Calibration of PurpleAir PA-I and PA-II Monitors Using Daily Mean PM_2.5 Concentrations Measured in California, Washington, and Oregon from 2017 to 2021

Round 1

Reviewer 1 Report

In the Manuscript ID: sensors-1739887, entitled “Calibration of PurpleAir PA-I and PA-II monitors using daily mean PM2.5 concentrations measured in California, Washington, and Oregon from 2017
to 2021”, the authors aim to determine a revised calibration factor for a PM low-cost sensor, using monitors located near regulatory PM2.5 Air Quality System (Washington, Oregon, and California; 2017 – 2021). The manuscript is of interest and well written: few comments are reported below:

 

• Line 38: Can the authors give an idea of: "[…] numbers were small before 2018"?
• Line 55: Please pay attention to "40% .."
• Line 106: Please pay attention to the citation.
• Lines 143-158: I suggest that the authors make these parts more immediate to understand.
• Line 234: Please pay attention to “ug/m3”.
• Line 281: Please pay attention: “PM2.5”.
• I suggest authors verify that acronyms are defined in full only once in the text (e.g. "RH", "FRM", AQS ").
• Paragraph 4.2. - Limitations: Did the authors also consider the seasonal variability of the data? Do the authors think that it could be useful to provide a "summer" and "winter" calibration factor?
• Paragraph 5. - Conclusions: I suggest that the authors report the benefits this work can bring to the scientific community.

Author Response

• Line 38: Can the authors give an idea of: "[…] numbers were small before 2018"?  There were fewer than 30 at the end of 2018.  "<30" added to MS
• Line 55: Please pay attention to "40% .." Corrected
• Line 106: Please pay attention to the citation. Named citation removed to leave numbered citation
• Lines 143-158: I suggest that the authors make these parts more immediate to understand. A further sentence was added to explain the coefficients: ".  The coefficients shown derive directly from the ALT-CF3 approach of selecting an average (geometric mean) diameter for the three smallest size categories and using the density of water. "
• Line 234: Please pay attention to “ug/m3”. Gk letter "mu" added.
• Line 281: Please pay attention: “PM2.5”. Subscript "2.5" added.
• I suggest authors verify that acronyms are defined in full only once in the text (e.g. "RH", "FRM", AQS "). All three acronyms now have only one definition in full the first time they appear/ (thank you for noticing that).
• Paragraph 4.2. - Limitations: Did the authors also consider the seasonal variability of the data? Do the authors think that it could be useful to provide a "summer" and "winter" calibration factor? This is an excellent idea for further analysis. This would be a major effort so we cannot add this to this study: however, we have added this idea for possible future work.
• Paragraph 5. - Conclusions: I suggest that the authors report the benefits this work can bring to the scientific community.  Another good idea--we have added a brief mention of benefits to our conclusion.

Reviewer 2 Report

This is a good study that describes an important topic: the quality of PM data provided by low-cost sensors. These low-cost sensors are widely deployed and the publicly available data is viewed by many. The methodology of the study is generally sound. The authors compare the low-cost results that are open access to values from the EPA's regulatory network. 

I have a number of specific comments incorporated into the PDF file that need to be addressed. In addition, it would be helpful for the authors to speculate on the nature of the empirical correction factor that is derived. What are the potential theoretical underpinnings of the CF? Is it as simple as particle density? 

Second, this is a very nice dataset, and it would be good to explore if the CF depends on other factors, including RH and the spatial distance between the low-cost sensor and the regulatory sensor. I understand this is out of scope for the present work, but some thoughts and speculation should be provided in the Conclusion on potential avenues for future work.

See detailed comments on PDF file.

Comments for author File: Comments.pdf

Author Response

PLEASE SEE ATTACHMENT

Author Response File: Author Response.pdf

Reviewer 3 Report

Sensors

 

Manuscript Number: sensors-1739887

 

Title: Calibration of PurpleAir PA-I and PA-II monitors using daily mean PM2.5 concentrations measured in California, Washington, and Oregon from 2017 to 2021

 

Reviewer' comments:

Overall statement of the article

Particulate matter (PM) is one of the regularly monitored air pollutants due to its substantial effect on human health [1, 2]. In the last decade the evidence on the potential role of coarse particles [PM2.5] as a risk factor to human health has accumulated in the epidemiological literature [3]. Indeed, Exposure to PM is linked to various health outcomes such as respiratory disease [4], cardiovascular disease [5] , and pregnancy outcomes [6]. Around 4.2 million of deaths are assigned to air pollution exposure Worldwide [1]. Therefore, assessing air pollution exposures in order to estimate human health risk and elaborate process to manage those risks becomes an major measure public health.

The paper sensors-1739887 “Calibration of PurpleAir PA-I and PA-II monitors using daily mean PM2.5 concentrations measured in California, Washington, and Oregon from 2017 to 2021 “, submitted to Sensors Journal for publication, put in evidence the necessity to determine the quality of measurements allowing to assess the concentrations of air particles.

In this paper, the authors  determine a revised calibration factor for PA-II monitors using all PurpleAir monitors located near regulatory PM2.5 Air Quality System (AQS) monitors using Federal Reference Methods (FRM) in the West Coast states of Washington, Oregon, and California between the dates of Jan 1, 2017 through Sept. 8, 2021. We also determine a new calibration factor for PA-I monitors by comparing to nearby outdoor PA-II monitors.

The paper contains 10 keywords: sensors, low-cost particle monitors, calibration factor, PurpleAir, particles, PM2.5, ALT-CF3, algorithm, PMS1003, PMS5003

This work is divided into 4 distinct parts: (i) Abstract, (ii) Introduction, (iii) Materials and Methods, (iv) Results, (v)  Discussion, and (v) Conclusions.

The paper has 2 tables and 2 figures. There are 19 references, which are from 1990 to 2021. The majority of the references are internationally evaluated and published in peer-reviewed journals.

According to the authors, “These results show that PurpleAir data can agree well with regulatory monitors 319 when an optimum calibration factor is found”.

 

Overall strengths of the article

The title is clear. The authors have clearly state the objectives of the article.

The topic of the works is of general interest, and the article reflects a current state of knowledge with a sufficiently critical and internationally evaluated literature. The paper is an informative paper.

The authors provide information on the subject, and discuss information reported in the scientific literature.  the structure of the paper should be revised with regard to conciseness.

In order to render the paper more impactful reviewer wonders to the authors to revise the introduction in its structure. This section needs some other references.

Is it possible to reinforce the methodology of the work at the theoretical level by mentioning some scientific references?

The reviewer asks the authors to further develop the results section and to increase the comparative analysis between their results and the information available in the scientific literature.

P3, L106 Please revise the reference (Wallace et al., 2021¹³).

Overall statement

The paper needs major revisions.

 

Reviewer’ References

1. Shtein, A., Karnieli, A., Katra, I., Raz, R., Levy, I., Lyapustin, A., ... & Kloog, I. (2018). Estimating daily and intra-daily PM10 and PM2. 5 in Israel using a spatio-temporal hybrid modeling approach. Atmospheric Environment, 191, 142-152. https://doi.org/10.1016/j.atmosenv.2018.08.002

2. Stafoggia, M., Bellander, T., Bucci, S., Davoli, M., De Hoogh, K., De'Donato, F., ... & Schwartz, J. (2019). Estimation of daily PM10 and PM2. 5 concentrations in Italy, 2013–2015, using a spatiotemporal land-use random-forest model. Environment international, 124, 170-179. https://doi.org/10.1016/j.envint.2019.01.016

3. WHO (World Health Organization), 2013. Review of Evidence on Health Aspects of Air Pollution – REVIHAAP Project: Final Technical Report. WHO Regional Offiffiffice for Europe, Copenhagen, Denmark.

4. Kloog, I., Ridgway, B., Koutrakis, P., Coull, B.A., Schwartz, J.D., Program, R., 2013. Longand short-term exposure to PM2.5 and mortality: using novel exposure models. Epidemiology 24, 555–561. https://doi.org/10.1097/EDE.0b013e318294beaa 

5. Zanobetti, A., Schwartz, J., 2005. The effect of particulate air pollution on emergency admissions for myocardial Infarction : a multicity case-crossover analysis. Environ. Health Perspect. 978, 978–982. https://doi.org/10.1289/ehp.7550 

6. Zeka, A., Melly, S.J., Schwartz, J., 2008. The effects of socioeconomic status and indices of physical Massachusetts. Environ. Health 7, 1–13. https://doi.org/10.1186/1476-069X-7-60 

 

Comments for author File: Comments.doc

Author Response

Please see attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

I appreciate the thoughtful responses of the authors to my comments. I respect the few instances were the authors argued against my recommendations. The manuscript is now ready for publication, with a few minor edits:

Line 196: remove one of the two end-of-sentence periods.

Lines 242-243: some formatting/editing issue with the insertion of the parenthetical.

Lines 321-325: too many new paragraphs.

Author Response

Thankyou for a review that was a model of careful concise review with flexibility in judging our responses.

 

Line 196: remove one of the two end-of-sentence periods.

Done

Lines 242-243: some formatting/editing issue with the insertion of the parenthetical.

Sentence changed  as follows:

We located 194 outdoor PA-II sites within 1 km of 43 outdoor PA-I sites. (One km was chosen instead of 0.5 km to increase the number of sites available for analysis).

Lines 321-325: too many new paragraphs.

I combined two paragraphs into one:

Also, correlations between PurpleAir and FRM monitors were stable across all distances from 1-10 km apart, indicating strong spatial uniformity of PM2.5 concentrations.

Reviewer 3 Report

Sensors

 

Manuscript Number: sensors-1739887-V2

 

Title: Calibration of PurpleAir PA-I and PA-II monitors using daily mean PM2.5 concentrations measured in California, Washington, and Oregon from 2017 to 2021

 

Reviewer' comments:

Overall statement of the article

Particulate matter (PM) is one of the regularly monitored air pollutants due to its substantial effect on human health [1, 2]. In the last decade the evidence on the potential role of coarse particles [PM2.5] as a risk factor to human health has accumulated in the epidemiological literature [3]. Indeed, Exposure to PM is linked to various health outcomes such as respiratory disease [4], cardiovascular disease [5] , and pregnancy outcomes [6]. Around 4.2 million of deaths are assigned to air pollution exposure Worldwide [1]. Therefore, assessing air pollution exposures in order to estimate human health risk and elaborate process to manage those risks becomes an major measure public health.

The paper sensors-1739887 “Calibration of PurpleAir PA-I and PA-II monitors using daily mean PM2.5 concentrations measured in California, Washington, and Oregon from 2017 to 2021 “, submitted to Sensors Journal for publication, put in evidence the necessity to determine the quality of measurements allowing to assess the concentrations of air particles.

In this paper, the authors  determine a revised calibration factor for PA-II monitors using all PurpleAir monitors located near regulatory PM2.5 Air Quality System (AQS) monitors using Federal Reference Methods (FRM) in the West Coast states of Washington, Oregon, and California between the dates of Jan 1, 2017 through Sept. 8, 2021. We also determine a new calibration factor for PA-I monitors by comparing to nearby outdoor PA-II monitors.

The paper contains 10 keywords: sensors, low-cost particle monitors, calibration factor, PurpleAir, particles, PM2.5, ALT-CF3, algorithm, PMS1003, PMS5003

This work is divided into 4 distinct parts: (i) Abstract, (ii) Introduction, (iii) Materials and Methods, (iv) Results, (v)  Discussion, and (v) Conclusions.

The paper has 2 tables and 2 figures. There are 19 references, which are from 1990 to 2021. The majority of the references are internationally evaluated and published in peer-reviewed journals.

According to the authors, “These results show that PurpleAir data can agree well with regulatory monitors 319 when an optimum calibration factor is found”.

 

Overall strengths of the article

The title is clear. The authors have clearly state the objectives of the article. The topic of the works is of general interest, and the article reflects a current state of knowledge with a sufficiently critical and internationally evaluated literature. The paper is an informative paper. The authors provide information on the subject, and discuss information reported in the scientific literature. 

Overall statement

The authors replied to the various comments made by the reviewers. They justified their answers and scrupulously worked on each of the comments. The paper should be accepted to be published in the journal after.

 

Reviewer’ References

1. Shtein, A., Karnieli, A., Katra, I., Raz, R., Levy, I., Lyapustin, A., ... & Kloog, I. (2018). Estimating daily and intra-daily PM10 and PM2. 5 in Israel using a spatio-temporal hybrid modeling approach. Atmospheric Environment, 191, 142-152. https://doi.org/10.1016/j.atmosenv.2018.08.002

2. Stafoggia, M., Bellander, T., Bucci, S., Davoli, M., De Hoogh, K., De'Donato, F., ... & Schwartz, J. (2019). Estimation of daily PM10 and PM2. 5 concentrations in Italy, 2013–2015, using a spatiotemporal land-use random-forest model. Environment international, 124, 170-179. https://doi.org/10.1016/j.envint.2019.01.016

3. WHO (World Health Organization), 2013. Review of Evidence on Health Aspects of Air Pollution – REVIHAAP Project: Final Technical Report. WHO Regional Offiffiffice for Europe, Copenhagen, Denmark.

4. Kloog, I., Ridgway, B., Koutrakis, P., Coull, B.A., Schwartz, J.D., Program, R., 2013. Longand short-term exposure to PM2.5 and mortality: using novel exposure models. Epidemiology 24, 555–561. https://doi.org/10.1097/EDE.0b013e318294beaa 

5. Zanobetti, A., Schwartz, J., 2005. The effect of particulate air pollution on emergency admissions for myocardial Infarction : a multicity case-crossover analysis. Environ. Health Perspect. 978, 978–982. https://doi.org/10.1289/ehp.7550 

6. Zeka, A., Melly, S.J., Schwartz, J., 2008. The effects of socioeconomic status and indices of physical Massachusetts. Environ. Health 7, 1–13. https://doi.org/10.1186/1476-069X-7-60 

 

Author Response

Thank you for your overall comment that the paper should be accepted.