the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 04 Aug 2026
Raman source using a hollow-core fiber for ozone monitoring in the lower troposphere
Abstract. The DIfferential Absorption microlidar for Boundary Layer Ozone (DIABLO) is designed for boundary-layer profiling. This instrument uses a diode-pumped, passively Q-switched Nd:YAG laser at 266 nm, coupled to a deuterium-filled hollow-core fiber (HCF) to generate the 289 nm Stokes wavelength via stimulated Raman scattering (SRS). The high Raman gain in the gas-filled HCF enables efficient UV conversion with significantly lower pump energy than required in traditional free-space Raman cells. At the output of the fiber, 1-ns pulses of 10 μJ at 266 nm and 7 μJ at 289 nm are obtained with a repetition rate of 1 kHz. First atmospheric measurements in central Paris provide ozone concentrations between 400 m and 1700 m. In contrast with existing UV DIAL ozone systems, which are typically bulky, complex, and operated in campaign-based research configurations, DIABLO targets a compact and low-energy architecture optimized for autonomous boundary-layer monitoring.
- Preprint
(9695 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2026-2631', Anonymous Referee #1, 10 Aug 2026
-
RC2: 'Comment on egusphere-2026-2631', Anonymous Referee #2, 01 Sep 2026
General comments
The manuscript "Raman source using a hollow-core fiber for ozone monitoring in the lower troposphere" presents an innovative architecture for miniaturizing UV DIAL systems. The integration of a deuterium-filled hollow-core fiber (HCF) to generate the 289 nm Stokes wavelength via stimulated Raman scattering (SRS) is a highly compelling approach to reducing the size and pump-energy requirements of traditional free-space Raman cells. The initial atmospheric profiles demonstrating measurements between 400 m and 1700 m in an urban boundary layer show the technique's fundamental viability. However, there is a significant gap between the successful demonstration of a novel optical architecture and the paper's stated goal of creating an instrument optimized for "autonomous boundary-layer monitoring" and "routine operation". A true operational network asset requires robust, uninterrupted 24/7 profiling. The current manuscript lacks critical laboratory characterization regarding the long-term stability of the transmitter, specifically concerning potential fiber degradation and optomechanical drift. Addressing these stability concerns, alongside providing independent vertical validation, is necessary before the system can be evaluated as a candidate for autonomous network deployment.
Specific comments
State of the art
Regarding the characterization of existing systems, the authors state in the abstract that current UV DIAL ozone systems are typically "operated in campaign-based research configurations". This generalization should be revised, as it underrepresents the current operational capabilities of the community. For example, instruments within the Tropospheric Ozone Lidar Network (TOLNet), such as the Small Mobile Ozone Lidar (SMOL) platform, routinely execute autonomous, continuous 24/7 vertical profiling over multi-month deployments with minor servicing needed every several weeks rather than being limited to short-term research campaigns. Acknowledging this existing transition toward operational network architectures will provide a more accurate baseline for comparison.
Transmitter Stability
The most critical omission in the manuscript is the absence of a long-term stability test of the transmitter under continuous operation. It remains unclear whether the 289 nm output energy remains stable at 7 µJ after 24 or 48 hours of continuous pulsing, or if it degrades rapidly. The authors must include a continuous laboratory stability plot tracking the output pulse energy of both the 266 nm residual pump and the 289 nm Stokes beam over at least a 24- to 72-hour period.
Optomechanical Drift and Coupling Stability
The authors note that coupling the 266 nm beam into the HCF is challenging due to a low numerical aperture (NA ~0.02) and tight angular acceptance, making the system "highly sensitive to alignment conditions". In an autonomous deployment, the system will be subjected to significant diurnal temperature swings and vibration during transportation and operations. Thermal expansion typically causes sub-micron drifts in optomechanical mounts, which could severely misalign the fiber coupling and destroy the Raman conversion efficiency. The authors should discuss how they expect to manage this issue.
Near-Field Blind Spot and Overlap Function
Although the instrument is designed for boundary-layer profiling, the lowest reported measurement altitude is 400 m. For urban air quality applications, capturing the lowest 400 m is arguably the most critical requirement for connecting boundary-layer dynamics to surface exposure. The authors state that this lower limit is constrained by incomplete overlap. However, given that the receiver has a relatively large full-angle field of view of 5 mrad and the transmitted beam divergence is estimated below 1 mrad, achieving full geometrical overlap should theoretically not be difficult at much shorter ranges. The authors should address whether there is an unusually large baseline separation between the transmitter and receiver causing this delayed overlap. If not, they should clarify if detector saturation or pulse pile-up at 1 kHz is the actual limiting factor and clearly outline the necessary receiver modifications required to profile this near-surface layer.
Validation with Ozonesondes
While the comparison with surface network data (such as the Paris 13ème and Tour Eiffel stations) and the CAMS model provides useful context, the manuscript lacks independent physical verification of the vertical structure. To properly validate the lidar's vertical profiling capabilities and confirm the observed boundary-layer ozone gradients, it would be highly beneficial to incorporate a couple of coincident ozonesonde launches during the measurement periods.
Gas Permeation Mechanism
The manuscript notes a deuterium pressure drop over a scale of days, which is currently attributed to permeation through the gas connections. While this pressure drop is not a primary concern for the overall viability of the technique, the authors should explore the physical mechanism more thoroughly and explain if thermal expansion and contraction as expected during typical field deployments and transportation could pose additional challenges.
Editorial comments
Section 1, Line 46: "...Network for the Detection for Atmospheric Composition Change..." – The correct name for NDACC is the Network for the Detection of Atmospheric Composition Change.
Section 2.2, Line 127: "...separed by 100 cm..." – Should be "separated".
Section 5, Line 271: "...(Chouza et al., 2025) Such developments..." – There is a missing period after the citation.
Citation: https://doi.org/10.5194/egusphere-2026-2631-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 126 | 52 | 16 | 194 | 13 | 9 |
- HTML: 126
- PDF: 52
- XML: 16
- Total: 194
- BibTeX: 13
- EndNote: 9
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The manuscript presents a new application of an O3 DIAL system based on SRS shifting of a 266 nm laser eam at 289 nm based on a HCF fiber filled with D2.
The measurements presented suffer from several drawbacks.
1) No systematic approach is provided to calculate (estimate through simulations) the systematic errors on O3 DIAL measurements (from aerosols, SO2 and NO2 concentrations) in the PBL as stated by numerous papers in the 80's and 90's. To this end a new section is required (see my specific comments in the manuscript). 2) No statisitical and systematic errors are provided in the retrieved O3 vertical profiles (cf. Fig. 6) taking into account NO2 and aerosol data profiles obtained from ceilometer and even by CAMS model. The reference to ceilometer profile data is not scientifically supported as no data are provided. 3) The gluing technique for the analog and photoc counting signals has to be presented. 4) The CAMS model is used as a "reference" without its inherent uncertainties discussed. 5) Several parts of the manuscript need further analysis and discussion and support based on published papers, as explicitely explained in the annotated manuscrpt. 6) The Conclusions part needs inprovement based on comments provided in the atached annotated manuscript.