the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 06 Aug 2026
The Ground-based Fog and Aerosol Spectrometer – Technical description and first applications
Abstract. High-precision measurements of cloud microphysical properties, such as particle size and phase, remain challenging. The most common method for in-situ single-particle measurements is via light scattering, where size is inferred from the scattered-light intensity using Mie-Lorenz theory. Here, we present the newly developed instrument, the Ground-based Fog and Aerosol Spectrometer (GFAS). Besides particle size and number concentration obtained from forward-scattering measurements, the GFAS also measures the change in polarization of backward scattered light, which allows for investigating particle morphology and composition. Within this work, we present a new methodology to calibrate detectors of light scattering spectrometers using a poly-disperse spray of pure water coupled with theoretical Mie-Lorenz modeling. Furthermore, we show the importance of proper calibration procedures to derive reasonable and interpretable data with respect to particle size and degree of linear polarization. In particular, we show that the new autocalibration method that uses a spray of pure water droplets leads to lower uncertainties compared to the standard glass bead calibration and is also usable for the polarized backward scattering detectors. Finally, we evaluate the GFAS using laboratory and field measurements. Under laboratory conditions, polarization measurements distinguish water droplets from nonspherical dust. Under ambient conditions, however, this distinction becomes considerably more challenging. For the in-situ observations, we find that cloud particle optical properties are distinctly different compared to those of pure water measured under laboratory conditions, indicating a potentially elevated refractive index in cloud droplets compared to pure water.
Competing interests: At least one of the (co-)authors is a member of the editorial board of Atmospheric Measurement Techniques. DB is a co-founder of DMT, which developed and commercially markets the instruments used in this study. DH is employed by DMT (now Envea). The remaining authors declare no competing interests.
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
(9560 KB) - Metadata XML
-
Supplement
(3252 KB) - BibTeX
- EndNote
Status: open (until 13 Nov 2026)
- RC1: 'Comment on egusphere-2026-3790', Anonymous Referee #1, 15 Oct 2026 reply
Interactive computing environment
Routines and codes for GFAS data processing and calibration L. Haberstock et al. https://github.com/SU-air/instrumentation-GFAS
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 268 | 125 | 51 | 444 | 83 | 44 | 48 |
- HTML: 268
- PDF: 125
- XML: 51
- Total: 444
- Supplement: 83
- BibTeX: 44
- EndNote: 48
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
I recommend rejection in its present form, unless the authors already have the additional measurements and modelling needed to address the concerns below. In that case, major revision could be appropriate. The main claim that water-droplet autocalibration provides lower uncertainty than calibration with reference glass beads is not sufficiently supported. The instrument response also needs further characterization before the polarization measurements can be interpreted in terms of particle properties. My main concerns are listed below in order of priority.
1. The discrepancy between measured and simulated bead responses remains unexplained.
The bead measurements should serve as the basis for calibrating the instrument response. Because the measured distributions are broader than the simulated distributions and the forward-detector calibration factors depend on bead size, the calibration should explicitly account for these size-dependent effects. Uncertainties in bead properties may contribute to the observed differences, but the measurements should be used to determine and validate the calibration rather than treated only as an independent test.
For the backward detectors, the authors report that coherent calibration factors could not be obtained because the measured intensity distributions for different bead sizes overlap strongly. However, Figs. 2b and 2c predict substantial size dependence of the backward-scattering cross sections, also for refractive indices representative of glass. This discrepancy needs to be investigated before concluding that beads are unsuitable for backward-channel calibration.
The authors should calculate the expected responses for the actual bead materials and size distributions, including an adequately modelled and experimentally validated instrument response, and compare these directly with the measurements.
2. The instrument response is not sufficiently characterized.
The optical system shown in Fig. 1 contains collection and collimating lenses, beam splitters, polarization filters, and a qualifier mask. Its response cannot simply be assumed to be uniform over the nominal collection solid angles. Particle position, local laser intensity, lens and polarizer transmission, optical aberrations, aperture clipping, and qualifier acceptance may affect the recorded signals.
Appendix C acknowledges several of these effects. Figure S2 also includes a simplified treatment of “laser ambiguity” using randomly selected factors between 0.5 and 1.0. However, the physical basis of this treatment is unclear. How does it represent the actual laser intensity profile and the trajectories of accepted particles? Is the same treatment applied when simulating the bead responses?
A useful approach would be to combine Mie calculations with an optical-engineering model in Zemax, FRED, or equivalent software. This would allow the detector response to be calculated as a function of particle size, refractive index, and position within the sampling volume. The predictions should then be tested against laboratory measurements.
3. The polarization calculation needs clarification.
The standard Mie polarization components refer to the local scattering plane, which changes with azimuth across the GFAS collection solid angle. The GFAS analyzers, however, have fixed orientations relative to the incident laser polarization.
The authors should explain how the incident polarization is resolved into the local scattering basis, how the scattered polarization is propagated through the collection optics, and how it is projected onto the fixed analyzer axes before integration. Integrating the two local Mie intensity components separately would generally not describe this measurement geometry.
The manuscript does not provide enough information to determine whether these transformations are included. This needs to be clarified before the calculated polarization ratios, the backward-channel calibration, and the interpretation of ambient measurements can be assessed.
4. The actual Mie-feature matching is not shown.
Figure 4b and Fig. S2 describe the procedure, but the authors do not show a measured water-droplet histogram together with the fitted model histogram. This is essential evidence for the proposed method.
The authors should show representative fits, identify the features that determine the calibration factor, and present residuals and the fitting objective as a function of that factor. They should also test whether the result is unique and how it depends on bin width, bin selection, spray size distribution, and the assumed illumination variability.
5. The authors have not demonstrated that autocalibration gives a more accurate result.
A smaller spread of fitted calibration factors does not necessarily mean that these factors are closer to the correct values. The fitting procedure may give a precise but biased result if the assumed instrument response is incomplete.
Both calibration methods should be tested against the same independent reference. For example, droplets with independently measured diameters could be used to compare the sizes obtained after bead calibration and after autocalibration. These validation measurements should be separate from those used to determine the calibration factors.
The conclusion that autocalibration reduces total calibration uncertainty is not established by the current analysis. Agreement between the two backward-channel methods cannot by itself demonstrate that either method produces more consistent factors, because both depend on the forward calibration and related theoretical assumptions. The uncertainty analysis in Appendix C should therefore be applied consistently to both methods, including instrument effects and uncertainties specific to the reference particles, and the methods should be assessed against independent validation data.
6. The interpretation of the field polarization measurements is premature.
The unresolved backward-channel bead responses, the unclear treatment of the polarization reference systems, and the lack of independent calibration validation leave substantial uncertainty in the interpretation of the polarization signals.
The authors suggest that differences between laboratory water droplets and ambient cloud droplets at SBO indicate a higher refractive index due to dissolved aerosol material. Although this is presented as a hypothesis, the instrument response needs to be understood before such differences can be assigned to particle properties.
The presence of dissolved material alone does not establish that its concentration is sufficient to explain the observations. During droplet growth by condensation, this material becomes increasingly diluted. The authors should quantify the refractive-index change required to reproduce the measured signals and assess whether the corresponding solute concentrations are realistic. They should also assess whether the differences can be explained by calibration or instrument-response effects.
These concerns affect the evidence supporting the main methodological conclusion and the interpretation of the field measurements. If the relevant data are available, the authors should have the opportunity to strengthen their analysis by providing the additional characterization and validation as outlined in my review, rather than changing the discussion alone.