Review of “Multi-decadal atmospheric carbon dioxide measurements in Central Europe, Hungary” by L. Haszpra, 2024

General:

The author of the manuscript presents and discusses long-term CO₂ measurements at a Central European station in Hungary. I enjoyed reading it. The manuscript is nicely written and organized and can be considered for publication with only minor changes outlined below.

Minor points:

L 10            …gives the technical description and its changes over time…

L 10            …physical environment

What does this mean? what about other conditions like to agricultural, biological situations?

L 16-17       You may state by how many days the growing period has lengthened over the entire measurement period.

L 63            yes, but what do you tell us about this? In the western European part, there are a couple of more southern stations!

L 68-69       I guess it is enough if you state this in section 2.2.

L 74            air intakes...

How many and at which levels? This information can be given already here. At least give the number of air intakes or move the sentence of line 131-132 to here.

L 94-96       These numbers are very low, what about the uncertainty of these values? The values tell us that there is no influence during winter from local emissions.

L 108-109   Reference: I could not find this information. You may consider a reference directly to the original publication where these classification are defined.

L 153-154   Also the NDIR systems can be run as absolute measurement devices. It is a question of calibration.

Fig. 3          Resolution of the graph should be increased!

L 201          What are the intake 1 and intake 2?

L 201-203   This is somewhat misleading as in this case, you could use all 2-minute values? Of course, this does not make sense as it is required to flush the Picarro measurement cell.

L 218          Calibration: this is rather infrequent with monthly calibration

L 220          What about the target tank concentration measurements? Or stability of standard gas raw measurements

Fig. 4          I guess you have averaged all days of the corresponding month of the complete data series. It would be good to add for each height uncertainty range by light shadowing using the same color.

Fig. 5          The same here, add the range of variation by color shadowing

L 311-312   You might move this sentence after the next sentence and start with: Furthermore, according to...

L 321-323   How was this calculated?

Fig. 6          The shift of the red dashed line….

if the increase is uniform

Fig. 8 Add see text for reference

Fig. 9          Which level have you used here? The lowest I guess. Anyhow write it.

L 415          NDVI values …add reference

Fig. 10        Define summer and winter by specifying the summer (months of...) as well as winter (months of ...)

L 439          This number could be checked by radiocarbon measurements in comparison with those of a marine boundary layer site.

L 451          Ocean phenomenon, how does it influence the terrestrial site as strongly as the marine boundary layer? (Fig. 13)

Fig. 13 Move the y-axis to the left

General comment

This paper presents 30 years of measurements of atmospheric carbon dioxide from a tall tower in Hungary. The Hegyhátsál station and its CO₂ measurements have already been well characterised in numerous previous publications and this paper adds an overarching view for the continuous long time series from 1994 to 2023 before joining the Integrated Carbon Observation System (ICOS) network. The changes in instrumentation and measurement set ups during that period are presented and sampling uncertainties evaluated. Using the 30 years’ worth of data, seasonal trends and features of the vertical gradient of CO₂ (10-115 m above ground) are investigated and put into a wider context. Changes such as the CO₂ seasonal trends and anomalies are tested for their statistical significance and explanations considered. Long-term trends and changes in growth rates are shown and connections are made to the El Niño-Southern Oscillation (ENSO).

The paper makes a valuable contribution to the body of work investigating and interpreting long time series of in situ carbon dioxide measurements from ground-based monitoring stations. Publication is recommended once the minor issues listed below have been addressed.

Specific questions/issues

1 Introduction

Line 23: “measurements were not convincing” – it should be clarified to whom they were not convincing, e.g. the scientific community at the time, or rather from a current view point or else. Also, the mentioned technical and representativity problems should be briefly explained, named or outlined.

Line 27: for context, consider naming the major measurement networks in which the monitoring stations were established, e.g. NOAA, GAW, etc. and/or add reference to World Data Centre for Greenhouse Gases.

Line 28/29: consider adding the first or longest running station as an example each for the list of “arctic regions, high mountain peaks, mid-oceanic islands”.  

Line 31: “they could not provide detailed information”, give an example of what is meant here with “detailed information”

Line 38: give a brief explanation what is meant by tall tower, e.g. “towers of up to 100s of metres high”

Line 39: influence by local vegetation: any type of localised influence should be relevant here, including local anthropogenic sources such as e.g. from industrial activity. Consider rephrasing.

Line 43-44: Five years of parallel measurements of K-puszta and Hegyhátsál: It would have been interesting to include the period of parallel measurements in this manuscript to extend the time series from the region, and add information to the spatial representativity of the measurements. If such an analysis already exists, please provide a citation or consider adding a short section that includes the measurements at K-puszta.

Line 50: determination of the carbon budget of the atmosphere: consider adding a reference, such as e.g. the Global Carbon project https://www.globalcarbonproject.org/index.htm

Line 58: model results depends on coverage: is temporal or spatial coverage meant here? What role does the quality of measurements play?

Line 62/63: information on emissions from Western Europe: looking at Fig.2 the given statement is supported, however in case of northerly/easterly/southerly conditions, the eastern stations can also provide emission information from other parts of Europe. In such conditions it also serves as a regional “background” to western/central Europe in model studies and inversions.

Line 68: delete “the” before “atmospheric”. Consider adding at the end of sentence “, which are also reported to WDCGG”.

2 Measurements and data

Line 78: high spatial representativeness: is it possible to give a quantitative measure or further description what is meant be “high” spatial representativeness?

Section 2.2. Monitoring system: Are any type of inlet filter used, to e.g. filter aerosol? If so, please provide technical details. If no filters are used, please comment on how fast/slow contamination by aerosol or particulate matter occurs in the lines or the KNF pumps.

Line 157/158: please provide details on the typical ratio of bypass/sample flow

Line 175-178: it could be useful to introduce, differentiate and define terms like e.g. “sampling uncertainty, which is a result of non-continuous sampling of the atmospheric variability within an hour” and “measurement, i.e. analytical, uncertainty”. The description “high frequency sampling” can potentially be misleading, as not e.g. 10 Hz data logging is meant but the cycling through to each inlet height in sequence. Consider rephrasing the sentence.

Line 180: see also previous point, consider being more specific, e.g. “sampling time/switching frequency through inlet heights”

Line 187: low-uncertainty data: in the context of the section it seems clear that reference is made to the sampling uncertainty that arises out of the large atmospheric variability when e.g. the night time boundary layer breaks up. However, the way it is presented here could give the impression that the data from boundary layer transition periods is not used in models because of the above-mentioned sampling uncertainty. This however is not the main reason and thus consider rephrasing these sentences.

Line 188: delete “when the demand appears”. This statement is a home truth, but possibly worth reiterating.

Line 191: consider changing adjective “remarkable” to “great”, quantifying also e.g. from x to xx m.

Section 2.4. An extended section and assessment on the NOAA flask and in situ measurement comparison would be welcome.

Line 236: in the larger than 3 sigma cases, the presumed reasons could be further investigated and reasons confirmed, e.g. utilising the high-resolution continuous data during the flask sampling periods.

Line 237: please specify the time period over which the mean deviation is calculated. Is the comparison robust throughout that time frame, or are there periods (years?) when the WMO compatibility goal is not achieved? Can a difference be seen for the different types of analysers and/or set ups? Can a significant relationship be observed for larger flask-in situ differences when atmospheric variability is high?

Line 241: elaborate or give examples what is meant by “technical issues”

Line 243: how many flask samples contribute to that mean and standard deviation?

Section 2.5. Data selection. Line 249/250: consider adding information in the supplementary material to support the statement made on the lack of directional difference, e.g. a wind rose showing the residuals of detrended CO2 concentrations

3 Results and discussion

Section 3.1. Line 264 and Line 311-312, Fig S2 and statement on nighttime boundary layer heights: please comment on the uncertainties in the BL height determination in the ECMWF ERA5 data product? The dynamics of the nighttime BL is given as main reason for the high summer amplitudes of CO2 (Fig 5), however it seems the effect of the daytime BL (and the dilution effect) will have a lot larger impact on the large CO2 amplitude.

Line 316/317: can the statement on the “increasing respiration due to significant increase in temperature” be further supported. Is the observed nighttime temperature increase consistent with the suggested corresponding increase in respiration rates (from literature or e.g. nearby flux measurement data)? Has the vegetation cover (type and cover spatially) around the station changed in the time period, and if so, could that play a role in the observed trends?

Line 321/322: although reference to Haszpra et al., 2015 is made, please briefly describe salient points of the aircraft campaign, e.g. how was the top of the boundary layer sampled/derived (aerosol, T-profiles)? How many flights contribute to that mean, and what is meant by the “top of the tower underestimates the mean planetary boundary layer”? What could be the reason for the observed mean differences with the aircraft-based measurements? Please also be more specific as to how these results can be “informative for those using models with coarse vertical resolution” (Line 324).  

Section 3.2. Line 368/9: please provide a citation/reference for this statement

Line 374/375: please provide data and/or a citation/reference for this statement. Also comment on the role of regional sources in winter, e.g. from fossil fuel combustion for heating and energy?

Line 381/381: delete “but to the author’s knowledge, the role of the changes in the dynamics of the atmosphere has not yet been studied in details.”

Line 397: as general statement okay, however please be more location specific. Are additional phenology data from e.g. the Hungarian Met Service or else available that could be used for comparison and support here?

Line 397-400: provide guidance as which papers are relevant for the region/Hungary, and/or provide additional ones that show trend in Hungary

Line 408: statement is made that trend is not significant, please provide details on trend and p value

Line 421-423: please comment on the uncertainty that is involved in the seasonal detrending, that could affect the trend analysis and thus possible detection of small trends

Line 425: decreasing winter peaks: please consider other reasons for decreasing winter peaks such as e.g. mild winters (less heating), events such as economic slowdown in the region and recently Covid-19 pandemic.

Line 427: vague statement, consider rephrasing

Section 3.3. Line 449: It is unclear what is meant by “emissions vary in much narrower ranges”. The previous sentence contains information on atmospheric concentration, here emissions are mentioned, which does not directly follow on from the data from Hegyhátsál.

Figure 12: please double-check the values of the growth rates for Hegyhátsál. Consider putting them in context with the global growth rate, as e.g. published in the WMO Greenhouse Gas Bulletin Nr.19 (Nov 2023) https://library.wmo.int/records/item/68532-no-19-15-november-2023?offset=1 . Global growth rates have been above 1 umol/mol/year since 1994, in the order of 1 to ~3.2 umol/mol/year.

Line 466-485: this section describes the observed pattern with ENSO, detailing lag times, however not sufficient detail is given what causes the effect on CO2 and how this information can be utilised and for whom it is relevant?

4 Summary

Consider changing the summary to a conclusion, highlighting e.g. the value in such long-term observational data time series, their usage by others (modellers…) etc. This would strengthen the manuscript.

Technical corrections

Abstract:

Line 9: be specific: “GAW ID code: HUN”. Consider adding WIGOS station identifier 0-348-4-16307, here or later in Section 2.1

1 Introduction

Line 19: consider changing “raised” to “suggested”

Line 22: add “that” before “it might”

Line 29: delete “However,” starting the sentence with “One of the main (…)”

Line 31: insert “background” after “global”

Line 45: exchange “several” with “many”

Line 63: exchange “development” with “expansion”

2 Measurements and data

Line 83: delete square brackets, and add “which includes e.g.” “after 6% other”. Delete comma and “etc” after “settlements”

Line 84: “lessivated” without capitalised L

Line 85: delete comma after “Alfisol”

Line 91: unclear what type of road “2x1-line” describes, how many lanes?

Line 169: consider changing “multi-elevation monitoring site” to “multi inlet height monitoring site”

Line 183/184: change to “Today models try to avoid these transition periods (…)”

Line 184: change to “CO₂ concentration tends to be low.”

Line 198: exchange “ventilated” with “flushed”

Line 200/201: change “Intake” to “intake”, not capitalising the word

Line 219: delete one full stop at the end of sentence

3 Results and discussion

Line 323: in second square brackets delete comma before “2”

Line 354: delete “overlying”

Line 401: first use of “NDVI”, please spell out

Line 436: consider changing “permanent” to “steady”

Line 446: change “bend” to “band”

Line 468: change “Barring Head” to “Baring Head”

Line 471: change “Plateau Rose” to “Plateau Rosa”

Line 492: delete “yet”