the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 22 Jun 2026
Gravimetrically Measured Water Content of Filter Sampled Fine Particulate Matter at Low Relative Humidity: Insight from the Surface Particulate Matter Network (SPARTAN)
Abstract. Accurate measurements of the composition of fine particulate matter (PM2.5) are important for understanding its sources and health impacts. Water, a portion of PM2.5 mass, is difficult to measure. We describe developments to the Surface PARTiculate mAtter Network (SPARTAN) to better characterize the chemical composition of PM2.5 on polytetrafluoroethylene filters. A robotic weighing facility is used to estimate water content of PM2.5 sampled at three sites (Fajardo, Puerto Rico; Bujumbura, Burundi; Abu Dhabi, United Arab Emirates), and the chemical composition is used to attribute water content to three hygroscopic categories: high growth (sodium chloride), medium growth (ammonium-nitrate-sulfate-potassium), and low growth (organics). The growth rates are tested using two additional sites (Beijing, China; Halifax, Canada). The water content at 35 % relative humidity (RH) is estimated to be 16.5 % (11.6 %–26.5 %, 95 % confidence) for the high growth category, 3.9 % (3.0 %–4.2 %, 95 % confidence) for medium growth, and 1.0 % (0.6–1.3 %, 95 % confidence) for low growth. We calculate the average water content at 35 % RH for 2442 filters from 24 globally distributed sites from December 2019 to September 2024 to be 2.06 % (0.85 %–5.47 %, range), where 58 % of the aerosol water is associated with the medium growth category (22 % low growth, 19 % high growth).
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
(1393 KB) - Metadata XML
-
Supplement
(445 KB) - BibTeX
- EndNote
Status: final response (author comments only)
- EC1: 'Comment on egusphere-2026-3224', Mingjin Tang, 01 Jul 2026
-
RC1: 'Comment on egusphere-2026-3224', Anonymous Referee #1, 09 Jul 2026
Aerosol water content plays an important role on atmospheric chemistry, especially for liquid reactions, but its mass concentration is difficult to measure directly. This study combined robotic weighing method and mass balance calculation (based on chemical compositions) to gravimetrically estimate water content of PM2.5 at different sites. The authors found water content can be attributed to three hygroscopic categories of chemical compositions following the Wilson equation. Overall, this study presents a remarkably detailed gravimetric analysis of aerosol water content, and the results are valuable for further studies in aerosol chemistry. I only have a few minor personal concerns for the authors' consideration.
- In Introduction, could you provide a short comparison on present methods of estimating aerosol water content in the atmosphere, for examples, thermodynamics model and/or hygroscopicity measurement?
- The results suggested that water content can be attributed to three hygroscopic categories of chemical compositions following the Wilson equation at 35%-55% RH conditions. How about higher RH conditions (eg. 70%-80%)?
- Its better to add a brief summary on the atmospheric implications of the gravimetric analysis results of water content in this study, such as in mass concentration determination, source apportionment, visibility effect, etc..
Citation: https://doi.org/10.5194/egusphere-2026-3224-RC1 -
RC2: 'Comment on egusphere-2026-3224', Anonymous Referee #2, 12 Jul 2026
The manuscript by Oxford et al. presents a method for measuring water content of PM2.5 filter samples collected by the SPARTAN network. If successful, extending a global filter sampling network to quantify aerosol water content would provide valuable information for atmospheric aerosol research. However, in its current form, I believe that the manuscript has several fundamental technical issues that should be addressed before the methodology can be considered sufficiently validated for publication.
1) Lack of validation using standard chemical compounds
Validation using well-characterized reference compounds is a standard practice in hygroscopicity measurements because it demonstrates that the measured water uptake is physically meaningful and that the measurement system performs correctly. Measurements of atmospherically important compounds, such as ammonium sulfate, sodium chloride, and sucrose, are routinely conducted for this purpose, with ammonium sulfate being one of the most commonly used standards. I was unable to identify such validation measurements in the manuscript. Therefore, measurements of standard compounds should be conducted and compared with well-established literature data to demonstrate the validity of the measurement approach.
2) Limited RH range
According to the Introduction, the purpose of the study is to quantify the water content of PM2.5. However, the measurements were limited to a relatively narrow RH range (35–55%). This RH range does not cover the deliquescence relative humidity (DRH) of major atmospheric inorganic aerosol components such as ammonium sulfate and sodium chloride, where substantial water uptake occurs. Furthermore, multiphase chemical reactions that are important in the atmosphere are generally promoted under higher RH conditions. The reported mass growth factors (Figure 3) remain below approximately 1.05 throughout the investigated RH range, whereas ambient aerosol particles are frequently exposed to much higher RH in the atmosphere. Consequently, the atmospheric significance of the reported measurements remains unclear. Most hygroscopicity studies measure water uptake up to approximately 85–90% RH in order to characterize the full hygroscopic behavior of aerosol particles. If the authors believe that restricting the measurements to 35–55% RH is sufficient for the intended scientific objectives, a strong scientific justification should be provided. Otherwise, I suggest expanding the measurable RH range.
3) Insufficient technical description to demonstrate the validity of the methodology
The technical description is not sufficient for readers to evaluate the validity of the proposed measurement approach. For example, the manuscript does not adequately describe how RH was generated and controlled. In addition, information regarding RH stability, such as representative RH time series during equilibration and the standard deviation of RH, is not provided. Combined with the absence of validation using standard compounds, these omissions prevent the reader from assessing whether the proposed methodology accurately measures aerosol-associated water. Consequently, the technical validity of the methodology has not yet been demonstrated.
Because these issues require substantial additional experimental validation rather than clarification or minor revision of the existing manuscript, I do not believe that they can be adequately addressed through a normal revision process. Therefore, I cannot recommend publication of the manuscript in its current form. Resubmission should be considered after the authors have fully addressed these fundamental technical concerns.
Citation: https://doi.org/10.5194/egusphere-2026-3224-RC2 -
RC3: 'Comment on egusphere-2026-3224', Anonymous Referee #3, 15 Jul 2026
Review of Oxford et al. “Gravimetrically measured water content of filter sampled fine particulate matter at low relative humidity: Insight from the Surface Particulate Matter Network (SPARTAN)
This manuscript describes experimental studies to estimate aerosol water content at low relative humidity using filters obtained from the SPARTAN network. The aerosol water growth categories were separated into low, medium, and high, and the chemical composition of the filters was also split into three hygroscopic categories. The hygroscopic growth models were applied to other sites in the network to gain an understanding of the water content on the filters at low RH (~35%) at each site. The strengths of the study include the experimental design, especially accounting for evaporation of mass on the filter over time and exploring several water growth models to apply. However, it is necessary put the results into context of mass reconstruction. For example, the highest water content at 35% relative humidity was 5.5%. What is the typical uncertainty or closure in mass reconstruction? This water fraction is significantly lower than the “residual” fraction, which is the difference in reconstructed and measured mass and is assigned to organics. However, this residual also reflects and includes other sampling biases. In addition, the mass evaporated was in some cases similar in magnitude to the water content. To what degree can these results be used to improve reconstructed mass estimates when there is a large residual due to unmeasured mass species? I recommend addressing these issues before publication. Additional comments are below.
Line 73. Consider changing sentence to “… to develop a relationship of mass…”
Line 124. Differences in what across monitoring networks?
Line 130. Consider changing “water mass” to “water mass fraction” since the values in the parentheses are water fractions.
Line 134. What does “lacked mass resolution” mean in this context?
Line 141. Do the authors mean “used gravimetric measurements” ?
Line 142. Are these RH values laboratory or ambient?
Line 159. When did the SPARTAN network begin operation?
Line 170. Include “a” after “plus”
Line 191. How long of a period occurs between sampling and analysis?
Line 275. How are method detection limits determined?
Line 286. Why were FT-IR OC not used instead of assigning the residual to organics?
Line 317. This terminology is somewhat confusing. I needed to reread this several times to get the difference between a group and a subgroup.
Line 325. Figure 1 reads “average aggregate mass” but here there seem to be two separate things, “average net mass” and “aggregate net mass”. Confusing.
Line 328. At what RH were the filters equilibrated?
Line 333. “relative humidity” and “RH” are used interchangeably. I suggest “RH” since it’s already been defined.
Line 336. Table 1. What does “7(1st) 8(all others)” mean?
Line 342. Why 20% or 35%? Why not use one value? How is RH measured and what is its uncertainty?
Line 346/347. The “high, low, and medium growth samples” has not yet been defined.
Line 349. So all filters were weighed from a given site at once?
Line 358. Explain what is meant by factors like the ambient conditioning unit? Is the the room A/C or are you referring to something else?
Line 363. What is meant by “achieves independence of ascending or descending sections”?
Line 372. For which case is this being discussed?
Line 372. It appears in Figure 1b that the slope of the line (extrapolated into the higher RH range) is used, not the 5th weighing group?
Line 480, Table 4. Do these values refer to a given site?
Line 569. Why does the “No” (white) not show up on the outer circles?
Citation: https://doi.org/10.5194/egusphere-2026-3224-RC3
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 78 | 20 | 9 | 107 | 13 | 4 | 5 |
- HTML: 78
- PDF: 20
- XML: 9
- Total: 107
- Supplement: 13
- BibTeX: 4
- EndNote: 5
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Page 4, line 111-112: The authors stated that "Measurements of water mass, however, remain absent from filter-based PM2.5 measurements". This is not true, as the paragraph starting at line 125 provided an overview of previous studies which measured watter mass of filter samples.
Page 4-5, line 125-145: there are a few more studies which measured watter mass of filter samples. Please refer to Section 3.2 in a previous reveiw article (Tang et al., 2019) for more information.
References: Tang, M. J., Chan, C. K., Li, Y. J., Su, H., Ma, Q. X., Wu, Z. J., Zhang, G. H., Wang, Z., Ge, M. F., Hu, M., He, H., and Wang, X. M.: A review of experimental techniques for aerosol hygroscopicity studies, Atmos. Chem. Phys., 19, 12631-12686, 2019. [24]