Degradation of anhydro-saccharides and the driving factors in real atmospheric conditions: A cross-city study in China

Zhou, Biao; Zhang, Kun; Wang, Qiongqiong; Zhu, Jiqi; Li, Li; Yu, Jian Zhen

doi:10.5194/egusphere-2025-5481

Preprints

https://doi.org/10.5194/egusphere-2025-5481

Preprints

18 Nov 2025

| 18 Nov 2025

Status: this preprint is open for discussion and under review for Atmospheric Chemistry and Physics (ACP).

Degradation of anhydro-saccharides and the driving factors in real atmospheric conditions: A cross-city study in China

Biao Zhou, Kun Zhang, Qiongqiong Wang, Jiqi Zhu, Li Li, and Jian Zhen Yu

Abstract. Anhydro-saccharides, as important components of organic aerosol, have been widely used as molecular markers for biomass burning. Previous studies have shown that levoglucosan degrades in the atmosphere, but most of the results are derived from laboratory experiments, little is known about the decay rates and their driving factors in the real complex ambient environment. In this study, a TAG-GC/MS was utilized to collect PM_2.5-bound saccharides in three typical cities across the major city clusters in eastern China (Zibo, North China Plain; Changzhou, Yangtze River Delta; and Hong Kong, Pearl River Delta region) during the autumn and winter seasons, with bihourly time resolution. With the relative rate constant method, we found the daytime (8:00–16:00 LST) decay rate of levoglucosan was fastest in Changzhou, reaching 0.13 ± 0.05 h^-1 , and the maximum decay rates of mannosan (0.14 ± 0.05 h^-1) and galactosan (0.15 ± 0.06 h^-1) were observed in Hong Kong. Results from the generalized additive model indicate that the daytime decay rate of anhydro-saccharides is primarily influenced by aerosol liquid water content, relative humidity, and atmospheric oxidation capacity, while temperature and solar surface radiation also contribute to an increase in the decay rates. This study provides valuable field data on the degradation rates of saccharides in real ambient environments and demonstrates that their degradation results are derived by the combined effects of multiple oxidation pathways.

Received: 05 Nov 2025 – Discussion started: 18 Nov 2025

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.

Download & links

Preprint (PDF, 1267 KB)

Supplement (1120 KB)

Download & links

Biao Zhou, Kun Zhang, Qiongqiong Wang, Jiqi Zhu, Li Li, and Jian Zhen Yu

Status: open (until 03 Jan 2026)

Post a comment Subscribe to comment alert

RC1:
'Comment on egusphere-2025-5481', Anonymous Referee #1, 13 Dec 2025 reply
The manuscript investigates the atmospheric degradation of three anhydrosaccharides (levoglucosan, mannosan, and galactosan) using hourly TAG-GC/MS measurements conducted in three regions in China (Zibo, Changzhou, Hong Kong). Daytime decay rates are quantified using a relative rate constant method with K⁺ from biomass burning (K⁺BB) as a reference tracer, and the influence of environmental factors is examined using generalized additive models (GAM). This study provides observational evidence for the atmospheric instability of commonly used biomass-burning tracers, complementing previous laboratory and modeling work. The dataset is strong, and the multi-city comparison offers useful insights.
Overall, this work falls within the scope of ACP and could be suitable for publication after the authors address several issues related to methodological assumptions, uncertainty treatment, and interpretation to strengthen the scientific robustness of the conclusions.
A central concern is the use of K⁺BB as an inert, co-emitted tracer for decay-rate calculations. Potassium has significant contributions from dust and sea salt, which may potentially introduce substantial uncertainty. In addition, the relative rate method assumes stable emissions during the 8-hour daytime window, but this assumption may not be valid and is not supported by field observations. In addition, field decay rates are derived from linear regressions of ln(Cᵢ/Cₖ) vs. time, yet no statistical acceptance criteria, uncertainty estimates, or confidence intervals are provided. The manuscript repeatedly acknowledges atmospheric mixing but the analysis does not attempt to separate dilution-driven concentration changes from true chemical degradation. The absence of boundary layer height information further complicates interpretation. These issues need to be explicitly addressed to enhance confidence in the derived decay rates.
The GAM analysis is a major component of the paper, but the modeling framework is insufficiently described, Key information on smoother specifications, k values, multicollinearity diagnostics, and handling of missing data is not sufficiently described. In addition, pooling data from all three cities into a single GAM may be statistically inappropriate given the large regional differences.
Detailed comments
Lines 15–33 (Abstract): please report the number of valid decay days per city to better contextualize the analysis.

Lines 165–170: the manuscript acknowledges that K⁺ has contributions from sea salt and dust, which challenges the assumption of K⁺ being solely biomass-derived. Please provide quantitative justification for the use of K+BB, especially for dust-influenced Zibo and marine-influenced Hong Kong.

Lines 201–219: The description of the GAM lacks essential details: spline type, basis dimension (k), link function, variable scaling, and treatment of missing values. Without this information, the analysis is not reproducible.

Lines 250–284: When interpreting Lev/Man or Man/Gal ratios, consider discussing potential influences of regional differences in burning practices or fuel types.

Lines 258–264: The manuscript notes r = 0.63 correlation between total K and levoglucosan, but this alone does not establish total K as a reliable biomass-burning tracer. Please include comparison of total K variability with expected sea-salt potassium contributions and an assessment of potential bias in derived decay rate k if K⁺BB is misrepresented.

Lines 315–318: It is noted that concentrations depend on diffusion, meteorology, and deposition, yet decay-rate interpretation does not account for boundary layer effects.

Lines 360–364: It is stated that some days show negative decay rates due to fresh BB emissions, which contradicts the assumption that emissions remain constant during 8:00–16:00. Please clarify how such deviations influence the decay rate estimates?

Lines 365–367: it is reported that only a subset of days exhibit “good linear fitting,” please define explicit acceptance criteria (e.g., R² cutoff).

Lines 381–389: The conclusion that differences in BDE cannot fully explain decay rate differences, but the explanation is brief. Consider expanding the discussion to include potential roles of aerosol phase state, aqueous-phase versus heterogeneous oxidation pathways, and differences in aerosol composition (e.g., inorganic ions influencing aerosol liquid water content and viscosity).

Lines 426–468: The model pools all three cities, but their meteorology differs dramatically (lines 391–399). Consider using city-specific GAMs or including city as a factor.

Lines 430–431: The GAM response variable is decay rate k, which itself may embed meteorological dilution effects. Please clarify whether the environmental variables are intended to explain chemical degradation alone or a combination of chemical and physical processes.

Lines 432–433: An adjusted R² = 0.70 is reported but there is no discussion about multicollinearity. Ox, SSR, and T are correlated (Fig. 5), and such multicollinearity may distort smooth functions. Please provide concurvity diagnostics.

Lines 487–488: Diurnal patterns attributed to chemical decay could reflect dilution when BLH increases during the day. Including BLH data (e.g., from ERA5 reanalysis) would strengthen the interpretations.

Fig. 6: confidence bands need clear labeling.

Reply
Citation: https://doi.org/10.5194/egusphere-2025-5481-RC1
RC2: 'Comment on egusphere-2025-5481', Anonymous Referee #2, 13 Dec 2025 reply

This paper presents results from near real-time measurements of anhydrosugars made at 3 sites in China during autumn and winter. The data is used to examine the decay rate of the anhydrosugars at each site. A generalized additive model was employed to examine the parameters influencing the decay rate.

Overall, this is a good paper. As anhydrosugars are often used as biomass burning markers, understanding their decay in the atmosphere is important. But I do feel the authors don’t provide enough details on the differences between the sites, how that could play a role in the ratios observed/fuels burned/burn practices, and what separates a day with a linear decay rate to one without. Also, often what is being plotted is not accurately described in the figure caption. I have tried to note and make suggestions about these along with number of other items. These are all outlined below in the specific comments which need to be addressed before this paper can be considered for publication.

Specific Comments:
Line 64 – I am not sure what the phrase universal of this phenomenon means. Should it be universal impact of this phenomenon?

Line 85 – Suggest changing during cold season to during various cold seasons

Lines 94-95 – Suggest changing during autumn and winter season across three typical cities in the three regions to during multiple autumn and winter season deployments in three different typical cities in three regions

Line 103 – Suggest removing the comma after February

Line 117 – Suggest adding a the before Changzhou

Line 119 – Suggest adding an a before WXT520, and a before BAM1020, a the before beta-ray, and an and before NOx

Line 120 – Suggest adding an a before MODEL 49i, an and before MODEL 450i, and an a before RT-4

Line 121 – Suggest adding an an before ADI2080

Line 122 - Suggest removing the The before Solar

Line 126 – Suggest adding an a before MODEL and an and before NOx

Line 127 – Suggest adding an a before MODEL 49i, an and before MODEL 42i, a respectively before OC/EC, and an a before MODEL ECOC-610

Line 129 – Suggest adding an a before MODEL S611

Line 130 – Suggest adding a the before Hong Kong

Line 131 – Suggest adding an a before Model 5030i

Line 132 – Suggest adding an an before ADI2080

Line 134- Suggest adding an and before NOx

Lines 134-135 – The authors note in Hong Kong that the meteorological parameters, O3, and NOx were measured by AWS tower. Is that the name of the instrument or the location? It seems like that might be a location and the name of the instruments are missing.

Line 135 – Suggest adding an an before Xact

Line 136 – Suggest adding an an before X-ray

Line 139 – Suggest adding an A before Detailed and a the before TAG

Line 141 – Suggest changing observation to observations and adding an a before deuterium-labeled

Line 155 – The authors note here the species used in there ISOROPPIA runs to calculate pH and ALWC. But I don’t recall measurements of NH3 being mentioned in the site description and field observation section. Aren’t measurements of HNO3 also needed in the calculation?

Line 179 – The V after Kumar can be removed and a period should be added after et al as part of the citation

Line 188 – The citation should be written as Wang et al. (2025)

Lines 262-263 – The phrase while the total potassium (total K) data is relatively complete can be removed from the sentence as it has already been previously stated

Line 314 – Suggest removing the was after LST

Lines 320-322 – The authors mention that the ratio of levoglucosan to potassium can be an indicator of the aging degree of biomass burning. But levoglucosan and potassium don’t have to be correlated. If there is or isn’t a relationship it is often based on the type of fuel being burned and the type of burning. All the measurements were made in autumn and winter, but the sites have very different characteristics. So, couldn’t there be some regional influences on this? I think the authors allude to this in the previous section and Figure S3. While the data from all sites clumps together in Figure S3, it is also on a log scale, so it covers a large range of ratios. Although the sites are described in section 2.1, the authors don’t go into the regional differences other than mention of the impacts they see from the monsoon. It would be helpful to provide these additional details. Maybe it would also be helpful to the reader to present time series of levoglucosan, potassium, and temperature along with the diurnal profile as the diurnal pattern could be driven by “special” days.

Lines 324-326 – Suggest removing this sentence as it has already been previously stated

Line 327 – Detailed information on what can be found in the previously published paper?

Lines 328-329 – I am not sure what the authors mean here by calculated formula and the range of (2)-(7) being noted.

Line 331 – Suggest changing galactosan and K+ to galactosan with K+

Line 333 - Suggest changing galactosan and K+ to galactosan with K+

Line 334 – I am not sure what the authors mean here by calculation formulas

Figure 2
-In caption suggest adding the phrase (left column) after galactosan and (right column) after ratios
-It is surprising in plot a in the left column that galactosan is higher than mannosan. Is this correct? Do the authors know why this is for Zibo?

Line 339 – Suggest adding a the before three

Lines 362-364 – The authors note that not all days showed good linear fitting and this could be due to the direct emissions and transmission of biomass burning. Are there any other characteristics that could be different across the sites and various study periods? More warmer days observed? Burning practices the same at each location? Fuel type at each location? It would be helpful to provide more context for the reader.

Lines 365-367 – The authors note how many days at each site could be fit with a linear decay pattern. But they don’t really mention anything else about the days without this pattern. What is different about the days that don’t have a linear decay?

Line 389 – Suggest adding to the end of the sentence the phrase which we explore in the next section

Line 391 – Suggest change sampling points to sampling sites

Line 429 – I believe the reference is missing

Figure 5
-In caption suggest adding a the before three
-In caption the parameters O3 through T are denoted by letters, but they are actually all shown together on plot b

Line 459 – Suggest changing Fig. 6(a) show to Fig. 6(a) shows

Line 476 – mannosan and galactosan are misspelled

Figure 6
-In caption the labels noted do not match what are in the plots, suggest changing text from degradation rate of to (a) levoglucosan, (b) mannosan, and (c) galactosan analyzed as a function of ALWC, T, RH, O3, and SSR using the GAM model.

Lines 481-488 and Figure 7 – In this section the authors note and show that ALWC, RH, and Ox show a positive correlation with levoglucosan degradation rate from the factors tested. But aren’t T, SSR, and Ox naturally linked to begin with? Ozone formation is related to temperature and sunlight. In addition, RH and ALWC are not the exact same thing, but are quite similar. I guess I am not totally following how there can’t already be a relationship among these factors whether or not levoglucosan decays?

Line 488 – Should promoting be prominent?

Line 524 – Suggest changing substance to species

Lines 597- 598 – I believe the format of this reference is not the same as the others as it lists the authors first initial then last name

Supplemental Information
Line 51 – Suggest changing use the stock to used the stock and changing standard to standards. A period is also missing from the end of the sentence.

Line 53 – There should be commas before and after respectively. The chemical abbreviations used are not defined. Suggest adding a the before final working

Line 54 – Suggest changing are showed to that are shown

Line 56 – Suggest changing with range to ranging

Lines 56-57 – Suggest removing the phrase (ranging from 5 to 25 uL)

Table S2
-In the caption should the word identified be added after compounds?
-For the solvent and quantification IS are they the same for all 3 anhydrosugars?

Tables S4
-In caption suggest adding the phrase average +/- standard deviation of before hourly concentrations

Table S6
-In the caption the authors call these the smoothing function parameters, but I don’t believe they have been called that elsewhere in the text. Maybe it would be helpful to note what equation they refer to?
-Should it be SSR or SSRD? I believe throughout the text it was SSR.

Figure S3
-In caption the way it is written does not match what is actually plotted. Suggest changing correlation between K+bb and decay rates of levoglucosan, mannosan, and galactosan to correlation of levoglucosan, mannosan, and galactosan as a function of K+bb

Figure S4
-I believe the y-axes labels are not correct. It looks like the decay rate of levoglucosan is missing and the decay rate of galactosan is used twice.

Figure S5
-In caption the way part of it is written does not match what is actually plotted. Suggest changing predicted values vs. residuals to residuals vs. predicted values and predicted values vs. observed values to observed values vs. predicted values. Also suggest changing for GAM to for the GAM model

Reply

Citation: https://doi.org/10.5194/egusphere-2025-5481-RC2
RC3:
'Comment on egusphere-2025-5481', Anonymous Referee #3, 21 Dec 2025 reply
This work investigated the decay rates and their driving factors of anhydro-saccharides (levoglucosan, mannosan, and galactosan) at daytime at three different major city clusters in eastern China, with field measurements by TAG-GC/MS, rate calculation using the relative rate constant method, and factor contributions from the generalized additive model. The results are interesting and robust, highlighting the strong daytime decay and the dominant roles of aerosol liquid water, oxidants, temperature, and humidity. There are some minor comments which require to be addressed.
Line 47, nitro phenolic compounds, in particular nitro catechols, are also tracers for biomass burning, and thus can be included here.

Line 94, the same or three different TAG-GC/MS instruments were employed in three sampling sites?

Line 141, deuterium-labeled internal standards of the three targeted compounds?

Line 217, what explanatory variables were selected finally?

Line 282-284, in addition to industrial coal combustion, residential coal combustion for heating was also important emission source in suburban areas in North China in winter, which possibly corresponds to the low-temperature combustion at Zibo.

Line 322-324, the contributions of environmental factors to daytime decay of anhydro-saccharides also happened at nightime in theory. Why only significant declines of the ratio of levoglucosan/K⁺_BB were observed at daytime?

Line 399-401, what are the potential mechanisms for the complex relationship between decay rate and temperature?

Reply
Citation: https://doi.org/10.5194/egusphere-2025-5481-RC3

Biao Zhou, Kun Zhang, Qiongqiong Wang, Jiqi Zhu, Li Li, and Jian Zhen Yu

Supplement

https://doi.org/10.5194/egusphere-2025-5481-supplement

Biao Zhou, Kun Zhang, Qiongqiong Wang, Jiqi Zhu, Li Li, and Jian Zhen Yu

Viewed

Total article views: 218 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
160	34	24	218	25	18	13

HTML: 160
PDF: 34
XML: 24
Total: 218
Supplement: 25
BibTeX: 18
EndNote: 13

Views and downloads (calculated since 18 Nov 2025)

Month	HTML	PDF	XML	Total
Nov 2025	81	14	12	107
Dec 2025	79	20	12	111

Cumulative views and downloads (calculated since 18 Nov 2025)

Month	HTML	PDF	XML	Total
Nov 2025	81	14	12	107
Dec 2025	79	20	12	111

Viewed (geographical distribution)

Total article views: 266 (including HTML, PDF, and XML) Thereof 266 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 26 Dec 2025

Short summary

Anhydro-saccharides are key organic markers for biomass burning, but their decay rates and driving factors in real ambient environments remain unclear. This study conducted an online field measurement of PM₂.₅-bound saccharides in three eastern Chinese cities. Daytime levoglucosan decay rates were calculated, and the driving factors were investigated using a machine learning model. Findings of this study offers a strong scientific basis to support air pollution management.


Total:	0
HTML:	0
PDF:	0
XML:	0