the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 20 Aug 2026
SEMPAS: A UV-visible Spectroscopy Instrument for Monitoring Shipping Emissions and Atmospheric Composition on the North Sea
Abstract. Ship emissions contribute significantly to global transport pollution, foremost NOx and SOx. To mitigate this issue, emission regulated areas have been established, such as in the North Sea. Monitoring systems are essential to ensure consistent compliance to the emission limits. The Ship Emission Monitoring with Passive Absorption Spectroscopy (SEMPAS) instrument has been developed to provide operational monitoring of individual ships; as they transit past the instrument's on-shore or off-shore installation location. SEMPAS applies the UV-visible Differential Optical Absorption Spectroscopy (DOAS) technique to identify SO2 and NO2 in the UV, serving as qualitative proxies for NOx and SOx emissions. Through an optical fiber bundle organised as a matrix, a coarse 2D scene of the ship plume is captured. Robust housing protects the hardware from the marine environment. The instrument is equipped with a camera, coupled with image recognition software and an Automatic Identification System (AIS) receiver to identify individual ships. The vessels are actively tracked along their trajectory, to perform multiple consecutive measurements of a single ship. In the absence of nearby ships, the instrument operates as a Multi-Axis (MAX-)DOAS system providing regular measurements of the vertical profiles of NO2 and HCHO. These can be used for atmospheric composition research, as well as satellite validation in a marine environment. In this study, first shipping results are presented from a test campaign in Zeebrugge, Belgium. The ship tracking algorithm is stable, while NO2 and SO2 measurements show large variations. A good agreement is found between MAX-DOAS measurements and satellite observations, with discrepancies dominated by local emissions from the port environment.
Competing interests: At least one of the (co-)authors is a member of the editorial board of Atmospheric Measurement Techniques.
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
(11163 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (extended)
- RC1: 'Comment on egusphere-2026-3783', Anonymous Referee #1, 21 Sep 2026 reply
Data sets
L2 data from the SEMPAS instrument in Zeebrugge, Belgium Gytha Mettepenningen, Caroline Fayt, Michel Van Roozendael https://doi.org/10.18758/iybb1vnm
Video supplement
Ship tracking with the SEMPAS instrument in Zeebrugge, Belgium Gytha Mettepenningen https://doi.org/10.5446/73638
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 140 | 82 | 72 | 294 | 65 | 60 |
- HTML: 140
- PDF: 82
- XML: 72
- Total: 294
- BibTeX: 65
- EndNote: 60
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
This manuscript details the development, testing and initial evaluation of a new UV-Vis spectroscopic instrument (SEMPAS) intended for use in monitoring open sea shipping emissions of NOx and SOx. The instrumental components of SEMPAS and the ship tracking algorithm employed are described in detail and ship plume measurements are reported from an installation in the port of Zeebrugee from March – May 2025. The authors also describe how the SEMPAS instrument performs in relation to other MAX-DOAS ground based instruments during the CINDI-3 campaign and include a comparison of SEMPAS columns retrievals with the TROPOMI satellite instrument.
This paper is of interest to the atmospheric research community and fits within the scope of AMT because it addresses the gap in ground-based continuous monitoring of shipping emissions and presents a method that could be used to flag ships that are in violation of NOx and SOx emission thresholds. Overall the paper is well written and is of good scientific quality and I recommend publication after the following comments are addressed:
Major comment
Section 5: A major goal of the manuscript is to demonstrate that SEMPAS will be able to quantitatively measure enhancements in SO2 and NO2 from ship plumes in the North Sea (to flag vessels for further sniffer measurements). However, the case studies shown in this section are not convincing of this due to remaining uncertainties in finding the plume center and changes in background SO2 and NO2 that occur during the measurements. The authors also note in the Abstract that “The ship tracking algorithm is stable, while NO2 and SO2 measurements show large variations.”
Even though this is an initial study and there will still be future refinements to the tracking/measurement algorithm, a more detailed discussion of the current issues and limitations in reference to the case studies is needed here. Some questions that should be addressed are:
1) How exactly will the mismatch in SO2 and NO2 peaks be addressed?
2) How will the sensitivity of SO2 enhancements over the background be achieved and is this technique sensitive enough to address the need for an instrument that has low enough detection limits to flag vessels with a FSC slightly above 0.1%?
3) It is noted on Lines 52-54 that “The NO2 to NOx ratio depends on plume dynamics and the local atmospheric O3 content, thereby contributing to the uncertainty in the derivation of emission factors from trace gases observed in the UV by remote sensing.” How will the issue of NO-NO2 conversion in relation to total NOx concentrations be addressed moving forward?
4) How do you intend to define the enhancements in relation to the background and is there a quantitative metric that you are planning to use for the flagging of ships?
The authors discuss potential improvements to SEMPAS in the Conclusions & Outlook section, but it would also be nice to see a full discussion of this within the context of the observations presented.
Minor comments:
Abstract (Lines 13-14): “The ship tracking algorithm is stable, while NO2 and SO2 measurements show large variations.” –> Suggest expanding on this in the abstract and describe the causes of the NO2 and SO2 variations in an additional sentence or two.
Instrument description (Line 86): “Using scattered sunlight, SEMPAS identifies atmospheric trace gases at distances up to ±10-20 km distance” –> How is the observed atmospheric path length calculated and how is this estimate of distance made? And under what conditions does the path length approach 10 km?
Instrument description (Lines 120-121): “The following calibrations are applied: pixels-to-wavelength mapping, pixels-to-fibers mapping, dark current, electronic offset and straylight.” –> Are the effects of detector non-linearity and pixel response non-uniformity (PRNU) assessed?
Instrument description (Lines 125-126): “This reaches an accuracy of a few picometers in wavelength alignment, necessary to identify the differential structures in the spectrum’s optical density.” –> Do you have an estimate of the +/- pm level accuracy of the wavelength registration, this would be interesting to know.
Instrument description (Lines 144-147): “Repeating this process at a range of wavelengths, the straylight behaviour over the spectral 145 range is characterised. For the SEMPAS system, this was done with the double Bentham monochromator from the Belgian Radiometric Characterization Laboratory (B.RCLab, https://brclab.aeronomie.be/laboratory/wavelength-scale-calibration) at BIRA-IASB, for both gratings.” –> What wavelengths were scanned and did the stray light vary across the spectrum? How did the spectral response functions look? I agree with the authors when they say the straylight characterization is a complex aspect of calibration so it would be nice to see a supporting figure showing some results from this calibration step
Instrument description (Line 155): “The optical window is cleaned manually on a monthly basis.” –> Is once a month good enough, are the instrument field of views regularly checked in some manner? How does the instrument schedule handle the build-up of moisture at the entrance window? An expansion of this section on regular instrument maintenance would be of interest to readers, especially since the future goal is to deploy the instrument on an offshore location.
Instrument description (Line 184): “The size and color of the ship have little influence on the performance” –> Was this quantified somehow? How was this conclusion reached? Under what conditions were ship detections not successful (eg. in Figure 12 presented later in the paper)?
Retrieval algorithm (Lines 296-298): “Within a time series of images acquired during ship tracking, the motion of the plume can sometimes be identified. This improves the delineation between plume and background regions and supports a more accurate estimation of ship emissions” –> Are there plans to include this feedback in the tracking algorithm?
CINDI Results (Lines 349-352): “The results demonstrate that the SEMPAS instrument, equipped with the UV–vis grating, achieves performance in the NO2 VIS 350 retrieval window comparable to that of established research-grade MAX-DOAS instruments. As the remaining instrumental characteristics are identical, these findings are expected to be representative for other configurations, such as the UV grating, as well” –> Can you demonstrate this claim? It is mentioned earlier that the sensitivity and signal-to-noise is lower in the UV window and the effect of straylight could also be different here, so I don’t know if this conclusion can be drawn. In addition, I think this section would be strengthened if there were comparisons of SO2 or HCHO from CINDI-3 also presented.
CINDI results (Lines 368-370): After the campaign, the optical alignment of the instrument was adjusted with a filter in the optical path, restoring the resolution of the instrument and reducing out-of-band straylight to less than 0.05%.” –> Again, it would be good to see some results/figures from these straylight tests/claims.
Conclusions (Lines 531-532): “A different detector could be introduced to bring more fibers into view and generate larger images.” –> This would be very interesting for observing the plume dynamics and could also be used to measure a background and enhancements simultaneously, are there any current plans to implement this moving forward?