the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 19 Aug 2026
The Heterodyne Detection Wind lidar Gadget HEDWIG for the German research aircraft HALO: Instrument design, wind retrieval and first airborne observations
Abstract. The Heterodyne Detection Wind lidar Gadget (HEDWIG) is a new airborne Doppler wind lidar developed at the German Aerospace Center (DLR) for high-resolution observations of atmospheric wind profiles. The instrument operates at an eye-safe wavelength of 1617 nm. A newly developed double-wedge scanner provides rapid and flexible beam steering, enabling both conical scanning for horizontal wind retrieval and parallel nadir-pointing observations for direct measurements of the vertical wind component. The complete instrument, including laser transmitter, receiver, scanner, data acquisition system and wind retrieval chain, is described in detail. Particular emphasis is placed on the optical and mechanical design of the scanner, the synchronized data acquisition, as well as the retrieval algorithms for line-of-sight (LOS) Doppler velocities and atmospheric wind vectors.
The instrument was successfully deployed for the first time aboard the German research aircraft HALO during the North Atlantic Waveguide, Dry Intrusion and Downstream Impact Campaign (NAWDIC) in winter 2026. Surface-return analyses demonstrate a systematic wind speed error below 0.1 m s-1 and a random error of about 0.2 m s-1 for single LOS measurements. First airborne observations further demonstrate the capability of HEDWIG to retrieve horizontal wind vectors together with high-resolution vertical wind profiles. The instrument therefore provides a unique observational capability for airborne studies of atmospheric dynamics as well as for satellite instrument validation as for instance the Doppler velocity product of the cloud profiling radar on EarthCARE, but also for future missions as WIVERN and Aeolus-2.
- Preprint
(2433 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2026-4731', Anonymous Referee #1, 14 Sep 2026
-
RC2: 'Comment on egusphere-2026-4731', Kristopher M. Bedka, 21 Sep 2026
Summary and Comment
This paper describes a new airborne Doppler wind lidar system developed at DLR called HEDWIG, retrieval algorithms applied to its data, and a brief demonstration of data from its first flight campaign NAWDIC. The system has high repetition rate and moderate low pulse energy, placing its TFOM in the middle of current airborne wind lidar system. HEDWIG features a novel Rissly prism scanning mechanism that enables a diversity of beam orientations for both vertical and horizontal wind profiling. HEDWIG performs 360° conical scanning with relatively short dwell times per azimuth for horizontal wind profiling, , with longer dwell times interspersed for direct measurement of vertical motions. The instrument installation and pointing is calibrated with land and ocean surface returns, demonstrating very low uncertainty. An example of 2 hours of flight data is shown to provide an initial demonstration of instrument performance.
The paper is extremely well written and concise. I have no critiques or comments about the writeup, which is very rare for me, so kudos to the authors for such a well-constructed and written paper. My only comment / concern is that I am not able to interpret the data shown in Figure 6 as well as I had hoped. One of the primary strengths of a heterodyne system is its ability to detect extremely high precision and resolution winds from boundary layer and clear sky aerosol returns. I look at the top three panels Figure 6 and I see what only looks to be to be returns from mid- to upper-tropospheric cloud. I see that the dwells for off-nadir LOS were just 1 second, providing 750 pulses for integration, which could be limited backscatter sensitivity. I think showing a satellite image and corresponding data from the WALES HSRL would be informative to characterize the scene being measured by HEDWIG. I look at panels d and e and I see quite strong SNR from the nadir measurements (which have longer dwells, better backscatter sensitivity and greater depth of successful retrieval) throughout the mid- to upper-troposphere. The character of the returns looks to me like there is abundant cloud cover. I also some indication of clouds atop the PBL from -20 to -28 longitude, which may have obscured the PBL. But all this is speculative and hard to interpret from the curtain plots. Addition of the MTG FCI imagery and WALES would allow us to see “the truth”, to see the depth of the PBL (if detectable at all given cloud cover), and assess the backscatter sensitivity off-nadir beams with short dwells. I understand that future algorithm work will merge data from successive scans which should fill in some gaps in the profiles. This will be critical for future applications that require a deep layer of wind measurement from cloud free backscatter signal.
I also see mentions of assessment of data quality with independent wind measurements. Were dropsondes being released from the HALO aircraft? This will enable us to get a picture of measurement precision after the surface return calibration. It would be informative to see at least one comparison against an independent wind measurement in this paper so we can get an initial perspective of the wind measurement precision, which will be later augmented with a larger sample of comparisons.Citation: https://doi.org/10.5194/egusphere-2026-4731-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 162 | 75 | 57 | 294 | 52 | 56 |
- HTML: 162
- PDF: 75
- XML: 57
- Total: 294
- BibTeX: 52
- EndNote: 56
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
General comments
The paper is well written and structured. It addresses scientific questions well within the scope of AMT. It presents a new airborne lidar tool, that will be of interest for the atmospheric community. The concept is not novel, for it relies on a long heritage of former developments by the same team, but the improvement is well positioned within the state-of-the art, as well as compared to the precedent DLR lidar version. The paper claims are clearly expressed, and are demonstrated throughout the paper with sufficient details. It would have been nice to have also a validation of the lidar measurements compared to independent observations, but it is clearly explained that this will be part of future investigations and reports.
I would recommend only minor revision before publication, considering the specific comments below.
Specific comments
In Table 1, the relationship between tau_p in nanoseconds and tau_pulse in meters (not introduced in the text) should be clarified.
In section 2 around figure 1, it could be relevant to mention the size and weight of the presented structure.
In Fig.2, the slave oscillator (SO) mentioned in caption should be shown in the drawing (or removed from caption). The label “housekeeping unit” is also unfamiliar (to me at least). I see the explanation is given in sect.2.3. Perhaps it could be given earlier as well.
Line 134: it is mentioned that a PBS is used for separating transmitted and received optical paths. I guess there is also a quarter-wave plate after emission (where exactly?) to rotate the polarization 90° between both paths?
Line 150: can you precise what is meant by “high dynamic performance”. How much time does it take for the scanner to go from conical-scan to nadir-pointing position, and how does it compare to the former 2 micron lidar version ?
Line 220: it is mentioned that the laser performance can sometimes deviate from nominal behavior: can these deviations be compensated automatically (maybe partly), or does it rely entirely on the supervision and caution of the lidar operator to adjust the laser parameters? The laser “bad shots” are mentioned also lines 247-250, and can be suppressed at the processing level. As a rough order of magnitude, what fraction of data does this represent?
Line 237-238 and 269-274: From my understanding, the aircraft motion is obtained in the Earth frame by the GNSS sensor, while the Doppler velocity nu_D is obtained in the aircraft frame. Therefore I would have expected the coordinate transformation to apply only on nu_D (nu_ac subtracted after transformation) and not on nu_LOS=nu_D-nu_ac (nu_ac subtracted before transformation) as written line 269. Perhaps I misunderstood something there?
Below figure 5: the random error is shown to be around 0.2 m/s over land surface. Could you add a few comments on this value? Is it limited by the transmitter FOM (hence a higher FOM would reduce the error), or by the signal processing / wind retrieval procedure cumulated uncertainties (hence improving the FOM would not change much the result)? Also, how does it compare to wind natural inhomogeneity at the scale of the instrument spatial resolution?
Technical corrections
line 36: the final s from “precises” should be taken off
line 48: remove space between “in” and “to”
line 89: probably a word is missing in “during the operating the airborne”
Figure 1: GNSS instead of GNNS (drawing)
line 211: typo where the number 1.5 is repeated two times