the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 10 Jul 2023
Measurement Report: Exchange Fluxes of HONO over Agricultural Fields in the North China Plain
Yifei Song
Yuanyuan Zhang
Pengfei Liu
Fengxia Bao
Xuran Li
Abstract. Nitrous acid (HONO) is a crucial precursor of tropospheric hydroxyl radicals but its sources are not fully understood. Soil is recognized as an important HONO source, but the lack of measurements of soil-atmosphere HONO exchange flux (FHONO) has led to uncertainties in modeling its atmospheric impacts and understanding the reactive nitrogen budget. To address this, we conduct long-period FHONO measurements over agricultural fields under fertilized (FHONO-NP) and non-fertilized (FHONO-CK) treatments. Our results show that nitrogen fertilizer use causes a remarkable increase in FHONO-NP and it exhibits distinct diurnal variations, with an average noontime peak of 152 ng N m-2 s-1. The average FHONO-NP within three weeks after fertilization is 97.7 ± 8.6 ng N m-2 s-1, around two orders of magnitude higher than before fertilization, revealing the remarkable promotion effect of nitrogen fertilizer on HONO emissions.
We also discuss other factors that influence soil HONO emissions, such as meteorological parameters and soil properties/nutrients. Additionally, we estimate the HONO emission factor of 0.68 ± 0.07 % relative to the applied nitrogen during the whole growing season of summer maize. Accordingly, the fertilizer-induced soil HONO emission is estimated to be 0.06 and 0.16 Tg N yr-1 in the North China Plain (NCP) and China, respectively, representing a significant reactive nitrogen source. Furthermore, our observations reveal that soil emissions sustain a high level of daytime HONO, enhancing the atmospheric oxidizing capacity and aggravating O3 pollution in the NCP. Our results indicate that in order to effectively mitigate regional air pollution, future policies should consider reactive nitrogen emissions from agricultural soils.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(1215 KB)
-
Supplement
(239 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1215 KB) - Metadata XML
-
Supplement
(239 KB) - BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Yifei Song et al.
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-1223', Anonymous Referee #1, 20 Jul 2023
This paper studied soil HONO emissions from fertilized soil in Chinese agricultural land. In general, the authors provide some interesting field flux data, and discussed potential influencing factors and atmospheric implications.
I have some comments need to be addressed before it can be accepted.
- Introduction, I would suggest the authors focus on literature review in fertilization caused soil HONO emissions, both in field and laboratory studies. The review in methods of HONO flux measurement are not necessary.
- L173, the unit should be m2.
- L177, the calculation method of cumulative HONO-N emissions should be introduced. Many data points were missed in Figure 3, how did you fill these gaps?
- Soil temperature data are encouraged to provide during field HONO measurements.
- L257-260, these sentences can be moved to the Methods section.
- L323-324, another reason could be high soil moisture decrease gas diffusion in soil profile.
- L380, delete the word of decreasing.
- L382-383, we estimate a relationship between the emission factor caused by fertilization and fertilization amount, please see the reference Wu et al., 2022 JGR: Atmospheres, 127, e2021JD036379. I would suggest the authors can discuss the results in this manuscript with our results.
- L445, I would not call it long-period measurements if the authors only showed ~ 1 month available data.
- Figure 5 is confusing with two units. The authors should separate them with different axis.
- The authors cited too many of their own works, while lacking compare with other studies, such as the work from groups of Jonathan Raff, Marja Maljanen, and Dianming Wu. I would suggest the authors can discuss and compare their results with that from other scientists.
Citation: https://doi.org/10.5194/egusphere-2023-1223-RC1 -
AC1: 'Reply on RC1, RC2, and RC3', Chaoyang Xue, 08 Oct 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1223/egusphere-2023-1223-AC1-supplement.pdf
-
RC2: 'Comment on egusphere-2023-1223', Anonymous Referee #2, 25 Jul 2023
This study by Song et al. reports long-period measurements of HONO fluxes above agricultural fields in the North China Plain (NCP). Experiments are conducted on both normal fertilizer use and no fertilizer use fields. The reported soil HONO emissions are carefully compared with existing literature. The influencing factors and atmospheric implications (nitrogen budget, ozone air quality) of soil HONO emission are also analyzed.
This study fills the largely missing measurement of soil HONO emissions during fertilizer periods in the NCP, which is much underappreciated in current studies of air quality in this region. This is a valuable and significant contribution to the community. The analyses are comprehensive and informative. This is also a well-structured, well-detailed, and well-written manuscript. I recommend publication in ACP after some minor revisions.
My little concern is about the application of the measured soil HONO emission factor (0.68%) to estimate the national scale soil HONO emissions (Section 3.6.2). While this estimate is important and valuable, I wonder to what extent this emissions factor can be applied to other regions? As the fertilizer type, approach, and meteorological parameters differ in other regions with NCP, the emission factor would likely not be constant, then the national estimate might be biased. Some discussions of the limitation of this approach and comparison with other studies on the national scale might be helpful.
Section 3.6.3: does it also include ozone contributed by soil NOx (not solely from soil HONO)? Please clarify.
Line 254: “it is worth noting that the high water content but no fertilization for the CK plots may contribute to the negative fluxes.” It is not clear to me (and possibly other readers) how high water content would lead to negative flux. Please clarify.
Citation: https://doi.org/10.5194/egusphere-2023-1223-RC2 -
AC1: 'Reply on RC1, RC2, and RC3', Chaoyang Xue, 08 Oct 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1223/egusphere-2023-1223-AC1-supplement.pdf
-
AC1: 'Reply on RC1, RC2, and RC3', Chaoyang Xue, 08 Oct 2023
-
RC3: 'Comment on egusphere-2023-1223', Anonymous Referee #3, 08 Aug 2023
In the manuscript by Song et al. HONO fluxes from freshly fertilized soil surfaces are compared to non-fertilized soils in open-top dynamic chambers (OTDC) in China. Very high fluxes were determined for fertilized soils in good agreement with previous similar experiments, which were however much higher compared to all other HONO flux measurements by the gradient and REA methods under normal fertilization conditions (see below). Since fluxes give more direct information on ground surface sources and sinks compared to former PSS approaches, HONO flux measurements are highly recommended. The manuscript contains interesting results (e.g. for the first time cumulative HONO emissions after fertilization) and also valuable recommendation how to reduce HONO emissions by fertilizer usage and should be published after my concerns have been considered.
Major concerns:
1) Method used
In Table 1 different flux results are compared, nicely showing that in most OTDC studies significantly higher fluxes were determined compared to well-established open flux measurements like the aerodynamic gradient and relaxed Eddy accumulation methods. There are two possible explanations for this observation:
a) Fluxes in open dynamic chambers are overestimated by different potential artifacts. First, the temperature in a Teflon chamber and of the soil surface – even in open dynamic chambers – will be significantly higher compared to real open surfaces by the greenhouse effect (ca. 4 min residence time, ca. 1000 W/m2 at noon, low heat capacity of the air…). This will not only increase biological activity, but has also an influence on the soil surface coverage by water affecting surface exchange processes. Higher temperature decreases the Henry’s law constant of HONO and reduces water adsorption, decreasing the volume of water for solution. Both will increase HONO fluxes. This cannot be considered for by using a reference chamber and/or by comparing fertilized with non-fertilized soils as done in the present study (which I highly appreciate!). Second, by changing surface humidity and temperature also artificial heterogeneous HONO formation on chamber surfaces may be enhanced (ca. 4 min residence time). Typically, this is aimed to be considered for by using the difference to the reference chamber covered with a Teflon foil on the ground. However, if for example soils emit electron-rich VOCs and NOx (for the latter see the results of the present study), which are known to heterogeneously form HONO (see e.g. George et al., 2005) than the artificial heterogeneous HONO formation on chamber surfaces will be higher in the soil chamber compared to the reference chamber. And fertilized soils will emit more NOx compared to non-fertilized soils. Also here the comparison of fertilized with non-fertilized soils may not help, since e.g. NOx emissions may be much higher on fertilized soils. Furthermore, irradiated Teflon surfaces are known to artificially form HONO under irradiation, which is still not completely understood (see e.g. Rohrer et al. 2005; doi: 10.5194/acp-5-2189-2005). The latter artifact however, may be considered for in the present study by the use of the reference chamber.
In conclusion to a) larger surfaces/chambers always bear the risk of overestimating HONO fluxes! Thus, I would highly recommend an intercomparison of the OTDC method with direct AG or REA measurements in the future (before publishing more OTDC results…) and at least to highlight in the present study that the fluxes determined in OTDCs may be overestimated.
b) The higher HONO fluxes may be caused by the very high fertilizer application amounts of ca. 300 kg N / ha applied (see line 142), which is much higher compared to typical amounts of <100 kg N/ha applied in most regions of the world and on soils in other HONO flux studies (see table 1). This is especially important, since the authors observed an extremely non-linear relationship between the HONO fluxes and the nitrogen application amount (see the nice results shown in Figure S6 in Xue et al., 2022a, which will bring the very different published flux results together). From this exponential relationship HONO fluxes at 300 kg N/ha will be almost two orders of magnitude higher compared to fluxes at the average global application amount of 75 kg N/ha (see reference in Xue et al., 2022a), which is exactly the ratio observed between former flux studies and those in the recent OTDC studies (see Table 1). This non-linear relationship should be highlighted and discussed to explain the different flux results. In addition, the authors should consider the exponential relationship for estimation of regional and global HONO emissions from fertilized soils. Here much lower contribution will be obtained for any typical fertilization amount.
2) Estimated implication:
This brings me to my second major concern, the estimated HONO emissions for China. In their implication section 3.6 the authors used a constant EF(HONO) of 0.68% HONO emitted per amount of N-fertilizer applied. However, if the HONO fluxes are exponentially increasing with the fertilizer amount (see again Figure S6 in Xue et al., 2022a) a constant ratio cannot be used! Since this ratio was determined at very high fertilizer amount of 300 kg N /ha the average EF(HONO) will be much lower, considering variable application rates typically lower than that extreme value, even in China (see e.g. Figure 2a in Potter et al., 2010, DOI: 10.1175/2010EI288.1). The authors should apply the exponential relation considering regional variable fertilizer application amounts. In addition, the authors should estimate HONO fluxes from this relationship for typical global nitrogen application amounts (<100 kg N /ha) to highlight that the results are only applicable to the more extreme Chinese fertilization conditions. Here I expect at least one order of magnitude lower HONO emissions.
The results shown in Figure S6 in Xue et al., 2022a may also identify “over-fertilization”, for which nitrite formation by biological ammonium oxidation may be faster than the biological nitrite oxidation. In this case nitrite may accumulate and HONO emissions may be controlled by solubility in soil water (see discussion of the temperature and humidity dependence in section 3.5.2) which should be avoided! Thus, the authors may use these results to also recommend more environment-friendly fertilization amounts in the future for China (similar to most other regions of the world) besides their recommendation of the DF method.
Minor comments in the order how they appear in the manuscript:
Line 51: please first use a reference to Song et al., 2022a before Song et al., 2022b (and exchange both in the list…). The same for other references, e.g. line 54: Xue et al., 2022b is used before Xue et al., 2022a.
Line 53: Better use the reference Kleffmann, 2007 (doi: 10.1002/cphc.200700016), since in the referred 2003 study daytime sources were not a main focus.
Line 56 and 57: Better use here only studies in which the daytime source could be quantified based only on experimental PSS data and not on uncertain assumptions. Here the first studies in which the PSS and the unknow HONO source could be quantified fully by experimental data were by Kleffmann et al., 2005 (doi: 10.1029/2005GL022524) and Acker et al., 2006 (doi:10.1029/2005GL024643). Thus, I would delete Kleffmann et al., 2003, Sparataro et al., 2013 and Su et al., 2008 (no exp. OH data, OH source by HONO only estimated).
Line 68: Add reference by Tang et al., 2020.
Line 78: use “von der Heyden et al.” throughout the text.
Line 78: add references by Ren et al., 2011, Zhou et al., 2011, Laufs et al., 2017.
Line 168, equation-1: First, the units are missing for the factor 1/60 (min/s) and second, a factor 109 ng/g is missing. But I would recommend to delete all factors and simply give SI units for all terms (i.e. F in m3/s; M in ng/mol, P in Pa, R in J/mol/K).
Line 172: K-1
Line 186, Figure S1: The shown correlation is not “high” but may be fair (see scatter in Figure S1). In addition, the intercept will be zero by definition (J(NO2) = 0 during night…). And finally, a measure for the actinic flux (J(NO2)) will not linearly correlate with a measure of the irradiance (“solar radiation”). Here a non-linear correlation is expected since only for the irradiance the cos-dependence has to be considered (check with the TUV model). A quadratic dependence forced by zero will better describe the relation between both terms.
Line 193: Isn’t the soil particle density depending on the soil type and may be very different to the value of 2.65 g/cm3 in Linn and Doran, 1984? Please give uncertainty for the WFPS.
Line 199-200: For the typical applied extraction by unbuffered KCl solution underestimation of nitrite is well-known, see Homyak et al., 2015 (https://doi.org/10.2136/sssaj2015.02.0061n). Here the nitrite concentrations should be considered as lower limit.
Line 214: Better give an average daytime J(NO2)? The total campaign average will be otherwise restricted by the zero nighttime values.
Line 236-239: Fluxes measured in other studies are not only similar in comparison to the CK plots but also to all (CK and NP) PFP and LEP data. I.e. only the fluxes observed shortly after fertilization (HEP) with extremely high N-amount are much higher (see major concern 2).
Line 273: add references of Meng et al., 2022 and von der Heyden et al., 2022 with similar conclusions.
Line 277-280: First, also in Ren et al. (CALNEX data) and Laufs et al. soils were studied which are regularly fertilized. I.e. they are not “non-fertilized”. Second, the fluxes from the two studies are also in agreement with the present study, if the normal PFP and LEP data - even of fertilized soils - are considered. Thus, these low fluxes represent more typical conditions, while only during the short HEP period (few weeks) extremely high fluxes (two orders of magnitude higher) are observed for "over-fertilized" soils, see major concerns.
Line 290-293: Delete the references by Laufs et al, 2017; Meng et al, 2022; Ren et al., 2011; Sörgel et al., 2015; Zhou et al., 2011, since in these studied two orders of magnitude lower fluxes were determined.
Line 320: exchange order of NO2-/ H+ (acid-base…)
Section 3.5.2: If the temperature and humidity are the driving forces of the soil emissions, than the shape of the diurnal HONO flux profile should be similar with that of the temperature (and the inverse of the relative humidity). However, in most studies the HONO fluxes maximize around noon (see also Fig. 5A of the present study…), while the temperature shows a maximum in the afternoon (heat capacity of the soil, when heated up by the sun during daytime). Thus, at least under “normal” conditions, there must be other explanations for the HONO fluxes. Here I recommend that the authors also show a correlation of the normal diurnal flux data (Figure 5A) with the product of J(NO2) x [NO2] as done in other studies. I expect that the correlation is much better compared when using the temperature.
Only for extreme HEP conditions (see Figure 5B, representative for a few weeks per year for “over-fertilized” soils…) the diurnal profile of F(HONO) may well follow that of the temperature (see also the correlation results shown in Fig. 6 for HEP).
Line 357-359: This may be valid only for untypical HEP conditions. For the more typical PFP and LEP conditions, I expect a photochemical origin (s. above: check by plotting F(HONO) against J(NO2) x NO2).
Section 3.6: Re-evaluate the whole section after considering the non-linear relation between F(HONO) and fertilization amount.
Conclusion 1): Add a conclusion that the average F(HONO) data for all PFP and LEP conditions are in excellent agreement with most flux studies under normal fertilization conditions.
Conclusion 2): Add that these results only hold for extremely high nitrogen application during short periods after fertilization in China.
Conclusion 3): please do not use a constant value of EF(HONO), see above.
References:
General:
Order the references with same first authors chronologically (i.e. increasing year of publication), see VandenBoer et al., Wang (Y.) et al., Wu et al., Xue et al.. In addition, if an author has two publications in one year (a and b) use first the reference a) in the text.
Line 500, 580, 640, 716, 731: Pöschl,
Line 522, 688: Dubé
Line 526: Lörzer
Line 528: the doi link is not working?
Line 533-534 and others (644-647, 650-653, 660-661, 708-711, 723, 729, 780): please unify the style for Chinese given names and use only the first letter. E.g. Line 533: should be Li, D. and not Li, D. D.!
Line 537, 543, 645: Häseler
Line 544: Jäger
Line 557 and 594: Min, K.-E.
Line 579: …-Röser
Line 579, 715, 731: Sörgel
Line 654: subscript the “x” in ROx
Line 688: Öztürk
Line 703: 57(9), 3516-3526 missing
Line 710: Petäjä
Line 723: Huang, X.-R. Y.
Line 727: paper number e2021JD036379 missing
Line 729: Müller
Line 730: Fröhlich-…
Line 767: HOx and subscript the “x“
Line 772: use paper number L15820 (delete n/a-n/a)
Table 1: Here max fluxes are compared with the maximum in the average diurnal flux profile (see footnote d). I recommend to use the latter also for all other studies, since outliers are removed by this approach. Even better, one could simply compare average fluxes. In this case in all studies (except the OTDC fresh fertilization studies) the average fluxes would be in the range 0.1-2 ng N/m2/s…
Fig. 1: In Figure 1(B) I do not see items 10. and 11.? Please add to the figure. In addition, please specify in the methods section which type of pump (5.) is used (Teflon membrane pump?). Here additional artifacts may appear depending on the type of pump used, see major issue 1a)…
Figure 2. Here a very strong rain event (ca. 150 mm) is shown around the 9. September, while J(NO2) is very high at that day? In contrast, for the 19. September there is almost no light intensity (J(NO2)) but no rain? Please check the rain data. Here I find different rain periods in Figure S2? In addition, is there no J(NO2) data in the period 11-15. of July, or is it very low?
Figure 3: Unfortunately, after the PEP period there is a data gap, after which the HONO fluxes are again very low in the LEP period. However, if I would fit any Gaussian profile into the HEP flux data, I would get much higher HONO fluxes in the early LEP period. Was the soil anyhow treated directly after the HEP period? How do you explain the sudden step in the flux profile? Explain the terms NP and CK in the figure caption (a figure should stand alone).
Figure 4/5/6: dito for PFP, HEP and NP in figure caption 4/5/6.
Figure 6: Can you also show a plot against J(NO2) x [NO2]? Could be also in the supplement.
Figure 8: Show a new figure considering the non-linear relation between F(HONO) and fertilization amount.
Figure S3: Describe the arrows in the lower figure in the caption. E.g. "The arrows indicate reduced HONO fluxes after rain events."
Citation: https://doi.org/10.5194/egusphere-2023-1223-RC3 -
AC1: 'Reply on RC1, RC2, and RC3', Chaoyang Xue, 08 Oct 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1223/egusphere-2023-1223-AC1-supplement.pdf
-
AC1: 'Reply on RC1, RC2, and RC3', Chaoyang Xue, 08 Oct 2023
-
AC2: 'Revised Manscript with Changes Tracked', Chaoyang Xue, 08 Oct 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1223/egusphere-2023-1223-AC2-supplement.pdf
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-1223', Anonymous Referee #1, 20 Jul 2023
This paper studied soil HONO emissions from fertilized soil in Chinese agricultural land. In general, the authors provide some interesting field flux data, and discussed potential influencing factors and atmospheric implications.
I have some comments need to be addressed before it can be accepted.
- Introduction, I would suggest the authors focus on literature review in fertilization caused soil HONO emissions, both in field and laboratory studies. The review in methods of HONO flux measurement are not necessary.
- L173, the unit should be m2.
- L177, the calculation method of cumulative HONO-N emissions should be introduced. Many data points were missed in Figure 3, how did you fill these gaps?
- Soil temperature data are encouraged to provide during field HONO measurements.
- L257-260, these sentences can be moved to the Methods section.
- L323-324, another reason could be high soil moisture decrease gas diffusion in soil profile.
- L380, delete the word of decreasing.
- L382-383, we estimate a relationship between the emission factor caused by fertilization and fertilization amount, please see the reference Wu et al., 2022 JGR: Atmospheres, 127, e2021JD036379. I would suggest the authors can discuss the results in this manuscript with our results.
- L445, I would not call it long-period measurements if the authors only showed ~ 1 month available data.
- Figure 5 is confusing with two units. The authors should separate them with different axis.
- The authors cited too many of their own works, while lacking compare with other studies, such as the work from groups of Jonathan Raff, Marja Maljanen, and Dianming Wu. I would suggest the authors can discuss and compare their results with that from other scientists.
Citation: https://doi.org/10.5194/egusphere-2023-1223-RC1 -
AC1: 'Reply on RC1, RC2, and RC3', Chaoyang Xue, 08 Oct 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1223/egusphere-2023-1223-AC1-supplement.pdf
-
RC2: 'Comment on egusphere-2023-1223', Anonymous Referee #2, 25 Jul 2023
This study by Song et al. reports long-period measurements of HONO fluxes above agricultural fields in the North China Plain (NCP). Experiments are conducted on both normal fertilizer use and no fertilizer use fields. The reported soil HONO emissions are carefully compared with existing literature. The influencing factors and atmospheric implications (nitrogen budget, ozone air quality) of soil HONO emission are also analyzed.
This study fills the largely missing measurement of soil HONO emissions during fertilizer periods in the NCP, which is much underappreciated in current studies of air quality in this region. This is a valuable and significant contribution to the community. The analyses are comprehensive and informative. This is also a well-structured, well-detailed, and well-written manuscript. I recommend publication in ACP after some minor revisions.
My little concern is about the application of the measured soil HONO emission factor (0.68%) to estimate the national scale soil HONO emissions (Section 3.6.2). While this estimate is important and valuable, I wonder to what extent this emissions factor can be applied to other regions? As the fertilizer type, approach, and meteorological parameters differ in other regions with NCP, the emission factor would likely not be constant, then the national estimate might be biased. Some discussions of the limitation of this approach and comparison with other studies on the national scale might be helpful.
Section 3.6.3: does it also include ozone contributed by soil NOx (not solely from soil HONO)? Please clarify.
Line 254: “it is worth noting that the high water content but no fertilization for the CK plots may contribute to the negative fluxes.” It is not clear to me (and possibly other readers) how high water content would lead to negative flux. Please clarify.
Citation: https://doi.org/10.5194/egusphere-2023-1223-RC2 -
AC1: 'Reply on RC1, RC2, and RC3', Chaoyang Xue, 08 Oct 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1223/egusphere-2023-1223-AC1-supplement.pdf
-
AC1: 'Reply on RC1, RC2, and RC3', Chaoyang Xue, 08 Oct 2023
-
RC3: 'Comment on egusphere-2023-1223', Anonymous Referee #3, 08 Aug 2023
In the manuscript by Song et al. HONO fluxes from freshly fertilized soil surfaces are compared to non-fertilized soils in open-top dynamic chambers (OTDC) in China. Very high fluxes were determined for fertilized soils in good agreement with previous similar experiments, which were however much higher compared to all other HONO flux measurements by the gradient and REA methods under normal fertilization conditions (see below). Since fluxes give more direct information on ground surface sources and sinks compared to former PSS approaches, HONO flux measurements are highly recommended. The manuscript contains interesting results (e.g. for the first time cumulative HONO emissions after fertilization) and also valuable recommendation how to reduce HONO emissions by fertilizer usage and should be published after my concerns have been considered.
Major concerns:
1) Method used
In Table 1 different flux results are compared, nicely showing that in most OTDC studies significantly higher fluxes were determined compared to well-established open flux measurements like the aerodynamic gradient and relaxed Eddy accumulation methods. There are two possible explanations for this observation:
a) Fluxes in open dynamic chambers are overestimated by different potential artifacts. First, the temperature in a Teflon chamber and of the soil surface – even in open dynamic chambers – will be significantly higher compared to real open surfaces by the greenhouse effect (ca. 4 min residence time, ca. 1000 W/m2 at noon, low heat capacity of the air…). This will not only increase biological activity, but has also an influence on the soil surface coverage by water affecting surface exchange processes. Higher temperature decreases the Henry’s law constant of HONO and reduces water adsorption, decreasing the volume of water for solution. Both will increase HONO fluxes. This cannot be considered for by using a reference chamber and/or by comparing fertilized with non-fertilized soils as done in the present study (which I highly appreciate!). Second, by changing surface humidity and temperature also artificial heterogeneous HONO formation on chamber surfaces may be enhanced (ca. 4 min residence time). Typically, this is aimed to be considered for by using the difference to the reference chamber covered with a Teflon foil on the ground. However, if for example soils emit electron-rich VOCs and NOx (for the latter see the results of the present study), which are known to heterogeneously form HONO (see e.g. George et al., 2005) than the artificial heterogeneous HONO formation on chamber surfaces will be higher in the soil chamber compared to the reference chamber. And fertilized soils will emit more NOx compared to non-fertilized soils. Also here the comparison of fertilized with non-fertilized soils may not help, since e.g. NOx emissions may be much higher on fertilized soils. Furthermore, irradiated Teflon surfaces are known to artificially form HONO under irradiation, which is still not completely understood (see e.g. Rohrer et al. 2005; doi: 10.5194/acp-5-2189-2005). The latter artifact however, may be considered for in the present study by the use of the reference chamber.
In conclusion to a) larger surfaces/chambers always bear the risk of overestimating HONO fluxes! Thus, I would highly recommend an intercomparison of the OTDC method with direct AG or REA measurements in the future (before publishing more OTDC results…) and at least to highlight in the present study that the fluxes determined in OTDCs may be overestimated.
b) The higher HONO fluxes may be caused by the very high fertilizer application amounts of ca. 300 kg N / ha applied (see line 142), which is much higher compared to typical amounts of <100 kg N/ha applied in most regions of the world and on soils in other HONO flux studies (see table 1). This is especially important, since the authors observed an extremely non-linear relationship between the HONO fluxes and the nitrogen application amount (see the nice results shown in Figure S6 in Xue et al., 2022a, which will bring the very different published flux results together). From this exponential relationship HONO fluxes at 300 kg N/ha will be almost two orders of magnitude higher compared to fluxes at the average global application amount of 75 kg N/ha (see reference in Xue et al., 2022a), which is exactly the ratio observed between former flux studies and those in the recent OTDC studies (see Table 1). This non-linear relationship should be highlighted and discussed to explain the different flux results. In addition, the authors should consider the exponential relationship for estimation of regional and global HONO emissions from fertilized soils. Here much lower contribution will be obtained for any typical fertilization amount.
2) Estimated implication:
This brings me to my second major concern, the estimated HONO emissions for China. In their implication section 3.6 the authors used a constant EF(HONO) of 0.68% HONO emitted per amount of N-fertilizer applied. However, if the HONO fluxes are exponentially increasing with the fertilizer amount (see again Figure S6 in Xue et al., 2022a) a constant ratio cannot be used! Since this ratio was determined at very high fertilizer amount of 300 kg N /ha the average EF(HONO) will be much lower, considering variable application rates typically lower than that extreme value, even in China (see e.g. Figure 2a in Potter et al., 2010, DOI: 10.1175/2010EI288.1). The authors should apply the exponential relation considering regional variable fertilizer application amounts. In addition, the authors should estimate HONO fluxes from this relationship for typical global nitrogen application amounts (<100 kg N /ha) to highlight that the results are only applicable to the more extreme Chinese fertilization conditions. Here I expect at least one order of magnitude lower HONO emissions.
The results shown in Figure S6 in Xue et al., 2022a may also identify “over-fertilization”, for which nitrite formation by biological ammonium oxidation may be faster than the biological nitrite oxidation. In this case nitrite may accumulate and HONO emissions may be controlled by solubility in soil water (see discussion of the temperature and humidity dependence in section 3.5.2) which should be avoided! Thus, the authors may use these results to also recommend more environment-friendly fertilization amounts in the future for China (similar to most other regions of the world) besides their recommendation of the DF method.
Minor comments in the order how they appear in the manuscript:
Line 51: please first use a reference to Song et al., 2022a before Song et al., 2022b (and exchange both in the list…). The same for other references, e.g. line 54: Xue et al., 2022b is used before Xue et al., 2022a.
Line 53: Better use the reference Kleffmann, 2007 (doi: 10.1002/cphc.200700016), since in the referred 2003 study daytime sources were not a main focus.
Line 56 and 57: Better use here only studies in which the daytime source could be quantified based only on experimental PSS data and not on uncertain assumptions. Here the first studies in which the PSS and the unknow HONO source could be quantified fully by experimental data were by Kleffmann et al., 2005 (doi: 10.1029/2005GL022524) and Acker et al., 2006 (doi:10.1029/2005GL024643). Thus, I would delete Kleffmann et al., 2003, Sparataro et al., 2013 and Su et al., 2008 (no exp. OH data, OH source by HONO only estimated).
Line 68: Add reference by Tang et al., 2020.
Line 78: use “von der Heyden et al.” throughout the text.
Line 78: add references by Ren et al., 2011, Zhou et al., 2011, Laufs et al., 2017.
Line 168, equation-1: First, the units are missing for the factor 1/60 (min/s) and second, a factor 109 ng/g is missing. But I would recommend to delete all factors and simply give SI units for all terms (i.e. F in m3/s; M in ng/mol, P in Pa, R in J/mol/K).
Line 172: K-1
Line 186, Figure S1: The shown correlation is not “high” but may be fair (see scatter in Figure S1). In addition, the intercept will be zero by definition (J(NO2) = 0 during night…). And finally, a measure for the actinic flux (J(NO2)) will not linearly correlate with a measure of the irradiance (“solar radiation”). Here a non-linear correlation is expected since only for the irradiance the cos-dependence has to be considered (check with the TUV model). A quadratic dependence forced by zero will better describe the relation between both terms.
Line 193: Isn’t the soil particle density depending on the soil type and may be very different to the value of 2.65 g/cm3 in Linn and Doran, 1984? Please give uncertainty for the WFPS.
Line 199-200: For the typical applied extraction by unbuffered KCl solution underestimation of nitrite is well-known, see Homyak et al., 2015 (https://doi.org/10.2136/sssaj2015.02.0061n). Here the nitrite concentrations should be considered as lower limit.
Line 214: Better give an average daytime J(NO2)? The total campaign average will be otherwise restricted by the zero nighttime values.
Line 236-239: Fluxes measured in other studies are not only similar in comparison to the CK plots but also to all (CK and NP) PFP and LEP data. I.e. only the fluxes observed shortly after fertilization (HEP) with extremely high N-amount are much higher (see major concern 2).
Line 273: add references of Meng et al., 2022 and von der Heyden et al., 2022 with similar conclusions.
Line 277-280: First, also in Ren et al. (CALNEX data) and Laufs et al. soils were studied which are regularly fertilized. I.e. they are not “non-fertilized”. Second, the fluxes from the two studies are also in agreement with the present study, if the normal PFP and LEP data - even of fertilized soils - are considered. Thus, these low fluxes represent more typical conditions, while only during the short HEP period (few weeks) extremely high fluxes (two orders of magnitude higher) are observed for "over-fertilized" soils, see major concerns.
Line 290-293: Delete the references by Laufs et al, 2017; Meng et al, 2022; Ren et al., 2011; Sörgel et al., 2015; Zhou et al., 2011, since in these studied two orders of magnitude lower fluxes were determined.
Line 320: exchange order of NO2-/ H+ (acid-base…)
Section 3.5.2: If the temperature and humidity are the driving forces of the soil emissions, than the shape of the diurnal HONO flux profile should be similar with that of the temperature (and the inverse of the relative humidity). However, in most studies the HONO fluxes maximize around noon (see also Fig. 5A of the present study…), while the temperature shows a maximum in the afternoon (heat capacity of the soil, when heated up by the sun during daytime). Thus, at least under “normal” conditions, there must be other explanations for the HONO fluxes. Here I recommend that the authors also show a correlation of the normal diurnal flux data (Figure 5A) with the product of J(NO2) x [NO2] as done in other studies. I expect that the correlation is much better compared when using the temperature.
Only for extreme HEP conditions (see Figure 5B, representative for a few weeks per year for “over-fertilized” soils…) the diurnal profile of F(HONO) may well follow that of the temperature (see also the correlation results shown in Fig. 6 for HEP).
Line 357-359: This may be valid only for untypical HEP conditions. For the more typical PFP and LEP conditions, I expect a photochemical origin (s. above: check by plotting F(HONO) against J(NO2) x NO2).
Section 3.6: Re-evaluate the whole section after considering the non-linear relation between F(HONO) and fertilization amount.
Conclusion 1): Add a conclusion that the average F(HONO) data for all PFP and LEP conditions are in excellent agreement with most flux studies under normal fertilization conditions.
Conclusion 2): Add that these results only hold for extremely high nitrogen application during short periods after fertilization in China.
Conclusion 3): please do not use a constant value of EF(HONO), see above.
References:
General:
Order the references with same first authors chronologically (i.e. increasing year of publication), see VandenBoer et al., Wang (Y.) et al., Wu et al., Xue et al.. In addition, if an author has two publications in one year (a and b) use first the reference a) in the text.
Line 500, 580, 640, 716, 731: Pöschl,
Line 522, 688: Dubé
Line 526: Lörzer
Line 528: the doi link is not working?
Line 533-534 and others (644-647, 650-653, 660-661, 708-711, 723, 729, 780): please unify the style for Chinese given names and use only the first letter. E.g. Line 533: should be Li, D. and not Li, D. D.!
Line 537, 543, 645: Häseler
Line 544: Jäger
Line 557 and 594: Min, K.-E.
Line 579: …-Röser
Line 579, 715, 731: Sörgel
Line 654: subscript the “x” in ROx
Line 688: Öztürk
Line 703: 57(9), 3516-3526 missing
Line 710: Petäjä
Line 723: Huang, X.-R. Y.
Line 727: paper number e2021JD036379 missing
Line 729: Müller
Line 730: Fröhlich-…
Line 767: HOx and subscript the “x“
Line 772: use paper number L15820 (delete n/a-n/a)
Table 1: Here max fluxes are compared with the maximum in the average diurnal flux profile (see footnote d). I recommend to use the latter also for all other studies, since outliers are removed by this approach. Even better, one could simply compare average fluxes. In this case in all studies (except the OTDC fresh fertilization studies) the average fluxes would be in the range 0.1-2 ng N/m2/s…
Fig. 1: In Figure 1(B) I do not see items 10. and 11.? Please add to the figure. In addition, please specify in the methods section which type of pump (5.) is used (Teflon membrane pump?). Here additional artifacts may appear depending on the type of pump used, see major issue 1a)…
Figure 2. Here a very strong rain event (ca. 150 mm) is shown around the 9. September, while J(NO2) is very high at that day? In contrast, for the 19. September there is almost no light intensity (J(NO2)) but no rain? Please check the rain data. Here I find different rain periods in Figure S2? In addition, is there no J(NO2) data in the period 11-15. of July, or is it very low?
Figure 3: Unfortunately, after the PEP period there is a data gap, after which the HONO fluxes are again very low in the LEP period. However, if I would fit any Gaussian profile into the HEP flux data, I would get much higher HONO fluxes in the early LEP period. Was the soil anyhow treated directly after the HEP period? How do you explain the sudden step in the flux profile? Explain the terms NP and CK in the figure caption (a figure should stand alone).
Figure 4/5/6: dito for PFP, HEP and NP in figure caption 4/5/6.
Figure 6: Can you also show a plot against J(NO2) x [NO2]? Could be also in the supplement.
Figure 8: Show a new figure considering the non-linear relation between F(HONO) and fertilization amount.
Figure S3: Describe the arrows in the lower figure in the caption. E.g. "The arrows indicate reduced HONO fluxes after rain events."
Citation: https://doi.org/10.5194/egusphere-2023-1223-RC3 -
AC1: 'Reply on RC1, RC2, and RC3', Chaoyang Xue, 08 Oct 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1223/egusphere-2023-1223-AC1-supplement.pdf
-
AC1: 'Reply on RC1, RC2, and RC3', Chaoyang Xue, 08 Oct 2023
-
AC2: 'Revised Manscript with Changes Tracked', Chaoyang Xue, 08 Oct 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1223/egusphere-2023-1223-AC2-supplement.pdf
Peer review completion
Journal article(s) based on this preprint
Yifei Song et al.
Data sets
Measurement Report: Exchange Fluxes of HONO over Agricultural Fields in the North China Plain Song et al. https://zenodo.org/record/8115973
Yifei Song et al.
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 644 | 159 | 25 | 828 | 47 | 9 | 18 |
- HTML: 644
- PDF: 159
- XML: 25
- Total: 828
- Supplement: 47
- BibTeX: 9
- EndNote: 18
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1215 KB) - Metadata XML
-
Supplement
(239 KB) - BibTeX
- EndNote
- Final revised paper