required to provide detailed and constructive comments, highlight how burdensome such poorly prepared manuscripts can be for reviewers. In its present form, the manuscript gives the impression that essential elements of PhD supervision and internal quality control were deferred to the external review process, which is not an appropriate use of peer review.
Formatting:
Introduce all instruments consistently: general instrument name (model, manufacturer, country) e.g. radiosonde (RS41-SGP, Vaisala, Finland)
Introduce all abbreviations: nitrogen (N), optical particle spectrometer (OPS), etc.
Introduction
The motivation should focus more on scientific needs and questions that the system can target than on available instrumentation. The logical flow seems interrupted by introducing instrument selection and the instrument's working principle, then returning to science, and then back to the instrument description.
Balloon:
Describe operation and ground preparation. How is the balloon recovered? Description of parachute etc.
Instruments:
The balloon reached an altitude of up to 30 km, which requires proof of instrument performance at these pressure levels.
POPS:
Why were these refractive indices chosen and not the PSL equivalent diameter taken? Add the inlet specification and estimate sampling efficiency
UCASS:
Show the calibration results for the UCASS (e.g. as histogram plot)
CO monitor:
Why use the same instrument for the entire campaign when it is not suitable for balloon applications? Simply delete everything about this failure from the document, as it is of no benefit to the reader
Detailed comments:
Line 79: cancel “that”
Line 80: “,” before b), exchange “.” with “,”
Line 86: radiosonde (RS41-SGP, Vaisala, Finland)
Line 87: add “measurements” after ozone, instrument naming see above
Line 89: attaches the rope to what? Or is the rope attached to the upper end of the rod?
Line 90: centered in relation to what and why? An imbalance in what and how is it mitigated with a simple rod? Is there a flexible weight on the rod?
Figure 1: Add a photograph of the system; the figure description is redundant to the text.
Table 2: “sampling time” is not useful for continuous measurements, consistent instrument naming (see above), add pump model and manufacturer
Line 92: The maximum weight of what?
Line 94: introduce all variables and units properly, e.g. aerosol particle number concentration (N in cm-3), carbon monoxide (CO in ppb) …
What about particle number size distribution?
Line 95: Are cloud particles also collected?
Line 97: cancel “platforms”
Line 101: an optical particle spectrometer is abbreviated OPS not OPC, the UCASS is used on UAS but it was designed for balloons
Line 103: “laser diode beam”
Line 104: the scattered light intensity is also correlated to the particle's chemical composition (refractive index) and shape
Line 105: What is calibrated with the pitot tube?
Line 108 to 116 should be at the beginning of the section or even in the introduction
Line 132: remain consistent with UCASS subscripts
Line 133: add the calibration results. What were the usual sizing deviations, and on what basis was it defined that it is a faulty unit or not?
Line 139: the POPS is an optical particle spectrometer, as it provides aerosol size information, and not only number
Line 142: Why was 1.45 +0.00i chosen? It does not seem representative of ambient atmospheric aerosol
Line 143: As PSL has a different refractive index from that chosen above, the diameter size bins provided by the manufacturer are not applicable
The sizing accuracy of the POPS is insufficient to support the use of 36 size bins, as shown by other studies, e.g. Pilz et al. (2022; https://doi.org/10.5194/amt-15-6889-2022 ).
Line 146: This size range refers to the refractive index of PSL and not to your chosen refractive indices. It has to be recalculated with Mie theory for your indices.
Line 152: Are the pumps capable of providing the critical pressure drop over the orifice and the volumetric flow at higher atmospheric levels? Please provide the pump specifications
Line 156 to 158: Please provide the equations or references for the impactor design. Provide a plot for the efficiency curves and specify what less steep curve means. The upper limit is probably more constrained by the sampling efficiency of the downward-facing inlets. Please discuss or provide calculations.
Line 195: Battery weight seems to be a constraint regarding scientific payload, so why not use fewer batteries when the entire battery capacity is not used?
Line 239: Introduce all chemical elements as in line 242
Line 268: Please specify what aerosol nucleation refers to, cloud droplet activation or secondary particle formation?
Line 275: The warm front appears to approach from the southwest and move to the northeast
Line 277: What is the meaning of the value (0.15 Jkg-1)?
Figure 3: Please reduce the figure size. The caption states a different CAPE of 1.6 Jkg-1 compared to the text. Please check
Lines 284 to 294 are redundant, as the same information is given above in Sec. 2.2.1. Please shorten.
Line 294: There seems to be an inconsistency between the flow rate comparison and the particle concentration deviations of the two methods. In Figure 4, the GPS ascent rates appear mainly higher than the TFS flow rate, with only a few outliers. Hence, the particle concentration based on the TFS should be mainly higher than that of the GPS. This seems true for UCASS A (red dots) in Fig. 6. Where are the – 30% deviation by the UCASS d from? The deviations seem to correlate well with the cloud layer between 900 and 800 hPa? Please elaborate!
Lines 295 to 309: It is obvious that UCASSa, with a detection size range of 0.3-32.5 µm, must provide higher particle number concentration outside of clouds than UCASSd, with a detection range of 9-51.5 µm. The effect of particle sampling losses due to impaction (which is textbook knowledge, e.g., Barron & Willeke) is negligible for comparing the two UCASS units with entirely different size ranges for measurements of the ambient aerosol size distribution. The number concentration of aerosol particles < 1µm is often orders of magnitude higher than above 9 µm.
Lines 320 to 325: Why does the overestimation of LWC from overrepresenting larger cloud particles align with the overcounting of smaller particles? Please elaborate
Line 344 to 345: To which particle size range do these concentrations refer?
Line 346: Which two instruments?
Line 353 to 356: In Figure 10, there is no comparison between the UCASS and the POPS possible for the overlap size range from 0.3 to 2.6 µm. Please add the entire size range of down to 0.3 µm for the UCASS a to Fig. 10 and discuss potential differences between the two instruments.
Figure 10: There is an inconsistency between the covered height range and the averaging time. With the balloon's average ascent rate of about 6 m/s, a 1-minute average should cover an altitude range of about 360 m, not 56 m, as the difference between 1761 and 1817 m. Please correct
Line 360: rephrase “aerosol-water activation” into “cloud droplet activation”. The POPS cannot differentiate between hygroscopic and hydrophilic particles.
Line 368: What is Jacobson’s modeled mode? It seems to be non-activated particles in a water-saturated environment. Provide more information
Figure 9 Caption: What means: “UCASSa, UCASSd (middle) have an optical particle diameter of 1.51µm and POPS (right) of 1.41µm.”?
Line 380: At a constant pump power consumption across declining ambient pressure during ascent, results probably in a continuously decreasing sample flow due to decreasing air density. Without monitoring the sample flow or at least the necessary critical pressure drop across the orifice, a constant sampling cannot be assumed. Hence, the aerosol population at lower altitude levels is potentially overrepresented in your impactor samples.
Line 385: It seems unlikely that the downward-facing inlets of the impactors collect cloud particles of 25 µm size when the balloon is ascending at 6 m/s, because of particle inertia. Please provide a calculation of the impactor inlet sampling efficiency.
Line 389 - 393: This is inaccurate, and the authors may improve foundational knowledge of aerosol science. A well-mixed atmospheric layer is usually defined by atmospheric stability and turbulence. The aerosol population in the boundary layer, on the other hand, is almost always a mixture of different aerosol sources (primary or secondary). Thus, heterogeneity in the aerosol population within a well-mixed atmospheric layer is expected. Aerosol particles of different origins coexist within an air mass and are not confined to single air parcels that coexist beside each other. The case does not point towards a complex atmospheric structure; in fact, the observed structure is rather common.
Line 396: The sentence is too strong as written. Ammonium sulfate is common in aged continental aerosol, but it is not universally present nor uniquely defining in every sampled air layer or region.
Figure 11: Decrease the size of the figure in height by plotting in two columns.
Line 403: Which Lagrangian back trajectory tool was used? Please provide references
Lines 409 to 417: The whole paragraph needs rephrasing as multiple sentences appear twice.
Line 419: introduce all abbreviations!
Line 418 to 426: Cancel the whole paragraph as it does not concern the presented platform but some sample treatment issues. Also, only a certain specific type of particles entering the impactor through diffusional processes (which are they???) is not in line with the fundamentals of aerosol physics.
Figure 12: Decrease the figure size.
Figure caption: “The presence of ammonium sulfate, a secondary aerosol species, indicates that the sampled air mass has undergone significant atmospheric aging.” is inaccurate. Ammonium sulfate can also exist in relatively fresh air masses. More indicators are necessary to derive atmospheric aging.
Figure 13: Combine with Figure 13.
Figure 14: Move to supplement. The trajectories do not add information. From the wind direction alone, it can be inferred that the boundary-layer aerosol at the sample location is influenced by different anthropogenic sources in the Rhine-Main region.
Line 436: Figure 15 is introduced but not discussed. Can be deleted
Line 451: Sentence appears twice
Line 476: No modularity was shown; only a single configuration was shown.
Line 479: How do you know that the cloud droplets were newly formed?
Line 481 - 483: This conclusion is not supported by the study. Only single samples of the aerosol population were shown, which are not representative of the population as a whole.
The particle origin was connected to a broad region, and the back trajectories did not add further information. Finally, it was shown that aerosol particles were activated into cloud particles, but no quantification was provided. In summary, there is not much added value.
Line 484: The process level view is not supported without a proper quantification of aerosol and cloud particle number concentration.
Line 487: Compared to tethered balloons, the ascent rate is quite high
Line 493: It was not mentioned before that the impactors remain open during descent.
Line 505: It does not seem cost-effective to single-use an instrument like the POPS, which costs about 20 k€, if the balloon gets lost.