the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 13 Dec 2023
Aerosol optical properties within the atmospheric boundary layer predicted from ground-based observations compared to Raman lidar retrievals during RITA-2021
Abstract. In this study, a Mie theory-based model was built to predict the vertical profile of the aerosol optical properties, including the aerosol scattering coefficient, backscatter coefficient, extinction coefficient, and lidar ratio. The model utilizes ground-based in-situ measurements of the aerosol chemical composition and particle size distribution, as well as the meteorological data from the Weather Forecasts (ECMWF) as input values. These are all parameters readily obtained for ACTRIS sites and the aim of this study was to investigate their suitability for generating representative estimates of the lidar ratio, and then further improve the lidar retrievals by utilizing these estimates. The measurements were performed during the Ruisdael land-atmosphere interactions Intensive Trace-gas and Aerosol (RITA) campaign in the Netherlands in 2021. The calculated dry aerosol optical properties were validated against a Nephelometer with good agreements (R2 ≈ 0.9). The predicted ambient vertical profiles of aerosol optical properties were compared to retrievals by a multi-wavelength Raman lidar. Predicted and retrieved backscatter coefficients were usually comparable under conditions of a well-mixed boundary layer. The extinction coefficients and lidar ratios were retrieved by the Raman lidar only at a height above 800 m. The estimated lidar ratio profiles based on in-situ data connected reasonably well to the lidar profiles within the boundary layer, with differences on average ± 30 %. Our study shows that for well-mixed boundary layers, a representative lidar ratio can be estimated based on ground-based in-situ measurements of dry size distribution and chemical composition taking into account the hygroscopic growth and ambient humidity. This allows to extend extinction profiles to lower altitudes, where they cannot be retrieved, or for use with simple elastic backscatter lidar to derive extinction profiles. The proposed method could be further applied to predict aerosol optical depth and also might be beneficial for large-scale or global radiation simulations.
- Preprint
(1741 KB) - Metadata XML
-
Supplement
(4829 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Review of egusphere-2023-2262', Anonymous Referee #2, 05 Jan 2024
The authors present a comparison of backscatter and extinction coefficients as inferred from ground-based in-situ measurements and lidar observations. The manuscripts describes the steps taken to convert in-situ measurements to optical properties and shows findings for a five-month measurement campaign in 2021. Given these specifics, the manuscript would potentially qualify as a Measurement Report. However, I don't find the work to be within the scope of ACP and there are several issues that lead me to recommend rejection:
- The authors fail to motivate why the scientific community should be interested in this work. Will the method be applied in future analyses of CAELI observations? Can it be adapted to other sites? How is it superior to the traditional approach of just assuming a lidar ratio?
- The presentation is rather unfocussed with an additional 42 figures in the supplement. The authors also mention instruments like the microwave radiometer and the ceilometer that are not really used later in the work. I suggest to identify key messages and trim the presentation accordingly.
- The presented measurements seem to be quite specific focussing almost exclusively on clean conditions. It would have been nice if there had been an effort to put the aerosol conditions during RITA into a long-term perspective, e.g., using long-term sun-photometer measurements.
- The authors admit that coarse-mode aerosols have a large impact on the scattering calculations. While this is somewhat minimised by the clean conditions considered in their work, it is likely to be of huge importance during other conditions. In that context, it would have been nice to get some long-term perspective on the occurrence of coarse-mode aerosols. The authors should also mention that Mie theory is inadequate to infer optical properties of dust particles.
- The optical profiles inferred from the ground-based in-situ measurements all look like scaled versions of the RH profile. This is not surprising as only RH might give some insight on vertical variation and the authors assume aerosol conditions to be constant with height. It doesn't seem fair that any discrepancies between measured and modelled optical properties are then attributed to errors in the RH profile. The authors should rather find a way to identify a maximum height up to which the vertical extension of ground-based in-situ measurements can give meaningful results. High-resolution sounding profiles could be a source for such an assessment and I wouldn't assume any connection between the ground and above the first inversion - particularly later in the year.
- The authors overstate their findings. They state "a representative lidar ratio can be estimated based on ground-based in-situ measurements". However, the presented results give the impression that they are by no way superior to an analysis by an average lidar operator. The suitability furthermore hinges on the assumption of vertical homogeneity which - though not unreasonable - should still be supported by some form of measurement. They continue "This allows to extend extinction profiles to lower altitudes, where they cannot be retrieved, or for use with simple elastic backscatter lidar to derive extinction profiles." Again, lidar operators have been doing pretty well with assuming lidar ratios based on experience. The authors would need to be more specific to support their statement. And conclude "The proposed method could be further applied to predict aerosol optical depth and also might be beneficial for large-scale or global radiation simulations." It is quite customary to simply assume constant lidar extinction coefficients from the lowermost trustworthy measurement height to the surface. This approach generally shows good agreement to Sun-photometer observations of aerosol optical depth. This approach also assumes vertically homogeneous aerosol conditions but is much more straightforward than the authors' work.
- The choice of references regarding the lidar technique in general and lidar ratios in particular is quite unusual. I suggest to consult with the corresponding co-authors to find more suitable references.
Citation: https://doi.org/10.5194/egusphere-2023-2262-RC1 -
RC2: 'Comment on egusphere-2023-2262', Anonymous Referee #1, 22 Jan 2024
General remarks: The manuscript discusses an interesting approach and brings together in situ observations of microphysical aerosol properties and chemical composition at ground (at Cabauw in The Netherlands) and optical modeling and comparison of the modeling results with lidar profiles of measured aerosol optical properties. There are many examples of such so-called closure experiments in the literature (since about 25 years), however, such exercises are still needed and thus the manuscript is a good addition to the literature in this field of optical closure studies.
In contrast to the other reviewer, I do not think that the paper should be rejected. It shows the present state of the art when combining ACTRIS observations from super sites… equipped with (a) aerosol monitoring tools and (b) remote sensing instruments. I also do not agree that this manuscript is a measurement report. Closure studies as presented here are more than just observations.
My main point of criticism is the following one: A lot of essential information is given in the rather extended supplement. That means one has to be very ‘active’ as reader and switch from main text to supplementary material and back and so one. A fluent reading is not possible. In the detail section, I will provide a few suggestions how this can be overcome, at least a bit.
Details:
The Abstract needs to be adjusted after finalization of the revision.
p2, l53: Cooney et al. and Melfi references are not appropriate here, in the context of aerosol extinction retrieval… Cooney and Melfi are pioneers in the field of Raman lidar developments because they introduced the temperature and water vapor Raman lidar technique.
p3, l81: be more specific already here, mention time periods.
p3, l84: I would prefer to include Figure S1 in the main text, and even Figure S2.
p5, l144: Avoid confusion (with lidar backscatter at 180°) already in the beginning, mention the angle range directly after … backscatter coefficient (7° to 170°).
p6, l174: When having a near range telescope you should be able to show extinction and lidar ratio values even down to 500 m height (after overlap corrections). And for heights above about 1000 m, you should be able to use the far range observations (after overlap correction) and then we would have much better, less noisy lidar ratios between 1000 and 2500 m height. So, why are these data not included? ….. should be stated…. I would recommend to use your own Raman analysis method instead of using the automated SSA software, and in this way, to optimize the lidar products in these optical closure studies.
Another question: What about the Raman lidar observations of the water vapor mixing ratio. In combination with ECMWF temperatures (usually very accurate) one could present them in the panels with ECMWF T and RH profiles. Even during daylight conditions, I could imagine that signals are good enough to show water vapor data up to 1000 m height.
The ECMWF RH profiles are rather uncertain (as usual for modelled water vapor profiles), so one needs more information about the ‘real world’ RH conditions. I would appreciate, if one shows radiosonde profiles in the respective panels in Figures 5,6,7,8, and if possible the Raman lidar RH profiles. The humidity has such a large and critical impact on the modelled optical properties, one needs to show better RH values, even if ECMWF RH values are considered in all the modelling, the reader should know about the quality (uncertainty) of these ECMWF RH profiles.
Figure 1 is certainly confusing for non-lidar scientists, especially regarding all the vertical white lines up to 5 km height. Is that just noise or is that strong backscatter from clouds…? Furthermore, what do you mean: an overview is given in Figure 1…., when nothing is explained? What is then the message to the reader? The Raman lidar observations need to be better indicated by thicker lines and brighter color, maybe yellow or orange.
p9, l232: I would include Table S3 in the main manuscript.
p9, l243: One should better highlight and explain, how the vertical profile is obtained…. Maybe, one should have a subsection (on vertical aerosol profile) , and show a sketch…, showing T and RH profiles, a well-mixed PBL, and maybe even T and RH profiles for a well-mixed layer, i.e. pot temp = const, RH increasing according to mix ratio = const. In addition, the optical properties as modelled at the surface (indicated by a big symbol) should be shown and finally the aerosol extinction profile, that is in agreement with the RH height profile structures.
Such a sketch would support the reader to understand the closure results…. in Figs. 5-8.
Results and discussion:
I would prefer to start with Figure S1 and S2 in the main text! Four case studies are then discussed. To provide all necessary details (to the field site, trajectories etc…) in the main text, one probably has to reduce the number of case studies. In the case of Figure 5, I would prefer to see in addition Fig. S9 (showing the full advantage of a lidar, clearly indicating many different aerosol layers, rather than any well-mixed layer), Fig. S10, providing information about the chemical composition, and Fig. S11, showing the origin of the pollution. However, we need at least different trajectories at 250 m (representative for surface aerosol conditions), 900 m, and also one for the 1200-2500m aerosol layer.
In this way (Fig. 5, S9, S10, S11), we would have a complete story and could much better discuss the results of the closure study, and why there is disagreement, especially for heights above 1200m.
I also believe that a full set of observed information (including a much better description of the humidity conditions and air mass transport at different heights) will allow a critical and much deeper debate on the applicability of the closure approach presented here and the especially concerning the limits of the approach.
And as mentioned, I would include a nighttime radiosonde RH profile (19 May, 23:30 UTC, Figure S6 shows it), and if Raman lidar mixing ratio data are available even Raman lidar based RH profiles.
Fig. 5: I do not see (a), (b), (c), (d), where did you put/place these letters? If there are only 355 nm extinction and lidar ratio profiles, then one should not show 532nm in the boxes (with line and symbol explanations), and these white boxes should not hide values. This holds for all other figures and panels as well.
The same for Figure 6, we need in addition, Fig S12, S13, and 14 (with three trajectories) in the main text. And on 9 Sep, it was probably dark over Cabauw at 21 UTC…. so please show Raman lidar RH profiles plus radiosonde RH profiles (9 Sep 23:30 UTC).
Now, we can discuss this closure study in very large detail, including the uncertainty in the model results caused by the ECMWF RH profile.
I would skip the Fig. 7 closure study. There is already the 19 May case, and the lidar ratio shows marine conditions.
Figure 8 is nice, could be presented with the figures S18-S20 here in the main manuscript, and again more trajectories for more heights (250 m, 800m, 1600 m) should be shown. Furthermore, Raman lidar and radiosonde water vapor profiles, if available.
Alternatively, one could try to combine Figs. 7 and 8 in ONE figure and show only the optical properties, and briefly discuss the results of these closure study.
Figure 9 shows just ONE 532 nm lidar ratio. I would remove this 532 nm value, so that only measured 355 nm lidar ratios are considered in Fig 9a and 9b.
The supplementary material is too much, no reader (except the reviewers) will study all details so one should reduce the amount of figures and plots to an absolute minimum.
Citation: https://doi.org/10.5194/egusphere-2023-2262-RC2
Data sets
Datasets for " Evaluation of the TOF-ACSM-CV for PM1.0 and PM2.5 measurements during the RITA-2021 field campaign" Xinya Liu, Bas Henzing, Arjan Hensen, Danielle van Dinther, and Ulrike Dusek https://doi.org/10.5281/zenodo.7924288
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 249 | 92 | 21 | 362 | 34 | 13 | 12 |
- HTML: 249
- PDF: 92
- XML: 21
- Total: 362
- Supplement: 34
- BibTeX: 13
- EndNote: 12
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1