From the journal Environmental Science: Atmospheres Peer review history

28-Jul-2023

Dear Dr Volckens:

Manuscript ID: EA-ART-06-2023-000086
TITLE: A national crowdsourced network of low-cost fine particulate matter and aerosol optical depth monitors: Results from the 2021 wildfire season in the United States

Thank you for your submission to Environmental Science: Atmospheres, published by the Royal Society of Chemistry. I sent your manuscript to reviewers and I have now received their reports which are copied below.

I have carefully evaluated your manuscript and the reviewers’ reports, and the reports indicate that revisions are necessary.

Please submit a revised manuscript which addresses all of the reviewers’ comments. Further peer review of your revised manuscript may be needed. When you submit your revised manuscript please include a point by point response to the reviewers’ comments and highlight the changes you have made. Full details of the files you need to submit are listed at the end of this email.

Please submit your revised manuscript as soon as possible using this link:

*** PLEASE NOTE: This is a two-step process. After clicking on the link, you will be directed to a webpage to confirm. ***

https://mc.manuscriptcentral.com/esatmos?link_removed

(This link goes straight to your account, without the need to log on to the system. For your account security you should not share this link with others.)

Alternatively, you can login to your account (https://mc.manuscriptcentral.com/esatmos) where you will need your case-sensitive USER ID and password.

You should submit your revised manuscript as soon as possible; please note you will receive a series of automatic reminders. If your revisions will take a significant length of time, please contact me. If I do not hear from you, I may withdraw your manuscript from consideration and you will have to resubmit. Any resubmission will receive a new submission date.

The Royal Society of Chemistry requires all submitting authors to provide their ORCID iD when they submit a revised manuscript. This is quick and easy to do as part of the revised manuscript submission process. We will publish this information with the article, and you may choose to have your ORCID record updated automatically with details of the publication.

Please also encourage your co-authors to sign up for their own ORCID account and associate it with their account on our manuscript submission system. For further information see: https://www.rsc.org/journals-books-databases/journal-authors-reviewers/processes-policies/#attribution-id

Environmental Science: Atmospheres strongly encourages authors of research articles to include an ‘Author contributions’ section in their manuscript, for publication in the final article. This should appear immediately above the ‘Conflict of interest’ and ‘Acknowledgement’ sections. I strongly recommend you use CRediT (the Contributor Roles Taxonomy, https://credit.niso.org/) for standardised contribution descriptions. All authors should have agreed to their individual contributions ahead of submission and these should accurately reflect contributions to the work. Please refer to our general author guidelines https://www.rsc.org/journals-books-databases/author-and-reviewer-hub/authors-information/responsibilities/ for more information.

I look forward to receiving your revised manuscript.

Yours sincerely,
Dr Lin Wang
Associate Editor, Environmental Science: Atmospheres

Environmental Science: Atmospheres is accompanied by companion journals Environmental Science: Nano, Environmental Science: Processes and Impacts, and Environmental Science: Water Research; publishing high-impact work across all aspects of environmental science and engineering. Find out more at: http://rsc.li/envsci

************

Reviewer 1

This is an excellent paper testing whether devices created in the laboratory could be used by homeowners to monitor optical depth and also PM2.5 as measured by Plantower PMS 5003 sensors. In general, the results were promising, with about 25% of the data lost. The use of gravimetric measurements of PM2.5 along with the PPMS 5003 measurement was a great strong point of the study. I had difficulty understanding which of the existing algorithms (CF_1, CF_ATM, the EPA (Barkjohn) model, or the alternative model "pm2.5 alt" was actually used here for the PMS 5003 measurement. Also wondered how the gravimetric measurement compared with the PMS 5003 sensor measurement. It was stated that "filter-corrected" values were used, but no indication of how the filter correction might have varied over location, time, etc.

I also wondered how their results might have related to the Ouimette et al., (AMT) paper claiming that the PurpleAir monitor (which uses two PMS 5003 sensors) acts as a nephelometer and therefore the first channel (which contains all particles measured >0.3 um in diameter) can be related to optical depth. Also whether the PurpleAir use of eta - 0.15 (based on Ouimette) matches up to these findings.

Comments in attached file

Reviewer 2

Review of “A national crowdsourced network of low-cost fine particulate matter and aerosol optical depth monitors: Results from the 2021 wildfire season in the United States” by Wendt et al.
The authors examined the PM2.5 and AOD relationship during wildfire season in the US, which is enabled by a national network sample of low-cost monitors. After validating the monitor samples, they demonstrated that the measured AOD from their developed monitors is generally consistent with AERONET records. More importantly, the authors pointed out the PM2.5:AOD ratio decreased strongly under smoke days in California, while the ratio were steady across non-smoke days and smoke days.
Overall, this is an interesting study and the devised low-cost monitors show reliable capability to observe surface PM2.5 and AOD column. This provides a valuable venue to study AQ affected by wildfires. Although this manuscript is well written, there are still several places that need to be further clarified. I am happy to see its publication after the authors have addressed the following concerns.
1.In Fig.4, how about the correlation between CEAMS and AERONET AOD under smoke and non-smoke days, respectively?
2. Moreover, is there any possible to show the time series between this two AOD records at some sites? This will be more powerful to justify the capability of CEAMS instruments.
3.The last paragraph in Section 3.2 is duplicated from the sentences at the end of Section 3.1.
4. The authors mentioned “This could point to shallower boundary layers for our sampling locations that”. Is there any reference or evidence for a shallower boundary layer?
6. In Fig.6, the authors show “we observed similar week-to-week PM2.5 concentrations”. I am very interesting whether this pattern has been affected by regional wind circulation. I didn’t find any meteorology analysis to explain this discrepancy between California and other regions.

Reviewer 3

Review of EA-ART-06-2023-000086 by Wendt et al.
This manuscript presents novel efforts to correlate and disentangle surface PM2.5 and AOD data for the purposes of improving PM2.5 exposure assessment and satellite data, with an emphasis on wildfire smoke. The methods and QA discussions are well written, and the novel devices used in this work show good correlation with AERONET AOD.
A major finding of this work is the demonstration of regional differences in PM2.5:AOD ratio and smoke PM2.5 impacts in CA.
The work’s interpretations of potential causes should be qualified as supported by limited evidence for specific days and locations, but are valuable as hypotheses. In particular, the hypothesis that smoke aloft may yield increased AOD but no net changes in surface PM2.5 may be useful for informing smoke guidance from public health agencies. This is a common point of confusion for the public when they see dark skies and assume their breathing zones are similarly impacted.
It should be noted that in CA, smoke events are often accompanied by unusual shifts in wind direction, and so PM2.5 on a smoky day will not necessarily equal A + B, i.e. the sum of the smoke (A) plus the typical non-smoke PM2.5 (B). Instead, it may correspond to a totally different background (A + C) that may be low and poorly characterized.
Equally important is the finding that CA non-smoke days possessed an unusually high PM2.5:AOD ratio. Any additional interpretations or need for future work should be emphasized.

Specific Comments
Abstract – “However, in California, median PM2.5 remained similar on smoky days relative to non-smoky days, while AOD increased, implying that the smoke was lofted above the surface.” The authors should consider instead “at the California study sites” and “may sometimes have been lofted,” or otherwise constrain the language to avoid statements about CA in general. As the authors cite in other studies, some CA locations frequently report elevated ground PM2.5 during smoke episodes.
Abstract – “In California, the median PM2.5:AOD ratio was 67.2 µg m-3 on non-smoky days” Adding another summary value for non-smoky ratios in other US regions is needed to emphasize how unusual this value is.
Introduction p.2 – Beta attenuation monitors should be included in the AQS monitors description along with gravimetric and optical, as many CA sites employ BAMs as AQS FEM.
Section 3.2, p.5, second column – The two sentences starting with “These simultaneous increases…” refer to Fig. 5, not Fig. 4, correct?
Section 3.2, p.5 – “Also of note, is that the non-smoky days in California had a higher ratio…” Please add comparison non-smoky PM2.5:AOD ratios from other US regions to this sentence.
Abstract and Section 3.2, p.5-6 – The hypothesis that CA PM aloft on smoke days reduced PM2.5 is compelling, and the limited data shown in Figures S9 and S10 do suggest aloft PM in specific sites on specific days. However, the other presented evidence is somewhat inconsistent: Figure 7 suggests PM2.5:AOD ratios skew visually low on CA smoke days compared to other regions, but the analogous 25%-median-75% ratio data in Table 1 suggest comparable smoky medians and skews across regions; can this apparent discrepancy be clarified? In addition, CA PM2.5 might not be expected to go up on smoke days in a predictable way relative to other regions, given that unusual background PM2.5 and wind patterns (e.g. easterly rather than westerly) may occur on such days. In general, the statement that “smoke was lofted” in the Abstract should be qualified as a hypothesis or a possibility (“may have been lofted”).
Section 3.2, p.5-6 – The explanation of shallow boundary layers causing unusually high CA PM25:AOD ratios on non-smoky days is plausible, though another explanation could be non-smoke PM types more prevalent in CA. The relatively high Angstrom exponents on many non-smoke days in Figs. S6 and S7 may refelct significant presence of submicron PM such as vehicle emissions, secondary aerosols, or fine sea salt, the latter of which is also possibly consistent with the low observed AOD.
Fig 6, S6, S7 – The utility of these figures would be improved if all three had the same vertical axes to help interpret the magnitudes of the variations. Optional: in my opinion the Angstrom exponent portion is valuable, so Fig.6 could be replaced with Fig. S6. By eliminating one of these figures, the authors could add a similar figure for a smoke episode from another US region (again with the same Y-axes), which may aid in interpretation.
Section 3.4, p.9 – Was any of the previous AMOD gravimetric or continuous PM performance testing conducted in these CA regions? If not, could any unique CA particle types or PM volatility have contributed to the CA PM2.5 or ratio anomalies observed in this study? This could be added to the Limitations.
Figure S2 – Please clarify the interpretation of this figure in the text. If a linear decreasing trend between median correction factor and number of smoke days is meant to be inferred, please provide a suggested explanation. Forcing 1 to appear in the Y-axis would be helpful.
Figure S9 – Please increase the font size of the Y-axis (altitude) and subtype definitions, or else move the subtype definitions to the caption.

REVIEWER REPORT(S):
Referee: 1

Comments to the Author
This is an excellent paper testing whether devices created in the laboratory could be used by homeowners to monitor optical depth and also PM2.5 as measured by Plantower PMS 5003 sensors. In general, the results were promising, with about 25% of the data lost. The use of gravimetric measurements of PM2.5 along with the PPMS 5003 measurement was a great strong point of the study. I had difficulty understanding which of the existing algorithms (CF_1, CF_ATM, the EPA (Barkjohn) model, or the alternative model "pm2.5 alt" was actually used here for the PMS 5003 measurement. Also wondered how the gravimetric measurement compared with the PMS 5003 sensor measurement. It was stated that "filter-corrected" values were used, but no indication of how the filter correction might have varied over location, time, etc.
-
Thank you for your positive review! We used the CF_1 (which we now state in the text) and then corrected it using the ratio between the average mass concentration from the filter and the average mass concentration measured by the Plantower. We discussed this method more in previous papers (e.g., Ford et al., 2019); thus the comparison is not included in detail in the main text and is instead in the supplement in Figures S2 and S3. There is variability in the filter correction by location and time. We found that, in general, sample periods with more smoke days had slightly lower ratios, often under 1. This could be due to the aerosol type, but we see similar results if we bin by mass (ie, smoke days generally had higher concentrations; thus, more smoke days mean more mass). This substantiates previous work that shows that Plantowers have a low bias at low concentrations (often reporting zero, as the reviewer notes) and a high bias at high concentrations.


I also wondered how their results might have related to the Ouimette et al., (AMT) paper claiming that the PurpleAir monitor (which uses two PMS 5003 sensors) acts as a nephelometer and therefore the first channel (which contains all particles measured >0.3 um in diameter) can be related to optical depth. Also whether the PurpleAir use of eta - 0.15 (based on Ouimette) matches up to these findings.

Thank you for the comment. We have a manuscript under review (in collaboration with Ouimette et al.) that demonstrates that the Plantower PMS5003 sensor is not a nephelometer but instead an imperfect optical particle counter. Our model and measurement data show that the Plantower is good at estimating the number concentration of 0.3 and 0.4 μm particles - counts in this size range make up the majority of the sensor’s “CF1” signal - and this signal correlates very well with the scattering coefficient of ambient PM2.5 (as measured by a true nephelometer). In other words, the PMS5003 correlates very well to a nephelometer for a typical accumulation mode aerosol, but the device is not a nephelometer. Because of these results, we do not recommend the use of “eta - 0.15” with the PurpleAir. We hope this manuscript will be published soon.

Comments in attached file
LW1 comment to “2.5 = ⋅ ” : Have you considered the claim in Ouimette (AMT, 2021) that the PurpleAir monitor acts more as a nephelometer than an optical particle counter? (I presume you have, since they are a related group at Colo State. Ouimette et al find that the PurpleAir first channel (>0.3 um) (i.e., all the particles they could count) was correlated with scattering coefficients as measured by the TS3563 nephelometer, with an eta of 0.15. So the PurpleAir API site uses that value to estimate scattering coefficients and visibility, visibility range, etc.
See previous comment for response to Ouimette et al. paper.
Regarding this equation, it is used widely in the satellite-derived PM2.5 field of study (e.g., van Donkelaar et al., 2007; 2013; 2015; 2019; Liu et al., 2005) and is not a reference to PurpleAir/Plantower data. It is a method used to relate the satellite AOD for the full atmospheric column to a surface concentration. Eta is not a fixed value; it varies by time and location (in general, eta is obtained by chemical transport model simulations and then constrained by lidar profiles and/or co-located PM2.5/AOD measurements at SPARTAN sites).


LW2 comment to “However, Plantower PMS5003 sensors are known to exhibit relatively high measurement bias, requiring calibration and field correction relative to reference methods.”: While this is true, except possibly for Tryner ref 45, all such estimates rely on the “proprietary” algorithm CF_1, or worse, CF_ATM. Yet both these algorithms return values of zero, particularly at low concentrations, when the number of particles > 0.3 um is never zero. A better algorithm is “pm2.5 alt,” available on the PA API site. No zeros, and no longer the 60% overestimate of the Plantower algorithms. 1.Bi, J., Wallace, L., Sarnat, J.A. and Liu, Y. (2021). Characterizing outdoor infiltration and indoor contribution of PM2.5 with citizen based low-cost monitoring data. Environmental Pollution 276:116793. https://pubmed.ncbi.nlm.nih.gov/33631689/ 2.Wallace, L., Bi, J., Ott, W.R., Sarnat, J.A. and Liu, Y. (2021) Calibration of low-cost PurpleAir outdoor monitors using an improved method of calculating PM2.5. Atmospheric Environment, 256 (2021) 118432. https//doi.org/10.1016/j.atmosenviron.2021.11843

We have changed this sentence to “However, Plantower PMS5003 sensors are known to exhibit high measurement bias at high concentrations and low bias at low concentrations, requiring calibration and field correction relative to reference methods.45,57–65” We have added the suggested references to the text.

LW3 comment to “ To date, crowdsourced deployments of low-cost aerosol pollution monitors have typically been limited to one measurement modality (PM2.5 or AOD).”: If Ouimette is right, PA measures some form of AOD. But he denies that PA is any good for measuring concentrations at specific sizes, as I think your group has shown to some extent, at least in showing the PM10 estimates are worthless. The fact remains that PA can do well in estimating PM2.5 by comparison with FEM/FRM sites, as shown by Barkjohn and others in your list of papers. 1.Wallace, L.A., Zhao, T., Klepeis, N.R. 2022 Indoor contribution to PM2.5 exposure using all PurpleAir sites in Washington, Oregon, and California. Indoor Air 32: (9) 13105 2.Wallace, L.; Zhao, T. 2023. Spatial Variation of PM2.5 Indoors and Outdoors: Results from 261 Regulatory Monitors Compared to 14,000 LowCost Monitors in Three Western States over 4.7 Years. Sensors 23, 4387. https://doi.org/10.3390/s23094387
See response #1 for comments about Ouimette et al. and the discussion on measuring concentrations. Additionally, if the Plantower is actually more of a nephelometer, it would be taking a direct measurement and could be used for determining an AOT, but it is not measuring a full atmospheric column AOD. The Globe sun photometers are a low-cost monitor that measures full atmospheric column AOD but do not measure PM2.5. Thus, there are no low-cost monitors that directly measure both PM2.5 mass (we have a filter measurement in addition to the Plantower) and a full atmospheric column AOD. We have changed the sentence to stress that we are referring to a full atmospheric column AOD measurement. We have also added these references to the paper.

LW4 comment to “to each PMS5003 PM2.5 measurement”: But “each PM2.5 measurement” is a single 2-minute average. You must mean you compared the filter amount to the 96-h average of the PM2.5 measurements.
We calculate the 96-h average from the 2-minute averages and then create a ratio to the 96-hr filter measurement, which we then apply to scale the 2-minute average. We have rewritten the sentence to clarify this.
“...we applied a filter correction to each PMS5003 CF1 PM2.5 measurement. We scaled each 2-minute PMS5003 PM2.5 measurement by the ratio between the corresponding 96-hour time-weighted average PM2.5 concentration measured via the filter sample and the 96-hour time-weighted average of all the Plantower PMS5003 PM2.5 measurements during the sampling period. Thus, the correction factor can be different for different locations/devices and for different sample periods but is a consistent scale factor per device sampling period.”


LW5 comment to “We scaled each PMS5003 PM2.5 measurements”: Did you state anywhere in this paper whether you used the CF_1 or CF_ATM or other (EPA Barkjohn model). The first two algorithms would result in overestimates. The EPA algorithm fixes that but has no ability to correct the various zeros produced by CF_1. Too late I suppose for this paper, but for future work, you might consider checking out the pm2.5 alt algorithm. There is a recent paper claiming that the CF_1 algorithm actually looks like PM2.5 = a(N1+N2) + c N3 +d, where a, c, and d can be determined easily from linear least-squares estimation. D turns out to be negative, on the order of -1 ug/m3, and the number of negatives produced by the model match up with the number of zeros reported by CF_1. (Wallace, 2023). 1.Wallace, L. (2023). Cracking the code— Matching a proprietary algorithm for a lowcost sensor measuring PM1 and PM2.5. Science of the Total Environment 893.164874. https://authors.elsevier.com/sd/articl e/S0048-9697(23)03497-6
Thanks for pointing us to your new paper; we have now added it as a reference. CF_1 and CF_ATM overestimate at high concentrations but underestimate at low concentrations (as the reviewer notes, it often reports zero). The Barkjohn EPA algorithm does attempt to fix the zeros because it adds an offset. All studies find that there is a bias, but they suggest different ways to correct the bias and provide different results under different conditions. We use the CF_1 values which we corrected using our filter measurements as we have done in the past (note, that we are not using a PurpleAir with two channels, just a Plantower).

“As previously mentioned, the Plantower PMS5003 is known to exhibit a high bias at high concentrations and a low bias at low concentrations. There are several different methods that are used to attempt to correct the bias (e.g.,62-64, 69).”

LW6 comment to “s. The mean ratio of the filter PM2.5 to Plantower PMS5003 PM2.5 (i.e., PMS5003 scaling factor) was 1.7”: Wait—I thought the CF_1 and CF_ATM algorithms overestimated PM2.5, as shown by all of your citations to Barkjohn, Kelly, Sayahi, etc. How does it happen that now they are underestimating? I can see from Figure S2 that indeed the ratio is >1 (although I would have appreciated filling in the values for every integer on the y-axis, which you have space to do.) It seems to me you need to state why it is that your result is so different from all other studies.
We correct our earlier statement to “has a high bias at high concentrations and a low bias at low concentrations”. Our concentrations were, in general, low (as shown by the distribution in Figure 3a) except during the wildfire smoke. As you mention earlier, there are a lot of 0s that are output that bring the average down at low concentrations. When there was more smoke, the ratio was <1 (Figures S2 and S3), suggesting that the Plantower had a high bias at higher concentrations. Thus, our results do not differ from previous papers. We have added the following to the text:

“This value of 1.7 suggests a low bias on average. However, this low bias is to be expected because concentrations were generally low across the whole campaign as shown in the PM2.5 distribution in Figure 3a. Box and whisker plots of these PMS5003 scaling factors as a function of the number of smoke-impacted days are provided in Fig. S2, and these correction-factor distributions for each individual AMODv2 device are provided in Fig. S3. As shown in these figures, scaling factors were less than one for sampling periods with more smoke days and higher concentrations, corroborating previous work showing a high bias at high concentrations (e.g., 45,57–62).”

LW 7 comment to “s. Thus, our results from the Midwest, Mountain West, and Northeast regions provide empirical support for source independent conversions from AOD to ground-level PM2.5 when smoke is not aloft; while our California sites suggest that smoke aloft can introduce source-dependent disparities, which would preclude the use of source-independent conversions of satellite AOD to ground level PM2.5 concentrations.” The data presently exists in about 20000 PurpleAir outdoor sites in the US to test the Ouimette et al hypothesis that the first channel (# particles >0.3 um) provides an estimate of AOD. Could you just compare measured vertical AOD to the existing PurpleAir estimate using eta = 0.15?
We do not entirely understand the reviewer’s comment here. We are not using PurpleAirs in this paper and do not have the API credits to download that much data (bring back the Thingspeak PurpleAir API!). One could possibly estimate the AOT of the surface layer from PM2.5 measurements, but not the full atmospheric column AOD from the Plantower PM2.5 measurements. One of our main findings here is that eta (the PM2.5/AOD ratio) has a lot of variability (regionally and temporally); thus, we know that using a consistent eta value to translate between PM2.5 and AOD would lead to very poor agreement. This is consistent with previous work on satellite-derived PM2.5 estimates that has shown that column AOD measurements and surface measurements are not always well-correlated (e.g., Ford and Heald, 2016) and that calculated values measured by more expensive instruments show significant temporal and spatial variability (e.g., Snider et al., 2015). Further, our goal here is not to estimate AOD, but discuss how more co-located ground-based measurements of PM2.5 and AOD can be used to gain insight that would allow satellite data to be better translated into information on surface air quality (especially in regions without any monitoring, although admittedly, those regions are shrinking with the prevalence of so many low-cost monitors!).

B. Ford and C. L. Heald, Exploring the uncertainty associated with satellite-based estimates of premature mortality due to exposure to fine particulate matter, Atmos Chem Phys, 2016, 16, 3499–3523.
G. Snider, C. L. Weagle, R. V. Martin, A. van Donkelaar, K. Conrad, D. Cunningham, C. Gordon, M. Zwicker, C. Akoshile, P. Artaxo, N. X. Anh, J. Brook, J. Dong, R. M. Garland, R. Greenwald, D. Griffith, K. He, B. N. Holben, R. Kahn, I. Koren, N. Lagrosas, P. Lestari, Z. Ma, J. Vanderlei Martins, E. J. Quel, Y. Rudich, A. Salam, S. N. Tripathi, C. Yu, Q. Zhang, Y. Zhang, M. Brauer, A. Cohen, M. D. Gibson and Y. Liu, SPARTAN: a global network to evaluate and enhance satellite-based estimates of ground-level particulate matter for global health applications, Atmos Meas Tech Discuss, 2015, 7, 7569–7611.
Referee: 2

Comments to the Author
Review of “A national crowdsourced network of low-cost fine particulate matter and aerosol optical depth monitors: Results from the 2021 wildfire season in the United States” by Wendt et al.
The authors examined the PM2.5 and AOD relationship during wildfire season in the US, which is enabled by a national network sample of low-cost monitors. After validating the monitor samples, they demonstrated that the measured AOD from their developed monitors is generally consistent with AERONET records. More importantly, the authors pointed out the PM2.5:AOD ratio decreased strongly under smoke days in California, while the ratio were steady across non-smoke days and smoke days.
Overall, this is an interesting study and the devised low-cost monitors show reliable capability to observe surface PM2.5 and AOD column. This provides a valuable venue to study AQ affected by wildfires. Although this manuscript is well written, there are still several places that need to be further clarified. I am happy to see its publication after the authors have addressed the following concerns.


1.In Fig.4, how about the correlation between CEAMS and AERONET AOD under smoke and non-smoke days, respectively?
We have added a table in the supplement with these values. Correlations were higher for smoke days for 440, 500, and 675 nm wavelengths but there were more smoke days (88) than non-smoke days (52). For the 870 nm wavelength, correlations were similar but there were overall fewer observations (32 no smoke and 41 smoke days).
Table S1. Correlations and number of observations (N) for each wavelength by smoke or no smoke day designation. Table is a companion to Figure 4 in the main text.
440 nm 500 nm 675 nm 870 nm
R2 N R2 N R2 N R2 N
No Smoke 0.94 88 0.90 88 0.79 88 0.88 32
Smoke 0.99 52 0.99 52 0.98 52 0.85 41

2. Moreover, is there any way to show the time series between these two AOD records at some sites? This will be more powerful to justify the capability of CEAMS instruments.
We have added two time series plots to the supplement which show the AOD measured by the AMOD hosted by the SARP participant and the nearby AERONET site.
This plot is for AMOD AD0323 and the Bozeman AERONET site. The time series has been sampled to times when both locations had valid AOD550nm measurements.

This plot is for AMOD AD0007 and the Georgia Tech AERONET site. The time series has been sampled to times when both locations had valid AOD550nm measurements.


3.The last paragraph in Section 3.2 is duplicated from the sentences at the end of Section 3.1.
Thank you for catching this. We have removed this section from 3.2.

4. The authors mentioned “This could point to shallower boundary layers for our sampling locations that”. Is there any reference or evidence for a shallower boundary layer?
We now cite Ayazpour et al. (2023), which does seem to suggest that the locations in California have lower planetary boundary layer heights (Figure 6 in the paper), although it is somewhat dependent on the product/method/definition used (ie, we had started by doing analysis on this using NARR data and did not see much evidence but several studies including the the Ayazpour et al. (2023) and McGrath-Spangler and Denning (2012) suggest that NARR PBLH in the West are too high). Ford and Heald (2012) also show that most of the columnar AOD is within the boundary layer for our southern California sites. Additionally, for smoke days, it is not really the height of the PBL that matters as much as the height in comparison to the smoke plume. Our previous work in Cheeseman et al., 2020 showed that California has high plume heights and frequently has smoke plume heights above the boundary layer (see Figures S4 and S5 in the supplement of the manuscript specifically).
“This could point to shallower boundary layers for our sampling locations in California that, coupled with potentially higher emission sources, may also explain the higher PM2.5 concentrations on non-smoky days compared to the other regions. As shown in 77, summertime planetary boundary layer heights along the California coast do appear to be shallower than many other regions across the CONUS. However, there is variability in the planetary boundary layer heights among different products with the greatest differences in the western US77,78. In addition to differences in the boundary layer, there may be differences in aerosol composition and size due to the mixture of different emission sources at the sites in California.”

Ayazpour, Z., Tao, S., Li, D., Scarino, A. J., Kuehn, R. E., and Sun, K.: Estimates of the spatially complete, observational-data-driven planetary boundary layer height over the contiguous United States, Atmos. Meas. Tech., 16, 563–580, https://doi.org/10.5194/amt-16-563-2023, 2023.
Cheeseman, M., Ford, B., Volckens, J., Lyapustin, A., & Pierce, J. R. (2020). The relationship between MAIAC smoke plume heights and surface PM. Geophysical Research Letters, 47, e2020GL088949. https://doi.org/10.1029/2020GL088949.
McGrath-Spangler, E. and Denning, A. S.: Estimates of North American summertime planetary boundary layer depths derived from space-borne lidar, J. Geophys. Res.-Atmos., 117, D15, https://doi.org/10.1029/2012JD017615, 2012.

6. In Fig.6, the authors show “we observed similar week-to-week PM2.5 concentrations”. I am very interesting whether this pattern has been affected by regional wind circulation. I didn’t find any meteorology analysis to explain this discrepancy between California and other regions.
Referee 3 also brings this up. We definitely think the winds can impact the PM2.5 concentrations. The sites in California are often impacted by sea breeze circulations, which can bring clean air or recirculate polluted air. We tried looking at some comparisons between smoky days and non-smoky days in California (example below). The wind directions (from NARR) are only slightly different, with slightly calmer winds in the morning on smoky days and slightly shallower planetary boundary layers (in the figure below, purple dots are the sampling locations, the top row is the 20th which was not a smoke day, the middle row is the 23rd which was a smoke day for the northern site, and the bottom row is for the 24th which was a smoke day for the southern sites).

There is a lot more analysis that could be done, looking at diurnal variability and multiple smoke days. We thus have added more acknowledgement that this is a limitation of the paper and further analysis of local meteorology would be useful for future work.
“These differences could be explained based on prior work, which found that an appreciable fraction of smoke plumes in California are injected into the free troposphere30 or could be due to changes in local meteorology on smoky days. Shifting wind patterns could alter not only the smoke concentration, but also the background concentration. Future work should further investigate the role of local meteorology on altering the composition and the vertical distribution of pollutants during smoke events.”

Referee: 3

Comments to the Author
Review of EA-ART-06-2023-000086 by Wendt et al.
This manuscript presents novel efforts to correlate and disentangle surface PM2.5 and AOD data for the purposes of improving PM2.5 exposure assessment and satellite data, with an emphasis on wildfire smoke. The methods and QA discussions are well written, and the novel devices used in this work show good correlation with AERONET AOD.
A major finding of this work is the demonstration of regional differences in PM2.5:AOD ratio and smoke PM2.5 impacts in CA.
We thank the reviewer for their comments.

The work’s interpretations of potential causes should be qualified as supported by limited evidence for specific days and locations, but are valuable as hypotheses. In particular, the hypothesis that smoke aloft may yield increased AOD but no net changes in surface PM2.5 may be useful for informing smoke guidance from public health agencies. This is a common point of confusion for the public when they see dark skies and assume their breathing zones are similarly impacted.
It should be noted that in CA, smoke events are often accompanied by unusual shifts in wind direction, and so PM2.5 on a smoky day will not necessarily equal A + B, i.e. the sum of the smoke (A) plus the typical non-smoke PM2.5 (B). Instead, it may correspond to a totally different background (A + C) that may be low and poorly characterized.
Thank you for this comment, we have noted this in the text. We added to the conclusion section:
“These differences could be explained based on prior work, which found that an appreciable fraction of smoke plumes in California are injected into the free troposphere30 or could be due to changes in local meteorology on smoky days. Shifting wind patterns could alter not only the smoke concentration, but also the background concentration. Future work should further investigate the role of local meteorology on altering the composition and the vertical distribution of pollutants during smoke events.”

Equally important is the finding that CA non-smoke days possessed an unusually high PM2.5:AOD ratio. Any additional interpretations or need for future work should be emphasized.
We have added this to the discussion on future work:
“Results from our network indicated that median PM2.5:AOD ratios varied regionally. Network sites in California had the highest PM2.5:AOD ratios on non-smoky days (median of 67.2 µg m-3) compared to the other regions (medians of 24.7-33.5 µg m-3). This could be due to differences in boundary layer heights, emission sources, or particle types. Future work should further explore drivers of the high ratios noted in California on non-smoky days as this has implications for air quality and exposure estimates.”

Specific Comments
Abstract – “However, in California, median PM2.5 remained similar on smoky days relative to non-smoky days, while AOD increased, implying that the smoke was lofted above the surface.” The authors should consider instead “at the California study sites” and “may sometimes have been lofted,” or otherwise constrain the language to avoid statements about CA in general. As the authors cite in other studies, some CA locations frequently report elevated ground PM2.5 during smoke episodes.
We have changed the Abstract language to the following:
“In a regional analysis of wildfire smoke, we observed elevated PM2.5 and AOD on smoky days at study sites in most regions, which led to similar PM2.5:AOD ratios regardless of smoke. However, at the California study sites, median PM2.5 remained similar on smoky days relative to non-smoky days, while AOD increased, implying that the smoke may have been lofted above the surface during the study period. At the California study sites, the median PM2.5:AOD ratio was 67.2 µg m-3 on non-smoky days, compared with 30.2 µg m-3 on smoky days. We show that paired PM2.5 and AOD measurements collected by a crowdsourced network can highlight anomalies in air quality during smoke events and provide insights into the relationship between satellite-based and ground-based air quality observations.”

Abstract – “In California, the median PM2.5:AOD ratio was 67.2 µg m-3 on non-smoky days” Adding another summary value for non-smoky ratios in other US regions is needed to emphasize how unusual this value is.
We have added the following sentence to the abstract:
“At study sites in other regions, the average ratio was 24.7 to 33.5 µg m-3 on non-smoky days and 20.3 to 29.4 µg m-3 on smoky days.”

Introduction p.2 – Beta attenuation monitors should be included in the AQS monitors description along with gravimetric and optical, as many CA sites employ BAMs as AQS FEM.
We have added this to the sentence as follows:
“For example, PM2.5 can be measured gravimetrically by instruments that sample air at a known flow rate and isolate particles with diameters smaller than 2.5 µm from the flow stream, 43,44 which are deposited on a filter; by using light-scattering sensors which estimate aerosol concentrations and size distributions based on how a controlled light source is scattered and absorbed by sampled air45, or by beta attenuation monitors which estimate concentrations based on absorption of beta radiation.”

Section 3.2, p.5, second column – The two sentences starting with “These simultaneous increases…” refer to Fig. 5, not Fig. 4, correct?
Yes, this has been corrected.

Section 3.2, p.5 – “Also of note, is that the non-smoky days in California had a higher ratio…” Please add comparison non-smoky PM2.5:AOD ratios from other US regions to this sentence.
We have changed the sentence to the following:
“Also of note, is that the non-smoky days in California had a higher ratio than all other regions on non-smoky days (67.2 µg m-3 compared to 24.7 – 33.5 µg m-3) and on smoky days (30.7 µg m-3 compared to 20.3 – 29.4 µg m-3).”

Abstract and Section 3.2, p.5-6 – The hypothesis that CA PM aloft on smoke days reduced PM2.5 is compelling, and the limited data shown in Figures S9 and S10 do suggest aloft PM in specific sites on specific days. However, the other presented evidence is somewhat inconsistent: Figure 7 suggests PM2.5:AOD ratios skew visually low on CA smoke days compared to other regions, but the analogous 25%-median-75% ratio data in Table 1 suggest comparable smoky medians and skews across regions; can this apparent discrepancy be clarified? In addition, CA PM2.5 might not be expected to go up on smoke days in a predictable way relative to other regions, given that unusual background PM2.5 and wind patterns (e.g. easterly rather than westerly) may occur on such days.
We want to clarify that we do not think that smoke is always lofted on smoke days in California. Our conclusions are that it likely happens more often here than in other regions (or just happened more likely during our sampling period). Our previous work in Cheeseman et al., 2020 does corroborate that smoke is often lofted above the boundary layer in California. Smoke days did not reduce total PM2.5 concentrations, but just the ratio between PM2.5 and AOD. In Figure 5, we show that PM2.5 concentrations on smoke and no smoke days are relatively similar for our California sites (although there is a longer tail with higher concentrations for smoke days), but AOD clearly increases. Thus, the distribution of the ratio does shift to lower values (as shown in row 3 of Figure 5 and in Figure 7). It is true that the distribution of ratios on smoky days is similar on smoky days across regions, but the difference between smoky and non-smoky days for the California sites is very different than the other regions. In general, we’d expect that the vertical distribution of the other sources would not change on smoky days unless there is a significant wind shift that alters the usual air mass. As mentioned in a previous comment, we have added this point to the text and suggested that this be a greater focus of future work.
“These differences could be explained based on prior work, which found that an appreciable fraction of smoke plumes in California are injected into the free troposphere30 or could be due to changes in local meteorology on smoky days. Shifting wind patterns could alter not only the smoke concentration, but also the background concentration. Future work should further investigate the role of local meteorology on altering the composition and the vertical distribution of pollutants during smoke events.”

Cheeseman, M., Ford, B., Volckens, J., Lyapustin, A., & Pierce, J. R. (2020). The relationship between MAIAC smoke plume heights and surface PM. Geophysical Research Letters, 47, e2020GL088949. https://doi.org/10.1029/2020GL088949.

In general, the statement that “smoke was lofted” in the Abstract should be qualified as a
hypothesis or a possibility (“may have been lofted”).
Reviewer 2 also made this suggestion, thus we changed the sentences in the abstract as follows:
“However, at the California study sites, median PM2.5 remained similar on smoky days relative to non-smoky days, while AOD increased, implying that the smoke may have been lofted above the surface during the study period.”

Section 3.2, p.5-6 – The explanation of shallow boundary layers causing unusually high CA PM25:AOD ratios on non-smoky days is plausible, though another explanation could be non-smoke PM types more prevalent in CA. The relatively high Angstrom exponents on many non-smoke days in Figs. S6 and S7 may reflect significant presence of submicron PM such as vehicle emissions, secondary aerosols, or fine sea salt, the latter of which is also possibly consistent with the low observed AOD.
We have added a sentence to include these other hypotheses.
“Alternatively, there may be differences in aerosol composition and size due to the mixture of different emission sources at the sites in California.”

Fig 6, S6, S7 – The utility of these figures would be improved if all three had the same vertical axes to help interpret the magnitudes of the variations. Optional: in my opinion the Angstrom exponent portion is valuable, so Fig.6 could be replaced with Fig. S6. By eliminating one of these figures, the authors could add a similar figure for a smoke episode from another US region (again with the same Y-axes), which may aid in interpretation.
We have made all of the vertical axes the same and replaced Fig 6 in the main text with Figure S6.

Section 3.4, p.9 – Was any of the previous AMOD gravimetric or continuous PM performance testing conducted in these CA regions? If not, could any unique CA particle types or PM volatility have contributed to the CA PM2.5 or ratio anomalies observed in this study? This could be added to the Limitations.
All the previous testing of the AMOD by our research group was conducted in Colorado. No previous performance testing was done in California by our team (although JPL has been testing AMODs in California). However, the continuous PM comes from a Plantower which has been extensively evaluated. The gravimetric PM is from a system that is the same in the UPAS and has been evaluated previously (Volckens et al., 2017 and Kelleher et al., 2018) However, we note that we have not done evaluation of the coincident measurements outside of Colorado and have added the following to the Limitations section:
“For our in-field AERONET comparison, we are also limited by the number of participant sites that were within 25 km of an AERONET site. Prior validation of the AMOD AOD (e.g., 61,67,68) was all completed in Colorado. Thus, our analysis here, which included sites throughout the CONUS, does suggest that the AMOD provides valid AOD values without any regional bias. However, extensive validation of the coincident PM2.5, AOD, and PM2.5:AOD ratio from the AMOD has not been completed outside Colorado. Thus, future work could include these comparisons, which could also allow for further exploration of the impact of different emission sources and particle types.”

Figure S2 – Please clarify the interpretation of this figure in the text. If a linear decreasing trend between median correction factor and number of smoke days is meant to be inferred, please provide a suggested explanation. Forcing 1 to appear in the Y-axis would be helpful.
We have remade this figure and changed the y-axis to a log scale. The interpretation of this figure is not as straightforward as we had hoped when creating it. We include it because it basically shows the bias in the Plantower because the number of smoke days are, for many regions, a proxy for mass loading (smoke days have higher concentrations at our sites as shown in Table 1). For sampling periods with no or few smoke days/low mass concentrations, the ratio is greater than one because the Plantower underestimates low concentrations. At higher concentrations, the Plantower overestimates concentrations; thus, with more smoke days the ratio is less than one. Previous studies (see remarks and responses to Reviewer 1) suggest that the aerosol type and size can bias the Plantower data. Thus, we were hoping to see a more evident narrowing of the distributions with more smoke days (the 4 smoke days distribution makes this less obvious) as there would be more mass and a more consistent aerosol type/size, thus there would be less variability in the ratio between the filter and Plantower.



We have added to the text the following sentences.
“As shown in Figure S2, scaling factors were greater than one for sampling periods with no or few smoke days (which correspond to lower concentrations, Table 1) and less than one for sampling periods with more smoke days and higher concentrations, corroborating previous work showing a high bias at high concentrations and a low bias at low concentrations (e.g., 45,57–62).”

Figure S9 – Please increase the font size of the Y-axis (altitude) and subtype definitions, or else move the subtype definitions to the caption.
Done.

07-Sep-2023

Dear Dr Volckens:

Manuscript ID: EA-ART-06-2023-000086.R1
TITLE: A national crowdsourced network of low-cost fine particulate matter and aerosol optical depth monitors: Results from the 2021 wildfire season in the United States

Thank you for your submission to Environmental Science: Atmospheres, published by the Royal Society of Chemistry. I sent your manuscript to reviewers and I have now received their reports which are copied below.

After careful evaluation of your manuscript and the reviewers’ reports, I will be pleased to accept your manuscript for publication after minor revisions.

Please revise your manuscript to fully address the reviewers’ comments. When you submit your revised manuscript please include a point by point response to the reviewers’ comments and highlight the changes you have made. Full details of the files you need to submit are listed at the end of this email.

Please submit your revised manuscript as soon as possible using this link :

*** PLEASE NOTE: This is a two-step process. After clicking on the link, you will be directed to a webpage to confirm. ***

https://mc.manuscriptcentral.com/esatmos?link_removed

(This link goes straight to your account, without the need to log in to the system. For your account security you should not share this link with others.)

Alternatively, you can login to your account (https://mc.manuscriptcentral.com/esatmos) where you will need your case-sensitive USER ID and password.

You should submit your revised manuscript as soon as possible; please note you will receive a series of automatic reminders. If your revisions will take a significant length of time, please contact me. If I do not hear from you, I may withdraw your manuscript from consideration and you will have to resubmit. Any resubmission will receive a new submission date.

The Royal Society of Chemistry requires all submitting authors to provide their ORCID iD when they submit a revised manuscript. This is quick and easy to do as part of the revised manuscript submission process. We will publish this information with the article, and you may choose to have your ORCID record updated automatically with details of the publication.

Please also encourage your co-authors to sign up for their own ORCID account and associate it with their account on our manuscript submission system. For further information see: https://www.rsc.org/journals-books-databases/journal-authors-reviewers/processes-policies/#attribution-id

Environmental Science: Atmospheres strongly encourages authors of research articles to include an ‘Author contributions’ section in their manuscript, for publication in the final article. This should appear immediately above the ‘Conflict of interest’ and ‘Acknowledgement’ sections. I strongly recommend you use CRediT (the Contributor Roles Taxonomy, https://credit.niso.org/) for standardised contribution descriptions. All authors should have agreed to their individual contributions ahead of submission and these should accurately reflect contributions to the work. Please refer to our general author guidelines https://www.rsc.org/journals-books-databases/author-and-reviewer-hub/authors-information/responsibilities/ for more information.

I look forward to receiving your revised manuscript.

Yours sincerely,
Dr Lin Wang
Associate Editor, Environmental Science: Atmospheres

Environmental Science: Atmospheres is accompanied by companion journals Environmental Science: Nano, Environmental Science: Processes and Impacts, and Environmental Science: Water Research; publishing high-impact work across all aspects of environmental science and engineering. Find out more at: http://rsc.li/envsci

************

Reviewer 1

Most comments have been satisfactorily answered. Still one or two that seem wrong or not fully justified to me (see attached Word file)

Reviewer 2

The authors have addressed my concerns and I think the current version is ready for publication. A very nice work!

Reviewer 3

All comments have been addressed satisfactorily. I recommend publication.

Thank you to the reviewers for their comments.

Responses to Reveiwer 1:

Commented [LW1]: Thanks, but I was not angling to get this reference in your paper, but pointing out that the CF_1 algorithm is terribly wrong, not only by overestimating PM2.5, but also by not even reporting values that are objectively >0 (as shown by the pm2.5 algorithm, which finds NO zeros in any dataset.) Now consider that all other algorithms (EPA, LRAPA, Australian woodsmoke, Utah group) depend on CF_1 (or CF_ATM), so they all struggle with the zeros. This sentence of yours is not meaningful, since the sensors themselves don’t exhibit high bias at high concentrations and low bias at low concentrations, but only some algorithms. The pm2.5alt algorithm manages to match up very well with hundreds of FRM/FEM stations in three Western states.

Commented [LW2]: See above comment. This sentence is so wrong, it sets my teeth on edge. At least state that these high and low biases are due to certain algorithms, but not implicit in the PMS 5003 sensors.

We have changed these two sentences to more specifically refer to the CF_1 algorithm. However, we do want to slightly push back on the comment that the bias is only algorithm dependent and is not due in any way to the measurement or design of the sensor. There is a lot of research on developing algorithms that may better correct or provide better agreement between Plantowers and FRM/FEM station PM2.5 measurements, but some of the bias at high concentrations is rooted in aerosol physics and the limitations of the sensor. As concentration increases, the scattering median diameter (and the mass median diameter) tends to increase (mostly due to coagulation because of higher number concentrations) and this increase in average size (say from an SMD of 0.4 to 0.6 um) causes a loss in the scattering signal for the PMS5003 (regardless of the algorithm).

“However, CF_1 PM2.5 concentrations from Plantower PMS5003 sensors are known to have a high bias at high concentrations and low bias at low concentrations, requiring calibration and field correction relative to reference methods.45,57–65”

“As previously mentioned, the CF_1 PM2.5 concentrations from Plantower PMS5003 sensors are known to have a high bias at high concentrations and a low bias at low concentrations.”

Commented [LW3]: I think you are overlooking the possibility that PurpleAir monitors may be useful here. You don’t need to start with all 20,000—just the 5 or 10 monitors that might be near each of the 100- odd IMPROVE sites. Then if that proves useful, you could add in more distant PurpleAir monitors. There is considerable homogeneity shown across fairly broad geographic areas (see the “Spatial Variation” paper cited above.” I think your comment that “we know that using a consistent eta value….would lead to very poor agreement” is perhaps subject to a test that can easily be carried out.
We think there may be a point of confusion here in what we are referring to as eta. We are using the eta (or PM2.5:AOD ratio) definition as used in Liu et al., 2005 and van Donkelaar et al., 2006, 2010, 2012, 2016, etc. This eta relates a surface PM2.5 measurement to a full atmospheric column aerosol optical depth measurement from satellites. It encapsulates not only the optical properties but the vertical profile of aerosols and other factors that may influence the relationship between surface level PM2.5 and full atmospheric column aerosol optical depth. We provide PM2.5:AOD ratios (eta values) in our paper (Figures 3c, 5, 6, Table 1) and show that it varies spatially and temporally and is not 0.15 (this variability in ratios/eta has been shown in the Snider et al., 2015 paper, the van Donkelaar papers, and from our AMODs in our previous paper Ford et al., 2019).
The reviewer is suggesting that we compare PurpleAir measurements to nephelometer scattering measurements at IMPROVE sites as previously done in the Ouimette et al. paper. We are not sure how this analysis directly relates to our work here on surface PM2.5 measurements compared to sun photometer measurements of full atmospheric column aerosol optical depth. IMPROVE sites do not include sun photometers, thus we would still need to compare the surface scattering estimates to some other column AOD measurement (like from a satellite), in order to discuss how surface measurements relate to full column measurements. While we do think that it would be an interesting analysis to compare surface extinction measurements from IMPROVE nephelometers to full atmospheric column aerosol optical depth measurements, that topic is really outside the scope of this paper, which is about and AOD measurements from our AMODs and how our results relate to interpretations of satellite AOD.


References:
B. Ford, J. R. Pierce, E. Wendt, M. Long, S. Jathar, J. Mehaffy, J. Tryner, C. Quinn, L. van Zyl, C. L’Orange, D. Miller-Lionberg and J. Volckens, A low-cost monitor for measurement of fine particulate matter and aerosol optical depth &ndash; Part 2: Citizen science pilot campaign in northern Colorado, Atmospheric Meas. Tech. Discuss., 2019, 1–20.
Y. Liu, J. A. Sarnat, V. Kilaru, D. J. Jacob and P. Koutrakis, Estimating ground-level PM2.5 in the eastern United States using satellite remote sensing, Environ. Sci. Technol., 2005, 39, 3269–3278.
G. Snider, C. L. Weagle, R. V. Martin, A. van Donkelaar, K. Conrad, D. Cunningham, C. Gordon, M. Zwicker, C. Akoshile, P. Artaxo, N. X. Anh, J. Brook, J. Dong, R. M. Garland, R. Greenwald, D. Griffith, K. He, B. N. Holben, R. Kahn, I. Koren, N. Lagrosas, P. Lestari, Z. Ma, J. Vanderlei Martins, E. J. Quel, Y. Rudich, A. Salam, S. N. Tripathi, C. Yu, Q. Zhang, Y. Zhang, M. Brauer, A. Cohen, M. D. Gibson and Y. Liu, SPARTAN: a global network to evaluate and enhance satellite-based estimates of ground-level particulate matter for global health applications, Atmos Meas Tech Discuss, 2015, 7, 7569–7611.
A. van Donkelaar, R. V. Martin, M. Brauer, N. C. Hsu, R. A. Kahn, R. C. Levy, A. Lyapustin, A. M. Sayer and D. M. Winker, Global Estimates of Fine Particulate Matter using a Combined Geophysical-Statistical Method with Information from Satellites, Models, and Monitors, Environ. Sci. Technol., 2016, 50, 3762–3772.
A. van Donkelaar, R. V. Martin and R. J. Park, Estimating ground-level PM2.5 using aerosol optical depth determined from satellite remote sensing, J. Geophys. Res. Atmospheres, 2006, 111, D21201.
A. van Donkelaar, R. Martin, C. Verduzco, M. Brauer, R. Kahn, R. Levy and P. Villeneuve, A Hybrid Approach for Predicting PM2.5 Exposure: van Donkelaar et al. Respond, Environ. Health Perspect., 2010, 118, a426–a426.
A. van Donkelaar, R. V. Martin, A. N. Pasch, J. J. Szykman, L. Zhang, Y. X. Wang and D. Chen, Improving the accuracy of daily satellite-derived ground-level fine aerosol concentration estimates for North America, Environ. Sci. Technol., 2012, 46, 11971–11978.

16-Sep-2023

Dear Dr Volckens:

Manuscript ID: EA-ART-06-2023-000086.R2
TITLE: A national crowdsourced network of low-cost fine particulate matter and aerosol optical depth monitors: Results from the 2021 wildfire season in the United States

Thank you for submitting your revised manuscript to Environmental Science: Atmospheres. I am pleased to accept your manuscript for publication in its current form.

You will shortly receive a separate email from us requesting you to submit a licence to publish for your article, so that we can proceed with the preparation and publication of your manuscript.

You can highlight your article and the work of your group on the back cover of Environmental Science: Atmospheres. If you are interested in this opportunity please contact the editorial office for more information.

Promote your research, accelerate its impact – find out more about our article promotion services here: https://rsc.li/promoteyourresearch.

We will publicise your paper on our Twitter account @EnvSciRSC – to aid our publicity of your work please fill out this form: https://form.jotform.com/211263048265047

How was your experience with us? Let us know your feedback by completing our short 5 minute survey: https://www.smartsurvey.co.uk/s/RSC-author-satisfaction-energyenvironment/

By publishing your article in Environmental Science: Atmospheres, you are supporting the Royal Society of Chemistry to help the chemical science community make the world a better place.

With best wishes,

Dr Lin Wang
Associate Editor, Environmental Science: Atmospheres

Environmental Science: Atmospheres is accompanied by companion journals Environmental Science: Nano, Environmental Science: Processes and Impacts, and Environmental Science: Water Research; publishing high-impact work across all aspects of environmental science and engineering. Find out more at: http://rsc.li/envsci

******

Sign up to contents alerts from Environmental Science: Atmospheres and stay up to date with the newest developments: http://rsc.li/esatmos-alerts

******

If you need to contact the journal, please use the email address esatmospheres@rsc.org

************************************

DISCLAIMER:

This communication is from The Royal Society of Chemistry, a company incorporated in England by Royal Charter (registered number RC000524) and a charity registered in England and Wales (charity number 207890). Registered office: Burlington House, Piccadilly, London W1J 0BA. Telephone: +44 (0) 20 7437 8656.

The content of this communication (including any attachments) is confidential, and may be privileged or contain copyright material. It may not be relied upon or disclosed to any person other than the intended recipient(s) without the consent of The Royal Society of Chemistry. If you are not the intended recipient(s), please (1) notify us immediately by replying to this email, (2) delete all copies from your system, and (3) note that disclosure, distribution, copying or use of this communication is strictly prohibited.

Any advice given by The Royal Society of Chemistry has been carefully formulated but is based on the information available to it. The Royal Society of Chemistry cannot be held responsible for accuracy or completeness of this communication or any attachment. Any views or opinions presented in this email are solely those of the author and do not represent those of The Royal Society of Chemistry. The views expressed in this communication are personal to the sender and unless specifically stated, this e-mail does not constitute any part of an offer or contract. The Royal Society of Chemistry shall not be liable for any resulting damage or loss as a result of the use of this email and/or attachments, or for the consequences of any actions taken on the basis of the information provided. The Royal Society of Chemistry does not warrant that its emails or attachments are Virus-free; The Royal Society of Chemistry has taken reasonable precautions to ensure that no viruses are contained in this email, but does not accept any responsibility once this email has been transmitted. Please rely on your own screening of electronic communication.

More information on The Royal Society of Chemistry can be found on our website: www.rsc.org