the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 13 Jul 2026
Technical note: A decoupled in situ framework for aerosol multi-wavelength optical properties: application for lidar retrievals
Abstract. Accurate constraints on aerosol complex refractive index (CRI), single scattering albedo (SSA), and lidar ratio (LR) are essential for improving lidar-based aerosol retrievals and reducing radiative-effect uncertainties. However, consistent derivation from in situ observations remains challenging in complex mixed-aerosol environments, where size distribution, absorption, and humidity effects are strongly coupled. Here we present a decoupled inversion framework that integrates SMPS–OPC size distributions, AE33-derived absorption coefficients, and ambient relative humidity observations from the Vipava Valley (Slovenia) in April 2016 to derive dry-state homogeneous-equivalent CRI and RH-corrected lidar-relevant SSA and LR at 355, 532, and 1064 nm. The framework reconstructs a continuous particle size distribution through instrument-response-aware SMPS–OPC geometric alignment and retrieves the reference real refractive index. AE33-derived absorption constraints and a singly subtractive Kramers–Kronig (SSKK) relation are then used to derive a spectrally consistent dry state CRI, after which SSA and LR are recalculated through an RH-corrected wet-state forward optical calculation. The retrieved dry-state mean n values were 1.448, 1.441, and 1.437, and the mean k values were 0.0312, 0.0238, and 0.0192 at 355, 532, and 1064 nm, respectively. The RH-corrected mean SSA values were 0.866, 0.852, and 0.774, and the corresponding mean LR values were 86.4, 58.4, and 39.2 sr. Sensitivity tests showed that the AE33 effective multiple-scattering correction affects absorption-sensitive products, with Ceff = 5 used as a site- and period-specific baseline. The framework provides a practical pathway for linking dry in situ CRI retrievals with RH-corrected multi-wavelength lidar-relevant aerosol optical products.
- Preprint
(1627 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 25 Aug 2026)
- RC1: 'Comment on egusphere-2026-3471', Anonymous Referee #1, 06 Aug 2026 reply
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 126 | 39 | 11 | 176 | 14 | 15 |
- HTML: 126
- PDF: 39
- XML: 11
- Total: 176
- BibTeX: 14
- EndNote: 15
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The study "Technical note: A decoupled in situ framework for aerosol multi-wavelength optical properties: application for lidar retrievals" by Yao et al. proposes a three-stage inversion for deriving lidar-relevant optical properties from ground-based in situ data. Stage 1 merges SMPS and Grimm OPC size distributions using an acceptance-angle response model and retrieves the real refractive index at 655 nm from shape consistency in the 350-750 nm overlap. Stage 2 constrains the imaginary part with AE33 absorption and propagates n(λ) from the 655 nm anchor via a singly subtractive Kramers-Kronig relation. Stage 3 applies k-Köhler growth and Bruggeman mixing to obtain ambient SSA and lidar ratio at 355, 532 and 1064 nm. The method is demonstrated on measurements from a 17-day campaign data over Slovenia, yielding k_dry=.0312/0.0238/0.0192, SSA = 0.866/0.852/0.774 and LR = 86.4/58.4/39.2 sr.
The separation of the geometric (PSD/n) problem from the absorption-constrained problem, with the dry-to-ambient conversion kept explicit, is sensible and addresses a genuine gap - boundary-layer lidar ratios, especially at 1064 nm, are poorly constrained. The manuscript is well organized and authors have commendably acknowledged several limitations about several limitations.
My main concern is that the products have not yet been shown to be quantitatively reliable. The adopted AE33 correction factor Ceff = 5 strongly influences the reported SSA and LR values but is not independently constrained. Moreover, the reported ± values appear to represent campaign variability rather than total retrieval uncertainty, while Monte Carlo retrieval uncertainty and structural model sensitivities are not clearly separated. Several useful checks appear possible using the existing campaign data and retrieval code.
1. The manuscript does not establish an independent basis for selecting Ceff = 5 as the baseline value. Section 3.2 shows that increasing Ceff from 1.57 to 5 changes median SSA355 from approximately 0.684 to 0.884 and median LR355 from approximately 131.9 to 83.2 sr. The manuscript notes that LR355 at Ceff = 5 falls within a literature-guided range of approximately 60-85 sr, but later states that the literature values are used only as contextual information and not as calibration or optimization constraints. Please clearly state the decision rule used to select Ceff =5 . The preferred options would be:
If no independent absorption or scattering reference exists, this should be stated explicitly. At minimum, the abstract should report the headline results as conditional values.
2. Equation (3) applies a spectrally invariant Ceff to the complete AE33 absorption spectrum. However, all available paired reference values in Table 1 decrease with wavelength. For example, the Granada ambient values are approximately 4.72 and 3.90 at 450 and 808 nm, respectively, while the soot measurements also exhibit wavelength dependence. A constant may therefore modify the relative absorption constraints at 355, 532, and 1064 nm. This could affect the retrieved spectral ordering of k, the SSA spectrum, and the pronounced decrease of LR toward 1064 nm. Please add a sensitivity calculation using one or more physically plausible wavelength-dependent Ceff (λ) profiles based on the paired reference measurements in Table 1. The authors should report whether the following conclusions remain robust:
The prupose for this test is not to prescribe a universal Ceff(λ), but to determine whether the major spectral conclusions are sensitive to the assumption of spectrally uniform scaling.
3. The synthetic forward-inverse experiment provides a useful numerical closure test, but the manuscript correctly acknowledges that it is not independent validation because the synthetic truth and retrieval use the same physical model and assumptions. Given the lidar-oriented framing, the availability of collocated or nearby lidar and sun-photometer observations during 7-23 April 2016 should be stated explicitly. The manuscript cites previous lidar work in the Vipava Valley environment, but it is unclear whether vertically resolved observations were available during the campaign period. If temporally collocated lidar measurements exist, please compare the RH-corrected 532 nm extinction or LR with the lowest range bins for which overlap correction is reliable. Such a comparison should be restricted to periods when the lower boundary layer can reasonably be considered vertically well mixed. If sun-photometer observations are available, comparison with AOD would require a defensible vertical integration of the locally derived extinction coefficient, for example using a lidar extinction profile or independently determined boundary-layer height. If neither lidar nor sun-photometer observations are available, this should be stated clearly in Section 2.1, and the abstract and conclusions should more explicitly describe the results as internally consistent, site-specific, ambient-equivalent estimates rather than independently validated lidar products.
4. The reconstructed PSD underpins the retrieval of n655, extinction, backscatter, SSA, and LR. The coarse tail may exert disproportionate control over the 1064 nm backscatter and therefore the reported LR1064.The manuscript shows PM2.5 and PM10, but it does not explain how these quantities were measured or derived. Please identify the instruments or procedures used to obtain PM data. If PM was measured independently of the Grimm OPC, integration of the reconstructed PSD using plausible mode-dependent densities could provide a valuable mass-closure test. Such a comparison should account for:
If the PM concentrations were themselves derived from the Grimm OPC, they should not be described as an independent validation of the reconstructed PSD. Similarly, the eBC series in Figure 8 appears to be derived using the same AE33 observations and the adopted Ceff = 5. It therefore cannot independently constrain Ceff unless an external refractory-BC or absorption reference was used. Please explain how eBC was calculated, including the assumed MAC and its wavelength dependence. The inlet configuration, sampling-line length, flow conditions, and effective upper-size transmission should also be reported. Without this information, it is difficult to assess whether the reported PSD extending to 20 µm, particularly the portion above approximately 5-10 µm, is quantitatively representative. Additionally, the synthetic test indicates that the real refractive index is less robustly recovered than the imaginary part and remains sensitive to PSD remapping and OPC-response assumptions. Please provide an objective measure of SMPS-OPC overlap quality and examine the sensitivity of to the selected overlap interval and plausible relative SMPS-OPC concentration offsets. The authors should also state whether retrievals with poor overlap or weakly constrained solutions were rejected. This would help demonstrate that the retrieved , rather than only the smoothness of the merged PSD, is adequately constrained.
5. The reported mean LR1064 = 39.2sr depends on several weakly constrained elements: the coarse-mode PSD retrieved from a single-wavelength OPC through an assumed response model, extrapolation of the AE33 absorption fit beyond its longest measurement wavelength, SSKK propagation furthest from the 655 nm anchor, and spherical Mie calculations in an environment affected by episodic coarse-mode and dust enhancement. Please add PSD-truncation tests by recalculating the optical products with upper-diameter cut-offs of 2.5, 5, and 10 µm. Cumulative size-resolved contributions to extinction and backscatter at 355, 532, and 1064 nm would further clarify which particle sizes control the reported LR spectrum. Particle nonsphericity should also be discussed explicitly, because nonspherical coarse particles may produce substantially different backscatter from equivalent spheres and could contribute to uncertainty in the low reported LR1064. This is particularly relevant to the contrast with Haarig et al. (2018) discussed in Section 4.2. Finally, the coarse-influenced regime has a higher median LR355 than the fine-mode-dominated regime, 88.4 versus 79.8 sr. Please explain this behaviour, preferably by decomposing extinction and backscatter contributions from the fine and coarse modes.
6. No perturbation magnitudes, probability distributions, or correlation structures are provided for the 500-realisation Monte Carlo analysis (L167–169). Please report these assumptions for each instrument. The meaning of the reported ± values should also be clarified. For example, LR355 = 86.4 ± 12.7sr appears to represent the campaign mean ± standard deviation rather than total retrieval uncertainty. Please distinguish among campaign temporal variability, Monte Carlo retrieval uncertainty, and structural sensitivity, including Ceff, the OPC response assumption, prescribed hygroscopicity parameters, particle sphericity, and the homogeneous-equivalent assumption. A short table separating these terms would substantially improve the usability of the products. Please also clarify whether the identical ±0.075 values for ndry at all three wavelengths represent campaign variability, Monte Carlo uncertainty, propagation from the common 655 nm anchor, or rounding.
7. Lines 347-355 present a detailed diurnal source interpretation involving 6 a.m. traffic and charcoal combustion, industrial emissions at noon, and nitrate/sulfate dominance, but no composition data or supporting citations are provided. Please support these claims or reduce the passage to what the optical data show. If SEM or mineralogical analyses of campaign filter samples exist, they could substantiate the dust interpretation. For example, the early-morning increase in k may be described as evidence of enhanced short-wavelength absorption without assigning a unique emission source. Separately, Section 2.1 lists only the SMPS, OPC, and AE33, whereas the results also use PM2.5, PM10, and station RH. Please describe these additional measurements and explain the visible data gaps in Figure 6, around 13-15 and 17-19 April, with valid-sample counts.
8. Text says outputs at 355, 532, 655 and 1064 nm; results show three wavelengths only.
9. Eq. (1), Line 143 - ‘x’ is undefined (presumably size parameter); λ is defined but does not appear. Please define all symbols and state the normalisation of w_optics(θ).
10. Please state how many samples fall in the RH < 40 %, 40-95 % and >95 % bins, and confirm no artificial step appears near RH = 40 % in Fig. 12c.
11. Line379-380 - "agrees with the CALIPSO version 4": V4 values are prescribed lookup entries, not observations. The manuscript makes this point correctly at Line 447-448; please make the two passages consistent.
12. Submitted as a Technical Note but running to ~23 pages with 12 figures. Please either condense (moving the regime analysis to supplementary material) or reclassify as a research article and expand the validation.
13. A short paragraph on recommended use - which products serve as lidar priors, under what conditions, with what uncertainty would materially increase impact.