From the journal Environmental Science: Atmospheres Peer review history
Investigation of indoor air quality in university residences using low-cost sensors
Round 1
Manuscript submitted on 05 nov 2022![](https://www.rsc-cdn.org/pubs-core/2022.0.147/content/NewImages/icon-letter.png)
![](https://www.rsc-cdn.org/pubs-core/2022.0.147/content/NewImages/chevron-down.png)
20-Nov-2022
Dear Dr Zhao:
Manuscript ID: EA-ART-11-2022-000149
TITLE: Investigation of Indoor Air Quality in University Residences Using Low-Cost Sensors.
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 major 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 Nønne Prisle
Associate Editor, Environmental Sciences: Atmospheres
************
![](https://www.rsc-cdn.org/pubs-core/2022.0.147/content/NewImages/icon-comments.png)
![](https://www.rsc-cdn.org/pubs-core/2022.0.147/content/NewImages/chevron-down.png)
Reviewer 1
This was an interesting paper looking at environmental quality in university halls of residence. I thought it was well written and presented, with some interesting results. I don’t have any major comments, but some minor ones before. My biggest concern was the choice of PM sensor, which is known to be poor and the authors discovered this. There is nothing that can be done about it, but the authors could write a section on the limitations of the study and this would be in it – including down time of their system etc.
Material and Methods: I would like to see a table for the EPA exposure concentrations for indoor environments and also a specification for the design of the unit. It would be fine to align the EPA requirements to the sensors purchased using a table for direct comparison. It would also be interesting to understand why you made the sensor choices you did.
Line 117: What was the flow rate (or air movement) of the fan and what was the dead volume inside the unit?
Line 162: I agree with this statement and I am pleased you say this. I would comment you are not calibrating for temperature PM detection (it is laser based, so your optics are affected by all sorts).
How often did you calibrate your sensors?
Line 196: Have you got the approval number?
Line 276: It does sound like self-heating. Please use this term as its normally how this is described. Somewhere, you might want to add that you could have calibrated this out.
Line 327: I would also expect windows to be open with cooking and residencies and it could be outside pollution from traffic etc. It would be useful to have a map showing the locations of the residencies. Whatever, this is an interesting result.
Line 370: I think you have to take into account the location of the room within the block, outside effects and its position within the path of ventilation. So care needs to be taken on any conclusions drawn.
Line 478: Can you comment on what a “Purple Air” system is? Is it a government unit with high precision? What was the sampling period for this unit?
Line 533: Please add the limitations of the study. For example, you only measured PM2.5 and did not include PM10, PM1 or ultrafines, which is becoming of increasing concern.
Reviewer 2
I think it’s great that the authors of this study provided students living in the university residences with access to real-time indoor air quality data! I also appreciate that these results illustrate how much ultrasonic humidifiers can increase indoor PM2.5 concentrations.
1. I would like to see the authors do some additional quality assurance and correction of their CO2 data from the Sensirion SCD30 sensor. I think low-cost NDIR CO2 sensors are great tools for assessing ventilation of indoor spaces, but such sensors are known to suffer from baseline drift. This issue of baseline drift is the reason why commercially-available monitors that use low-cost NDIR CO2 sensors, like the Aranet4, recommend that the baseline be “recalibrated” to the outdoor concentration of 420 ppm on a regular basis. Fortunately, it sounds like the authors collected the data they would need to perform a baseline correction on the SCD30 sensor data. On lines 181-184, the authors state that they opened the windows and collected SCD30 sensor data while the CO2 concentration decayed to the outdoor ambient value of 420 ppm. What CO2 concentration did each sensor report when measuring the outdoor concentration of 420 ppm? The data from each sensor could be corrected, by adding or subtracting some constant value from each reported concentration, so that the corrected CO2 concentration is equal to 420 ppm when the sensor is measuring outdoor air. Another issue is that low-cost NDIR CO2 sensors are sensitive to variations in atmospheric pressure but the SCD30 doesn’t measure atmospheric pressure. Unless provided with data indicating otherwise, sensors like the SCD30 assume that the atmospheric pressure is equal to 101.3 kPa. At lower atmospheric pressures, air density in lower, and there are fewer molecules per unit volume, so the sensor will underestimate the true CO2 mole fraction. Tryner et al. (DOI: 10.1016/j.buildenv.2021.108398) found that the SCD30 underestimated the CO2 concentration by about 20% in a location where the atmospheric pressure was 85 kPa, even after a baseline correction was applied. The atmospheric pressure in Edmonton is 94 kPa. Did the device that the authors built collect atmospheric pressure data? If so, were those pressure data used to correct the CO2 data for the difference between atmospheric pressure at sea level and the atmospheric pressure in Edmonton? If not, the authors should add a disclaimer that CO2 data from the SCD30 were not corrected for atmospheric pressure and that the reported CO2 values might be slightly underestimated. My guess is that, even if CO2 values were not corrected for pressure, CO2 concentrations are not underestimated by more than 10%.
2. Line 88: I suggest revising “> 5000 ppm” to “> 1000 ppm”. I agree that indoor locations with elevated CO2 concentrations pose a risk to human health, but 5000 ppm is a very high CO2 concentration that I don’t think is common.
3. Lines 116-117 and Lines 166-167: “…The SDS011 sensor could also measure PM10…” Despite the manufacturer’s claims, the SDS011 cannot measure PM10. See Kuula et al. (DOI: 10.5194/amt-13-2413-2020, especially Figure 5, the text on page 2418, and the references to Budde et al. and Laquai therein). I suggest not stating that this sensor can measure PM10 anywhere in the manuscript. There’s no need to comment on the ability to measure PM10 with the SDS011 since PM10 was not a focus of this study.
4. Line 169: “RH is considered the primary source of measurement error for PM sensors.” The rest of this paragraph is spot-on, but I don’t like the phrasing of this statement that RH is the “primary” source of measurement error for PM sensors. I’d argue that the key sources of measurement error are really differences in the size distribution and refractive index between the particles used to calibrate the sensor and the particles being measured by the sensor. The authors explain clearly in the rest of the paragraph that changes in RH can lead to measurement error because RH can affect particle size and refractive index. I suggest that the authors just revise this first sentence as “RH can be a major source of measurement error for PM sensors.”
5. Line 208: “…an the window or the corridor.” I think there is a typographical error here.
6. Table 1, Note a: I think “in-suit” should be “in-suite”.
7. Section 3.2: Please move Figure 4, Figure 5, and some of this text to the SI. It would be sufficient to comment briefly here on data availability, the Wi-Fi issues that the authors’ experienced, and the dates when many students left their residences, but I don’t think all of the information in this section is critical to readers, and I think this manuscript would benefit from being shorter.
8. Lines 322-325: The PM2.5 concentrations that the authors report here are very low (1 to 4 μg/m3). Are the authors certain that these concentrations are distinguishable from each other and from zero? Did the authors do any tests to evaluate the limit of detection for the SDS011 sensor? In other words what is the lowest PM2.5 concentration that the SDS011 can distinguish as being different from zero? Did the authors test the SDS011 measuring HEPA-filtered air or “clean” air in the office without the humidifier present?
9. Lines 332-337: How were averages and standard deviations calculated for the t-test? Was the time-averaged PM2.5 concentration reported by each of the 8 sensors in Residence A during the occupied period treated as a single measurement for the occupied period? Were the sample mean and sample standard deviation calculated from the time-averaged PM2.5 concentrations measured by each of the n sensors in each residence? Or was something else done?
10. Figure 6 caption: It would be helpful to remind readers in the figure caption that all residences A-D were mechanically ventilated, that residences A–B prohibited cooking, and that residences C–D allowed cooking.
11. Line 359: “closed” should be “close”
12. Lines 361-364, 365-368, 378-380 and Table S4: Did the authors adjust their α values for these multiple comparisons? One option is to use Tukey’s method for multiple comparisons.
13. Lines 386-387: “Results of this study confirm that MV with filtration can significantly reduce indoor particle levels and maintain good IAQ.” Do the authors know the MERV ratings of the filters used in the HVAC systems for these residences?
14. Figure 7 caption: Please note here in the figure caption that Residences A–D were mechanically ventilated and that Residence E was naturally ventilated.
15. Lines 417-420: This is an important result!
16. Lines 436-438: “Student residents should, where possible, use water with low levels of mineral and other dissolved components in ultrasonic humidifiers to limit inhalation exposure to potentially harmful components.” Or they should use evaporative humidifiers?
17. Lines 476-478: What correction was applied to the PurpleAir data? The US EPA correction?
18. Lines 491-494: “Dai et. Al, suggested minimizing particle filter efficiencies of MV systems are 86%, 85%, 74%, 58%, and 62% for severe cold, cold, hot summer and cold winter, moderate and hot summer, and warm winter zones, respectively.” I don’t understand what this sentence means.
19. Lines 496-498: “Based on the weekday’s data, the outdoor PM2.5 removal efficiency of HVAC and HRV filters was found at 76-84% and 80-90%, which indicates the high filtration efficiency of both ventilation systems.” How was this calculation performed? I don’t think it’s appropriate to do this sort of quantitative comparison using indoor and outdoor concentrations measured using two different brands and models of low-cost PM sensors. There’s no guarantee that the SDS011 and the PMS5003 respond similarly to the outdoor PM2.5. It’s useful to point out the correlation between outdoor and indoor PM2.5 that is shown in Figure 11, but the authors should remove these estimates of filtration efficiency from the manuscript.
20. Lines 503-505: “In this study, the MV systems in the two studied buildings were successful in keeping PM2.5 concentration below the recommended daily acute exposure limit.” Were the outdoor PM2.5 concentrations above the recommended daily acute exposure limit? If the outdoor concentrations were already below the recommended daily acute exposure limit, then the MV systems didn’t have to do anything.
![](https://www.rsc-cdn.org/pubs-core/2022.0.147/content/NewImages/icon-response.png)
![](https://www.rsc-cdn.org/pubs-core/2022.0.147/content/NewImages/chevron-down.png)
Response to Reviewers
Referee: 1
Comment to the Author
General comments:
This was an interesting paper looking at environmental quality in university halls of residence. I thought it was well written and presented, with some interesting results.
Response: We appreciate the reviewer for the positive feedback. We have made responses below to the comments of the reviewer, and also made corrections in our manuscript. We believe the manuscript is much improved now.
Specific comments:
I don’t have any major comments, but some minor ones before. My biggest concern was the choice of PM sensor, which is known to be poor and the authors discovered this. There is nothing that can be done about it, but the authors could write a section on the limitations of the study and this would be in it – including downtime of their system, etc.
Response: Thank you for the comments.
Firstly, regarding the choice of this sensor, we found that this sensor was easily embedded with Arduino and ESP8266 microcontrollers. Additionally, the SDS011 sensor measurement capacity range was up to 1000 ug/m3 for PM2.5.
At the stage of project design, and also now, we did not feel that SDS011 was particularly a bad choice compared to other low-cost sensors on the market. All the LCSs require corrections, and we achieved satisfactory accuracies after sensor calibration. Additionally, SDS011 is already widely being used in deployments around the world, ranging from sensor networks to grassroots citizen science projects (e.g., Aeroqual), and its performance has been verified by a few previous studies as already cited in the manuscript.
With regard to the downtime, it was not the fault of the sensor itself, but rather the compatibility of the residence network and the ESP board that we used. As non-specialists in IoT and programming languages, we would encounter the same problem with whichever LCS we had chosen. We also agree that there are potential limitations of the study and the sensor itself. So considering these facts and findings from our study, we described the possible limitations in sections 2.2, 3.1, and 3.2 of the manuscript.
Following Reviewer 2’s comment, we have moved the original Figure 5 to the SI, figure S1. Related to that, we have reformatted our discussions regarding sensor limitations in Section 3.2 (sensor data quality) in the manuscript.
Material and Methods: I would like to see a table for the EPA exposure concentrations for indoor environments and also a specification for the design of the unit. It would be fine to align the EPA requirements to the sensors purchased using a table for direct comparison. It would also be interesting to understand why you made the sensor choices you did.
Response: Thank you for bringing up this point. Please see the responses below for the specification or criteria recommended by the United State Environmental Protection Agency (US EPA) for the low-cost sensors and our low-cost sensor specification. Following the comment, we have added the below sentences and table in the SI (Table S1), and also linked it to the manuscript.
“The Low-cost sensor (LCS) choice was made considering the United State Environmental Protection Agency (US EPA) requirements. Besides that, a sensor is considered useful when it can detect the target pollutants over the full range of concentrations generally present in the environment of interest, for instance, particular matter, PM2.5 (0-40 µg/m3) and carbon dioxide, CO2 (350-600 ppm). However, the US EPA recommended some criteria for selecting LCS mainly for measuring the outdoor pollutants, with no particular parameters set up yet for the indoor environment.
The choice of our LCS was determined by the target pollutants (i.e., PM2.5, and CO2), their detection range, and the price of the sensor unit.”
Parameter Regulatory exposure limit US EPA recommendation LCS specification
Particulate matter (PM2.5) a35 µg/m3 (24-hour average) c5 µg/m3 d0-999.9 µg/m3
e0.6 µg/m3
Carbon dioxide (CO2) b1000 ppm (24-hour average) c100 ppm d0-40000 ppm
eNot measured
Response time - Less than 1 minute 1 Sec
Cost - $100 to $2500 $150
Note: aRecommended by US EPA2 and bRecommended by Canada health.3,4 This recommended CO2 exposure limit is in line with standards from other countries (i.e., Korea, Japan, France, Norway, Portugal, and Germany) and organizations (i.e., The American Society of Heating, Refrigerating and Air-Conditioning Engineers).4
cUseful detection limit recommended by US EPA, dDetection range, and eDetection limit of LCS.
Line 117: What was the flow rate (or air movement) of the fan and what was the dead volume inside the unit?
Response: The sensor had an inlet, from which we measured the flow rate to be 20-35cc. However, the flow rate was found variable as the system is not fully airtight and thus, it was also assumed that air can get into the sensor unit through other openings of the sensor.
We are not entirely sure what the reviewer means by dead volume. It was estimated that the volume of the space around the detection chamber is roughly 3 to 4 cc.
We have now added the flow rate in the manuscript that we measured in the manuscript Lines (119-120):
“A fan is used to draw air into the measurement cavity through an inlet with a flow rate of 20-35ccm.”
Line 162: I agree with this statement and I am pleased you say this. I would comment you are not calibrating for temperature PM detection (it is laser-based, so your optics are affected by all sorts). How often did you calibrate your sensors?
Response: Thanks. Yes, we didn’t do the calibration for temperature but just did the inter-comparisons between the sensors (Section 3.1, table 2). We did the PM sensor calibration before the sensor deployment and after the end of the study. Lines (162-165) has been modified for better clarification.
“However, we acknowledge that the calibration may not fully represent the entire spectrum of indoor aerosol, as the performance of LCS was found to vary under different conditions ( i.e, RH) and emitting sources (i.e, size, density, and refractive index of the PM).”
Line 196: Have you got the approval number?
Response: We confirmed the human research ethics approval prior to starting the study and the approval no. is Pro00112541. To include this information, the sentence has been revised to Lines (205-206) and also added in the acknowledgment:
“Prior to the recruitment, human research ethics approval was obtained from the University of Alberta Research Ethics Office (Pro00112541).”
Line 276: It does sound like self-heating. Please use this term as it's normally how this is described. Somewhere, you might want to add that you could have calibrated this out.
Response: Thank you for the suggestions. We have now modified all the related words into ‘self-heating”. We didn’t do the calibration of the SDS30 sensor for temperature. However, we did a calculation against absolute humidity to figure out how much overestimation of the temperature was measured by our SDS30 sensor. It is assumed that it is affected by internal heating or conditions within the sensor.5 The following sentence has been modified (Lines: 289-291) following the suggestion:
“However, we found that the hourly average temperature in the studied residences was recorded at 26.5 ± 2°C, which is 2-3°C higher than the actual temperature, and is assumed to be influenced by self-heating and internal conditions within the sensor unit 5.”
Line 327: I would also expect windows to be open with cooking and residencies and it could be outside pollution from traffic etc. It would be useful to have a map showing the locations of the residencies. Whatever, this is an interesting result.
Response: Thanks for pointing this out. However, as per our agreement with the sponsor, we can’t add the map indicating the location of the residence as it will reveal the residence's identity. But here to answer your query we would like to mention that the distance of the major road from residences A and B (approx. 38m), residence C (approx. 300m) and residence D (approx. 20m). So in terms of the location from the road, it is hard to discuss and compare the effect of traffic emission on indoor PM2.5 levels as the distance of A, B & D from the main road is very close except the residence C. On the other hand, during the comparison of the study period, the outside temperature was very low (- 10℃ to -32℃), so the probability of opening the window was zero percent (confirmed by the questionnaire's responses).
To better clarify, we followed the reviewer’s comment and changed the sentence (Lines: 343-346) to:
“According to the questionnaire’s response, due to the low outdoor temperature (-10℃ to -32℃) none of the participants who cooked also opened the window during this time period; thus, the impact of any outdoor sources during the cooking period would be negligible.”
Line 370: I think you have to take into account the location of the room within the block, outside effects, and its position within the path of ventilation. So care needs to be taken on any conclusions drawn.
Response: Thank you for the comments. We would like to clarify that in residence E, the window (NV) was the only way for air exchange from the indoor to outdoor and vice versa which has been mentioned in the manuscript. We agree that it would have been helpful if we knew the position of each participant in the block as well as the ventilation path. However, this information was not included in our survey, thus we are unable to obtain it. Following your comment, we have added the following sentence to the manuscript.
The sentence was added (Lines: 373-375) following the suggestion:
“The actual ventilation rates would be affected by a number of more complex factors, such as the specific location of the room in the block, as well as the ventilation path. However, we did not include this question in the survey.”
Line 478: Can you comment on what a “Purple Air” system is? Is it a government unit with high precision? What was the sampling period for this unit?
Response: PurpleAir (PA) sensors are widely used by individuals, community groups, and other organizations including state and local air monitoring agencies to measure particulate matter (PM). It is not a government-running system. The nearby government PM monitoring station is about 3.7 km away from our studied sensor whereas the PA sensor is 500 m away. We decided to consider PA PM2.5 data to get a better comparison between the indoor and outdoor concentrations of the residences. The sampling period was from February 24, 2022, to March 6, 2022.
The following sentences have been modified (Lines: 449-506) for further clarification:
“Fig.10 illustrates the time profiles of the hourly average data of the residence sensors with that of the nearest outdoor PurpleAir (PA) sensor which is 500 m away from the studied indoor sensor. PA sensor is a widely deployed LCS network across the globe.6 PA's PM sensor operates under a similar principle to our LCS but is manufactured by a different vendor. The specific PA we selected was deployed and maintained by Alberta Capital Airshed, the local organization overseeing the local air quality monitoring network. The comparison was done for the sampling period of February 24, 2022, to March 6, 2022. ”
Line 533: Please add the limitations of the study. For example, you only measured PM2.5 and did not include PM10, PM1, or ultrafine, which is becoming of increasing concern.
Response: Agreed. We have accordingly added a sentence (Lines: 172-174) to emphasize this point:
“While PM10, PM1, or ultrafine PM have also been of increasing concern and studied in numerous literature, but we did not investigate these in our work.”
Referee: 2
Comments to the Author
General comments:
I think it’s great that the authors of this study provided students living in the university residences with access to real-time indoor air quality data! I also appreciate that these results illustrate how much ultrasonic humidifiers can increase indoor PM2.5 concentrations.
Response: We appreciate the time and effort that the reviewers have dedicated to providing your valuable feedback on the manuscript. We have been able to incorporate changes to reflect most of the suggestions provided by the reviewers, which we believe is much improved after revision.
Here is a point-by-point response to the reviewers’ comments and concerns.
Specific comments:
1. I would like to see the authors do some additional quality assurance and correction of their CO2 data from the Sensirion SCD30 sensor. I think low-cost NDIR CO2 sensors are great tools for assessing ventilation of indoor spaces, but such sensors are known to suffer from baseline drift. This issue of baseline drift is the reason why commercially-available monitors that use low-cost NDIR CO2 sensors, like the Aranet4, recommend that the baseline be “recalibrated” to the outdoor concentration of 420 ppm on a regular basis. Fortunately, it sounds like the authors collected the data they would need to perform a baseline correction on the SCD30 sensor data. On lines 181-184, the authors state that they opened the windows and collected SCD30 sensor data while the CO2 concentration decayed to the outdoor ambient value of 420 ppm. What CO2 concentration did each sensor report when measuring the outdoor concentration of 420 ppm? The data from each sensor could be corrected, by adding or subtracting some constant value from each reported concentration, so that the corrected CO2 concentration is equal to 420 ppm when the sensor is measuring outdoor air.
Response: Thanks. Although we did open the window at the end of the SDS30 sensor’s CO2 intercomparison study, the concentration did not reach 420 ppm due to the recirculation of indoor air from the building. As shown in Figure S1 (taken post-campaign). However, all the sensors were consistent in the measurement of the building's air background. Given that all the sensors achieved a reasonable indoor air background after months of deployment, we assume that the background drifting was not very significant.
The sentence (lines: 188-190) has been modified as below in the manuscript:
“ The natural decay of CO2 concentration was monitored until it reached approximately 580 ppm. However, the CO2 concentration did not reach the ambient background of approximately 420 ppm7 which was assumed due to the recirculation of indoor air from the building.”
Another issue is that low-cost NDIR CO2 sensors are sensitive to variations in atmospheric pressure but the SCD30 doesn’t measure atmospheric pressure. Unless provided with data indicating otherwise, sensors like the SCD30 assume that the atmospheric pressure is equal to 101.3 kPa. At lower atmospheric pressures, air density in lower, and there are fewer molecules per unit volume, so the sensor will underestimate the true CO2 mole fraction. Tryner et al. (DOI: 10.1016/j.buildenv.2021.108398) found that the SCD30 underestimated the CO2 concentration by about 20% in a location where the atmospheric pressure was 85 kPa, even after a baseline correction was applied. The atmospheric pressure in Edmonton is 94 kPa. Did the device that the authors built collect atmospheric pressure data? If so, were those pressure data used to correct the CO2 data for the difference between atmospheric pressure at sea level and the atmospheric pressure in Edmonton? If not, the authors should add a disclaimer that CO2 data from the SCD30 were not corrected for atmospheric pressure and that the reported CO2 values might be slightly underestimated. My guess is that, even if CO2 values were not corrected for pressure, CO2 concentrations are not underestimated by more than 10%.
Response: Thank you. You have raised an important point here. However, we didn’t correct the SCD30 sensor data against the atmospheric data. As suggested by the reviewer, a sentence has been added (lines: 191-193) in the manuscript:
“ A previous study has found that the SCD30 could underestimate the CO2 concentration measurement under reduced pressure (approximately 85 kPa). 8. However, we didn’t correct the SCD30 sensor’s CO2 data because the pressure in student residences is unlikely to reach such low values .”
2. Line 88: I suggest revising “> 5000 ppm” to “> 1000 ppm”. I agree that indoor locations with elevated CO2 concentrations pose a risk to human health, but 5000 ppm is a very high CO2 concentration that I don’t think is common.
Response: Thank you for this suggestion. We agreed with you that the 5000 ppm CO2 concentration is not common indoors but not impossible. Azuma et al, found that the physiological changes occur at CO2 exposure levels between 500 and 5000 ppm9. Therefore, the following modifications were done accordingly (Lines: 86 - 91):
“ Growing evidence shows that exposure to a CO2 concentration of 500-5000 ppm can cause various physiological changes in circulatory and autonomic systems, and numerous unpleasant responses were also reported, including drowsiness, lethargy, stuffiness, and a feeling that the air is stale 9. Impairment of the cognitive performances including decision-making and problem-solving, could begin at 1000 ppm even with short-term exposure 10–14.”
3. Lines 116-117 and Lines 166-167: “…The SDS011 sensor could also measure PM10…” Despite the manufacturer’s claims, the SDS011 cannot measure PM10. See Kuula et al. (DOI: 10.5194/amt-13-2413-2020, especially Figure 5, the text on page 2418, and the references to Budde et al. and Laquai therein). I suggest not stating that this sensor can measure PM10 anywhere in the manuscript. There’s no need to comment on the ability to measure PM10 with the SDS011 since PM10 was not a focus of this study.
Response: Agreed. Thank you for this suggestion. The sentence (lines 117-119) has been modified in the manuscript:
“According to the manufacturer's information, the SDS011 sensor can measure particles with diameters between 0.3 and 10 µm in the air by using the principle of light scattering.”
For the sentence in Line 166, we have decided to retain the sentence, but add the literature that the reviewer suggested as a rationale for why we have excluded PM10. The sentence (lines: 169-172) has now been( modified as below:
“ As per manufacturer specification, the SDS011 sensor could also measure PM10. However, it has been found that PM10 measurement can be inaccurate compared to the reference instruments, especially if PM distribution shifts towards larger particles 15,16. Herein, we focus only on the PM2.5 data in our study. ”
4. Line 169: “RH is considered the primary source of measurement error for PM sensors.” The rest of this paragraph is spot-on, but I don’t like the phrasing of this statement that RH is the “primary” source of measurement error for PM sensors. I’d argue that the key sources of measurement error are really differences in the size distribution and refractive index between the particles used to calibrate the sensor and the particles being measured by the sensor. The authors explain clearly in the rest of the paragraph that changes in RH can lead to measurement error because RH can affect particle size and refractive index. I suggest that the authors just revise this first sentence as “RH can be a major source of measurement error for PM sensors.”
Response: Thanks. The sentence has been revised.
5. Line 208: “…and the window or the corridor.” I think there is a typographical error here.
Response: Thanks for pointing it out. It has been corrected.
6. Table 1, Note a: I think “in-suit” should be “in-suite”.
Response: Thanks. It has been corrected.
7. Section 3.2: Please move Figure 4, Figure 5, and some of this text to the SI. It would be sufficient to comment briefly here on data availability, the Wi-Fi issues that the authors’ experienced, and the dates when many students left their residences, but I don’t think all of the information in this section is critical to readers, and I think this manuscript would benefit from being shorter.
Response: Thank you for the suggestion. However only figure 5 has been moved to the SI as 1st reviewer suggested to mention about the limitations of our study and the sensor itself in the original manuscript.
8. Lines 322-325: The PM2.5 concentrations that the authors report here are very low (1 to 4 μg/m3). Are the authors certain that these concentrations are distinguishable from each other and from zero? Did the authors do any tests to evaluate the limit of detection for the SDS011 sensor? In other words what is the lowest PM2.5 concentration that the SDS011 can distinguish as being different from zero? Did the authors test the SDS011 measuring HEPA-filtered air or “clean” air in the office without the humidifier present?
Response: Thanks. Yes, we measured the limit of detection for the SDS011 sensor. A sentence has been added (lines: 274-279) in the manuscript following the reviewer's suggestion:
“We also operated the SDS011 sensors in the office without the humidifier for 24 hours to measure the limit of detection (LOD) for PM2.5, which was found to be 0.6 μg/m3. The method used to calculate the LOD was adapted from previous literature.17,18 LOD was the lowest concentration above which hourly averaged concentrations exceed their standard deviations by a factor of 3 more than 95% of the time. ”
9. Lines 332-337: How were averages and standard deviations calculated for the t-test? Was the time-averaged PM2.5 concentration reported by each of the 8 sensors in Residence A during the occupied period treated as a single measurement for the occupied period? Were the sample mean and sample standard deviation calculated from the time-averaged PM2.5 concentrations measured by each of the n sensors in each residence? Or was something else done?
Response: The occupied period time-averaged PM2.5 concentrations of residence A were measured for each of the 8 sensors for the calculation of the t-test. And the mean and standard deviation of the time-averaged PM2.5 concentrations were measured by each of the n sensors in each residence.
10. Figure 6 caption: It would be helpful to remind readers in the figure caption that all residences A-D were mechanically ventilated, that residences A–B prohibited cooking, and that residences C–D allowed cooking.
Response: Agreed. The figure caption has been corrected.
“ Effect of occupancy on indoor PM2.5 concentration (6A) and CO2 concentration (6B): Residences A-D were mechanically ventilated where residences A and B prohibited cooking, and residences C and D allowed cooking.”
11. Line 359: “closed” should be “close”
Response: Thanks. It has been corrected.
12. Lines 361-364, 365-368, 378-380 and Table S4: Did the authors adjust their α values for these multiple comparisons? One option is to use Tukey’s method for multiple comparisons.
Response: Thanks. We didn’t adjust α values as we didn’t do the multiple comparisons tests. We did a two-sample t-test. For example, for the occupancy study, we did a comparison between occupied vs unoccupied residence A or occupied vs unoccupied residence B, etc. Similarly, for the ventilation effect study, we did MV of residence A vs NV of residence E, or MV of residence B vs NV of residence E, etc.
13. Lines 386-387: “Results of this study confirm that MV with filtration can significantly reduce indoor particle levels and maintain good IAQ.” Do the authors know the MERV ratings of the filters used in the HVAC systems for these residences?
Response: The MERV ratings of the filters used in the MV of the residence A-D are M8; M8 & M13; M8; and M8 & M14, respectively. The information has been added to the SI table S2.
14. Figure 7 caption: Please note here in the figure caption that Residences A–D were mechanically ventilated and that Residence E was naturally ventilated.
Response: Thanks. The figure caption has been corrected. The caption has been modified to:
“ Effect of MV (residence A-D) and NV (residence E) on indoor PM2.5 (7A) and CO2 (7B) concentration.”
15. Lines 417-420: This is an important result!
Response: Thank you!
16. Lines 436-438: “Student residents should, where possible, use water with low levels of mineral and other dissolved components in ultrasonic humidifiers to limit inhalation exposure to potentially harmful components.” Or should they use evaporative humidifiers?
Response: Indeed. This is an important suggestion. The following modifications were done accordingly as per reviewer comments (lines: 458-461):
“ Student residents should be encouraged to use water with low levels of mineral and other dissolved components to limit inhalation exposure to potentially harmful components. Alternatively, an evaporative humidifier can be used because it does not generate as much PM as an ultrasonic humidifier 19.”
17. Lines 476-478: What correction was applied to the PurpleAir data? The US EPA correction?
Response: Thanks. The purpleAir (PA) data was not corrected. The US EPA recommended PM2.5 concentration correction model used ambient RH (relative humidity) data to improve the PM data accuracy of the sensor 20. Studies found that when RH exceeds 75% then it could bias (significantly increase) the PM2.5 measurement of the sensor PA sensor 5,21. However, during our studied period the outdoor RH was less than 55 %, so we can possibly exclude the impact of RH on the PurpleAir sensor PM2.5 data. On the other hand, we used PA data only for comparing the PM2.5 concentration trend, not for quantitation compared to the indoor PM2.5 concentration. The manuscript has added a sentence following the comments (lines: 506-508):
“ We did not correct the PA PM2.5 concentration data as it was considered for comparison with the indoor PM2.5 concentration trend, not for quantitative comparison.”
18. Lines 491-494: “Dai et. Al, suggested minimizing particle filter efficiencies of MV systems are 86%, 85%, 74%, 58%, and 62% for severe cold, cold, hot summer and cold winter, moderate and hot summer, and warm winter zones, respectively.” I don’t understand what this sentence means.
Response: This sentence was included in the manuscript to show the required particle filter efficiencies of MV systems recommended by Dai et al. in different seasons (particularly in cold/winter) to compare our studied residence’s (A and B) MV particle filter efficiencies. However, the sentence seems not very relevant to our observations, thus it has been deleted. Thanks for pointing this out.
19. Lines 496-498: “Based on the weekday’s data, the outdoor PM2.5 removal efficiency of HVAC and HRV filters was found at 76-84% and 80-90%, which indicates the high filtration efficiency of both ventilation systems.” How was this calculation performed? I don’t think it’s appropriate to do this sort of quantitative comparison using indoor and outdoor concentrations measured using two different brands and models of low-cost PM sensors. There’s no guarantee that the SDS011 and the PMS5003 respond similarly to the outdoor PM2.5. It’s useful to point out the correlation between outdoor and indoor PM2.5 that is shown in Figure 11, but the authors should remove these estimates of filtration efficiency from the manuscript.
Response: Thank you for your comments. Yes, we agreed with you that the SDS011 and the PMS5003 might respond differently to the outdoor PM2.5 to some extent and a quantitative comparison is infeasible between them. However, given the I/O correlation and the fact that indoor PM was significantly lower, we wanted to comment on the efficiency of filtration. So the following sentence has been modified (Lines: 521-524) for further clarification:
“However, based on the weekday’s PM2.5 concentration trend, the indoor PM2.5 concentration level was found low compared to the outdoor PM2.5 concentration which indicates the efficient filtration efficiency of both ventilation systems in these two residences.”
20. Lines 503-505: “In this study, the MV systems in the two studied buildings were successful in keeping PM2.5 concentration below the recommended daily acute exposure limit.” Were the outdoor PM2.5 concentrations above the recommended daily acute exposure limit? If the outdoor concentrations were already below the recommended daily acute exposure limit, then the MV systems didn’t have to do anything.
Response: From figure 11, if we particularly choose February 28, 2022, as an example (weekdays), the outdoor 24-hour average PM2.5 concentration was recorded at 41.2 μg/m3 which is higher than the daily acute exposure recommended limit. On the other hand, where indoor PM2.5 concentration was in residence A 24-hour average concentration was 2.1 to 6.6μg/m3 and in residence, B was 1.1 to 6.6μg/m3. There were some other events when we observed that the outdoor PM2.5 concentration was higher than the daily acute exposure limit but the indoor concentration was very low. So, we can confidently conclude that the MV systems in residences A and B successfully kept the PM2.5 concentration low.
Again, the filtration efficiency of the ventilation system not only depends on how much they remove the pollutant from the outdoor air but also on how much from the indoor air.
Reference
1. Guidebook - Final - Epa - Epa/600/R-14/159 June 2014 Epa/Ord Air Sensor Guidebook Ron Williams and - Studocu. https://www.studocu.com/en-us/document/universal-technical-institute/pile-design/guidebook-final-epa/18013270. Accessed December 1, 2022.
2. USEPA Quality assurance guideline.pdf.https://www3.epa.gov/ttnamti1/files/ambient/pm25/qa/m212.pdf
Accessed August 22, 2022.
3. Canada H. Consultation: Proposed Residential Indoor Air Quality Guidelines for Carbon Dioxide. October 29, 2020. https://www.canada.ca/en/health-canada/programs/consultation-residential-indoor-air-quality-guidelines-carbon-dioxide/document.html. Accessed August 15, 2022.
4. Canada H. Carbon dioxide in your home. March 19, 2021. https://www.canada.ca/en/health-canada/services/publications/healthy-living/carbon-dioxide-home.html. Accessed December 10, 2022.
5. Zimmerman N. Tutorial: Guidelines for implementing low-cost sensor networks for aerosol monitoring. Journal of Aerosol Science. 2022;159:105872.
6. Malings C, Tanzer R, Hauryliuk A, et al. Fine particle mass monitoring with low-cost sensors: Corrections and long-term performance evaluation. Aerosol Science and Technology. 2020;54:160–174.
7. Stassen I, Dou J-H, Hendon C, Dincă M. Chemiresistive Sensing of Ambient CO2 by an Autogenously Hydrated Cu3(hexaiminobenzene)2 Framework. ACS Central Science. 2019;5:1425–1431.
8. Tryner J, Phillips M, Quinn C, et al. Design and testing of a low-cost sensor and sampling platform for indoor air quality. Building and Environment. 2021;206:108398.
9. Azuma K, Kagi N, Yanagi U, Osawa H. Effects of low-level inhalation exposure to carbon dioxide in indoor environments: A short review on human health and psychomotor performance. Environment International. 2018;121:51–56.
10. Du B, Tandoc MC, Mack ML, Siegel JA. Indoor CO2 concentrations and cognitive function: A critical review. Indoor Air. 2020;30:1067–1082.
11. Satish U, Mendell MJ, Shekhar K, et al. Is CO2 an Indoor Pollutant? Direct Effects of Low-to-Moderate CO2 Concentrations on Human Decision-Making Performance. Environmental Health Perspectives. 2012;120:1671–1677.
12. Allen JG, MacNaughton P, Satish U, Santanam S, Vallarino J, Spengler JD. Associations of Cognitive Function Scores with Carbon Dioxide, Ventilation, and Volatile Organic Compound Exposures in Office Workers: A Controlled Exposure Study of Green and Conventional Office Environments. Environmental Health Perspectives. 2016;124:805–812.
13. Norbäck D, Nordström K, Zhao Z. Carbon dioxide (CO2) demand-controlled ventilation in university computer classrooms and possible effects on headache, fatigue and perceived indoor environment: an intervention study. International Archives of Occupational and Environmental Health. 2013;86:199–209.
14. Azuma K, Kagi N, Yanagi U, Osawa H. Effects of low-level inhalation exposure to carbon dioxide in indoor environments: A short review on human health and psychomotor performance. Environment International. 2018;121:51–56.
15. Kuula J, Mäkelä T, Aurela M, et al. Laboratory evaluation of particle-size selectivity of optical low-cost particulate matter sensors. Atmos Meas Tech. 2020;13:2413–2423.
16. Budde M, Schwarz A, Müller T, et al. Potential and Limitations of the Low-Cost SDS011 Particle Sensor for Monitoring Urban Air Quality. December 2018.
17. Wallace L, Ott W, Zhao T, Cheng K-C, Hildemann L. Secondhand exposure from vaping marijuana: Concentrations, emissions, and exposures determined using both research-grade and low-cost monitors. Atmospheric Environment: X. 2020;8:100093.
18. Wallace L, Bi J, Ott WR, Sarnat J, Liu Y. Calibration of low-cost PurpleAir outdoor monitors using an improved method of calculating PM2.5. Atmospheric Environment. 2021;256:118432.
19. Lau CJ, Loebel Roson M, Klimchuk KM, Gautam T, Zhao B, Zhao R. Particulate matter emitted from ultrasonic humidifiers—Chemical composition and implication to indoor air. Indoor Air. 2021;31:769–782.
20. Barkjohn KK, Gantt B, Clements AL. Development and application of a United States-wide correction for PM2.5 data collected with the PurpleAir sensor. Atmospheric Measurement Techniques. 2021;14:4617–4637.
21. Jayaratne R, Liu X, Thai P, Dunbabin M, Morawska L. The influence of humidity on the performance of a low-cost air particle mass sensor and the effect of atmospheric fog. Atmos Meas Tech. 2018;11:4883–4890.
Round 2
Revised manuscript submitted on 23 dec 2022![](https://www.rsc-cdn.org/pubs-core/2022.0.147/content/NewImages/icon-letter.png)
![](https://www.rsc-cdn.org/pubs-core/2022.0.147/content/NewImages/chevron-down.png)
02-Jan-2023
Dear Dr Zhao:
Manuscript ID: EA-ART-11-2022-000149.R1
TITLE: Investigation of Indoor Air Quality in University Residences Using Low-Cost Sensors.
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 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 Nønne Prisle
Associate Editor, Environmental Sciences: Atmospheres
************
![](https://www.rsc-cdn.org/pubs-core/2022.0.147/content/NewImages/icon-comments.png)
![](https://www.rsc-cdn.org/pubs-core/2022.0.147/content/NewImages/chevron-down.png)
Reviewer 2
I thank the authors for taking my comments into account when revising their manuscript. I ask that the authors make two additional, minor revisions for the sake of technical accuracy and reproducibility.
1. Lines 193-194: “However, we didn’t correct the SCD30 sensor’s CO2 data because the pressure in student residences is unlikely to reach such low values.” Please state the typical atmospheric pressure in Edmonton here. 94 kPa?
2. Lines 510-512: “We did not correct the PA PM2.5 concentration data as it was considered for comparison with the indoor PM2.5 concentration trend, not for quantitative comparison.” There is still not enough information here on the PurpleAir data. The Plantower PMS5003 sensors used in the PurpleAir report PM2.5 concentrations two different ways: “PM2.5 CF=1” and “PM2.5 ATM,” and these two values differ at concentrations above 30 µg m-3. Which of these two data streams are the authors showing in Figure 10? In other words, which variables from the PurpleAir data log are shown in Figure 10? pm2.5_cf_1 or pm2.5_atm? How did the authors obtain the PurpleAir data? Did the authors download .csv files from ThingSpeak or did they access the data through the PurpleAir API? If the authors used data from the PurpleAir API that were just labeled ‘pm2.5’, it’s likely that these where pm2.5_atm values. See the API documentation: “pm2.5 returns average for channel A and B but excluding downgraded channels and using CF=1 variant for indoor, ATM variant for outdoor devices” (https://api.purpleair.com/#api-sensors-get-sensor-data). As discussed by Barkjohn et al. [DOI: 10.5194/amt-14-4617-2021], uncorrected PurpleAir data typically overestimate ambient PM2.5 concentrations in North America by ~40%. That’s one reason why the PM2.5 concentrations from the PurpleAir monitor that are shown in Figure 10 are so much higher than the PM2.5 concentrations reported from inside the residences. The authors should really apply the simple linear EPA correction to these data. In early 2021, Barkjohn et al. published an equation that can be used to correct pm2.5_cf_1 values: PM2.5 = 0.524 × pm2.5_cf_1 − 0.0862 × RH + 5.75 (Equation 10 in the publication). Later in 2021, Barkjohn et al. presented a piecewise correction equation that was designed to correct pm2.5_atm values: PM2.5 = 0.52 × pm2.5_atm − 0.086 × RH + 5.75 when pm2.5_atm is below 50 µg m-3 and PM2.5 = 0.786 × pm2.5_atm − 0.086 × RH + 5.75 when pm2.5_atm is between 50 and 229 µg m-3 (see Slide 14 in https://cfpub.epa.gov/si/si_public_record_report.cfm?dirEntryId=353088&Lab=CEMM). I checked, and the corrected PurpleAir PM2.5 concentrations will still be higher than the indoor PM2.5 concentrations, as the authors state on lines 514-517 (“As shown in Figure 10, the indoor and outdoor PM2.5 concentrations exhibited a clear correlation albeit the indoor concentrations were lower and appeared to be delayed compared to those outdoors, likely due to the infiltration process”) and 529-532 (“However, based on the weekday’s PM2.5 concentration trend, the indoor PM2.5 concentration level was found low compared to the outdoor PM2.5 concentration which indicates the efficient filtration efficiency of both ventilation systems in these two residences.”). Even though both of these statements will still be true after the EPA correction is applied, it will lend more credibility to the authors’ results if these statements are not made in reference to a plot of PurpleAir PM2.5 concentrations that are artificially high due to well-known biases associated with the design and operation on the Plantower PMS5003 sensor.
![](https://www.rsc-cdn.org/pubs-core/2022.0.147/content/NewImages/icon-response.png)
![](https://www.rsc-cdn.org/pubs-core/2022.0.147/content/NewImages/chevron-down.png)
Response to Reviewers
Referee: 2
General comments:
I thank the authors for taking my comments into account when revising their manuscript. I ask that the authors make two additional, minor revisions for the sake of technical accuracy and reproducibility.
Response: We would like to thank the reviewer for the comments. Care has been taken to address the concerns as per the specific comments below and also made corrections in our manuscript.
Specific comments:
1. Lines 193-194: “However, we didn’t correct the SCD30 sensor’s CO2 data because the pressure in student residences is unlikely to reach such low values.” Please state the typical atmospheric pressure in Edmonton here. 94 kPa?
Response: Thanks. The sentences have been modified (line 193-195) in the manuscript following the reviewer’s comments:
“However, we didn’t correct the SCD30 sensor’s CO2 data because the atmospheric pressure in Edmonton was 90 to 96 kPa during the study period (December 2021 to April 2022)1 ”
2. Lines 510-512: “We did not correct the PA PM2.5 concentration data as it was considered for comparison with the indoor PM2.5 concentration trend, not for quantitative comparison.” There is still not enough information here on the PurpleAir data.
The Plantower PMS5003 sensors used in the PurpleAir report PM2.5 concentrations in two different ways: “PM2.5 CF=1” and “PM2.5 ATM,” and these two values differ at concentrations above 30 µg m-3. Which of these two data streams are the authors showing in Figure 10? In other words, which variables from the PurpleAir data log are shown in Figure 10? pm2.5_cf_1 or pm2.5_atm? How did the authors obtain the PurpleAir data? Did the authors download .csv files from ThingSpeak or did they access the data through the PurpleAir API? If the authors used data from the PurpleAir API that were just labeled ‘pm2.5’, it’s likely that these were pm2.5_atm values. See the API documentation: “pm2.5 returns average for channel A and B but excluding downgraded channels and using CF=1 variant for indoor, ATM variant for outdoor devices” (https://api.purpleair.com/#api-sensors-get-sensor-data).
Response: Thanks. PurpleAir data were downloaded through the PurpleAir API as .csv files. However, In figure 10, PM2.5_atm data were shown.
As discussed by Barkjohn et al. [DOI: 10.5194/amt-14-4617-2021], uncorrected PurpleAir data typically overestimate ambient PM2.5 concentrations in North America by ~40%. That’s one reason why the PM2.5 concentrations from the PurpleAir monitor that are shown in Figure 10 are so much higher than the PM2.5 concentrations reported from inside the residences. The authors should really apply the simple linear EPA correction to these data. In early 2021, Barkjohn et al. published an equation that can be used to correct pm2.5_cf_1 values: PM2.5 = 0.524 × pm2.5_cf_1 − 0.0862 × RH + 5.75 (Equation 10 in the publication). Later in 2021, Barkjohn et al. presented a piecewise correction equation that was designed to correct pm2.5_atm values: PM2.5 = 0.52 × pm2.5_atm − 0.086 × RH + 5.75 when pm2.5_atm is below 50 µg m-3 and PM2.5 = 0.786 × pm2.5_atm − 0.086 × RH + 5.75 when pm2.5_atm is between 50 and 229 µg m-3 (see Slide 14 in https://cfpub.epa.gov/si/si_public_record_report.cfm?dirEntryId=353088&Lab=CEMM). I checked, and the corrected PurpleAir PM2.5 concentrations will still be higher than the indoor PM2.5 concentrations, as the authors state in lines 514-517 (“As shown in Figure 10, the indoor and outdoor PM2.5 concentrations exhibited a clear correlation albeit the indoor concentrations were lower and appeared to be delayed compared to those outdoors, likely due to the infiltration process”) and 529-532 (“However, based on the weekday’s PM2.5 concentration trend, the indoor PM2.5 concentration level was found low compared to the outdoor PM2.5 concentration which indicates the efficient filtration efficiency of both ventilation systems in these two residences.”). Even though both of these statements will still be true after the EPA correction is applied, it will lend more credibility to the authors’ results if these statements are not made in reference to a plot of PurpleAir PM2.5 concentrations that are artificially high due to well-known biases associated with the design and operation on the Plantower PMS5003 sensor.
Response: Thanks. Purple air PM2.5_atm data is corrected and plotted in figure 10. Data correction is done using PM2.5 = 0.52 × PM2.5_atm − 0.086 × RH + 5.75 as PM2.5_atm was below 50 µg m-3 during the comparison period of PM2.5 of indoor and outdoor air. As a response to the reviewer comments, figure 10 showing the comparison of PM2.5_atm data before and after correction has been given and will be found in the attached pdf file of the "response to the reviewer'. However, in the manuscript, only the corrected data of PM2.5_atm is shown in the revised figure 10.
And the sentences (lines: 507-511) have been modified to support the correction of purple air PM2.5 data:
“Although PA PM2.5 concentration data were considered for comparison with the indoor PM2.5 concentration trend, not for quantitative comparison. However, a correction equation recommended by US EPA2 was applied to correct the PA sensor PM2.5 data to make it more comparable to our indoor LCS PM2.5 data, thus reducing data accuracy concerns.”
Additionally, the figure 10 caption has been changed to the following in the manuscript:
“Hourly variations of indoor PM2.5 level with outdoor PM2.5 level, during weekdays and weekends. The outdoor PurpleAir data was treated with corrections recommended by the US EPA.\cite{PAPM2.5datacorrection.”
Reference
Station pressure-Hourly data for Edmonton.
URL:https://edmonton.weatherstats.ca/charts/pressure_station-hourly.html
Accessed January 03, 2023.
Sensor data cleaning and correction: Application on the AirNow Fire and Smoke Map.
URL:https://cfpub.epa.gov/si/si_public_record_report.cfm?dirEntryId=353088&Lab=CEMM
Accessed January 03, 2023.
Round 3
Revised manuscript submitted on 04 jan 2023![](https://www.rsc-cdn.org/pubs-core/2022.0.147/content/NewImages/icon-letter.png)
![](https://www.rsc-cdn.org/pubs-core/2022.0.147/content/NewImages/chevron-down.png)
11-Jan-2023
Dear Dr Zhao:
Manuscript ID: EA-ART-11-2022-000149.R2
TITLE: Investigation of Indoor Air Quality in University Residences Using Low-Cost Sensors.
Thank you for submitting your revised manuscript to Environmental Science: Atmospheres. I am pleased to accept your manuscript for publication in its current form. I have copied any final comments from the reviewer(s) below.
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 Nønne Prisle
Associate Editor, Environmental Sciences: Atmospheres
![](https://www.rsc-cdn.org/pubs-core/2022.0.147/content/NewImages/icon-comments.png)
![](https://www.rsc-cdn.org/pubs-core/2022.0.147/content/NewImages/chevron-down.png)
Reviewer 2
I thank the authors for carefully revising their manuscript and writing a thorough response to my questions and concerns!
Transparent peer review
To support increased transparency, we offer authors the option to publish the peer review history alongside their article. Reviewers are anonymous unless they choose to sign their report.
We are currently unable to show comments or responses that were provided as attachments. If the peer review history indicates that attachments are available, or if you find there is review content missing, you can request the full review record from our Publishing customer services team at RSC1@rsc.org.